key: cord- -z sxr dx authors: coveri, andrea; cozza, claudio; nascia, leopoldo; zanfei, antonello title: supply chain contagion and the role of industrial policy date: - - journal: j doi: . /s - - - sha: doc_id: cord_uid: z sxr dx the covid- pandemic triggered a major disruption in global value chains (gvcs) that pushed the global economy into a recession that promises to be worse than the crisis. this article illustrates the mechanisms through which the covid- pandemic affected gvcs in the context of a changing configuration of the global economy. in particular, it is argued that gvcs became the main transmission channels of “economic contagion”. finally, we posit that the pandemic provides an opportunity to revive the role of industrial policy as to govern the landslides of a world economy constantly pressured by globalization and deglobalization forces. the diffusion of covid- represents the most serious pandemic since the spanish flu over the - period. after the outbreak in china in the second half of february , the pandemic spread worldwide in a few weeks. the large lockdown of production that followed has immediately shown its disruptive impact and promises to trigger the most serious economic crisis since the great recession of . among the most affected countries are in fact the largest industrialized economies in the world, primarily the united states, spain, italy, germany, the united kingdom and france, as well as china itself and japan, responsible overall for over % of global gdp (baldwin and tomiura ) . this short contribution illustrates the impact of the covid- pandemic on the current configuration of global economy (sect. ), stressing the role of global value chains (gvcs) as the main transmission channels of "economic contagion" worldwide (sect. ). in this context, we provide a focus on italy, which has been the first economy to be affected in europe, and most substantially. section highlights the challenges imposed by the pandemic and discusses the key role that industrial policy should play to set the global economy on a more resilient and sustainable path. finally, the last section draws some conclusions. supply-and demand-side contagion channels pandemic occurs in the midst of a transformation that has been characterizing the world economy over the last four decades. since the s, a series of technological, political and institutional factors gave rise to what baldwin ( ) called the " nd unbundling", meaning a powerful push towards the geographical dispersion of value-added tasks required to get the realization of final products. this vertical fragmentation of production on a global scale has given rise to global value chains (gvcs), along which east asia and china in particular represent a crucial node (baldwin ; gereffi and fernandez-stark ; zanfei et al. ) . in other words, the sophisticated interconnection of the productive structures of the different countries at the international level led the interdependence of the economies to increase remarkably. as a consequence, a production bottleneck in a hub of the dislocated supply chain can trigger a synchronized economic slowdown and, therefore, a global downturn (baldwin and weder di mauro ; boehm et al. ; inoue and todo ; sforza and steininger ; unctad ) . therefore, the first channel through which the pandemic has affected the global economy concerns the interruption of the supply chains, hitting what has been identified as the productive heart of the world, that is east asia, and china in particular (buckley and strange ) . in this regard, consider that chinese industrial production has fallen by . % on an annual basis between january and february , i.e. the biggest drop in production in china since the post-deng xiaoping era, especially in the transport equipment (− . %), general equipment (− . %), textiles (− . %) and machinery (− . %) (data from national bureau of statistics of china). the second, and related, channel concerns the amplification effect of the contagion induced by the chains of global subcontracting of intermediate goods, since even the manufacturing sectors of the least affected countries have had difficulty in acquiring (importing) the intermediate inputs necessary for domestic production (baldwin and tomiura ; bonadio et al. ) . in this context, the loosening of the production lockdown in china (in march), while it was being strengthened in europe and the united states (in april), adds further complexity to this scenario. on the one hand, the attempts of western importing companies to reduce their dependence on intermediate goods from china, the so-called decoupling, might turn out a bad strategy, as china is now recovering faster than the rest of world. on the other hand, the global spread of covid- is ultimately harmful also for china, as it imports a large part of the intermediate goods necessary for production for its domestic market, as well as for foreign markets (world bank ). in addition, the demand-side of the gvcs-pandemic nexus should not be underestimated. firstly, governments' measures aimed at containing the virus, such as the drastic reduction of people mobility and the shutdown of almost all commercial and leisure activities, entail an immediate reduction of consumptions. some of them will be postponed but many others will be probably never recovered. the heaviest effects will concern the services sector, in particular transport, tourism, accommodation and catering sector, thus affecting some countries (e.g. italy) more than others. secondly, the sharp slowdown in production is expected to generate an increase in the unemployment rate, which will result in a reduction in households' disposable income, starting with those who are hired on a temporary basis. if this reduction mainly affects the poorer segments of the population-those with a higher marginal propensity to consume-the consequences on aggregate consumption will be even more significant. in addition, due to the uncertainty caused by the speed and spread of the infection, it is reasonable to expect a reduction in investment rates and an increase in precautionary saving as a form of protection for a dark future. thirdly, a lower utilization rate of production capacity by firms could make it more difficult for them to amortize fixed costs (a trivial example is the rental cost of buildings). this in turn could lead to an increase in unit costs, a reduction in the profit rates and therefore a further contraction in investment expenditure. finally, the global reduction in consumption and investments amplifies the value added contraction simultaneously, further restricting foreign outlet markets and therefore slowing down the dynamics of net exports of both final and intermediate goods. although our focus is on economic consequences of the pandemic, it might be worth mentioning an important feedback effect. in fact, covid- may not only affect gvcs, but its diffusion might have been favored by the increasing resort to international production networks. one often emphasized mechanism is through increase in pollution. some scientists suggest that industrial production is altering ecosystems and, above all, reducing biodiversity (wallace ) . this reduction would result in a greater probability of the cross-species transmission of viruses. however, the probability of virus transmission is not only likely to be connected to industrial production in general, but more specifically to the modern hyper-globalized world economy. this link from gvcs to covid- holds for at least two reasons. first, consider the extent to which the gvcs have increased the concentration of manufacturing production in certain geographical areas, particularly in east asia but also in latin america. this concentration is accompanied by an accentuated specialization of such regions in the most energy-intensive manufacturing stages of production, taking advantage of cost reductions with little consideration of environmental issues. this is often due to the very low cost of local labor, which in turn depends on the large pool of workers in emerging countries and the less restrictive legislation compared to western economies. environmental regulation is also less restrictive than in other parts of the world, attracting polluting manufacturing activities which could contribute to increasing the likelihood of new viruses. second, the logic of gvcs meant a transnational dislocation of production which has at least partially required greater workers' mobility. given that gvcs are dominated by global actors, the international mobility of their workers, especially managers of multinational corporations (mncs), has likely played a role on the spread of the virus. although anecdotal, the first detected cases of "patient one" in germany, a manager infected by a colleague from shanghai who allegedly brought the virus to germany on january , (rothe et al. ) ; or the circumstance that the first case in the italian village of codogno was a manager of unilever, are emblematic. the shock caused by covid- pandemic occurred at a time when the global economy was already on the verge of a recession. figure reports the progressive reduction in the gdp growth rates of major economic regions-especially european ones-from to along with the imf growth projection to (imf ). according to the latter, the outbreak of the pandemic will lead to a global gdp contraction of % for . notably, the world gdp contraction during the great recession in was equal to . %. ( ) with regard to the european union, imf ( ) estimates a . % reduction in gdp, with italy called to suffer the biggest backlash: a . % decrease with respect to the previous year. to fully acknowledge the magnitude, consider that in , i.e. the worst year of the past global financial crisis, italy's gdp fell by just under % and at that time the unemployment rate in italy soared to almost % at the end of (data from oecd). moreover, preliminary data provided by the italian national institute of statistics (istat) show a decrease of . % in industrial production in march as compared to february for italy-much worse than in other industrialized countries, as reported in fig. -with a drop of . % in durable consumption goods and of . % in capital goods on a yearly basis. in addition, preliminary estimations provided by prometeia report that the pandemic would lead to an overall reduction of the italian industrial production for march and april equal to % . as stressed in the previous section, the strong interdependence of the economies has been crucial in spreading the economic contagion worldwide, following a domino scheme. in this context, the fact that china was the first economy affected by the covid- represented a factor of paramount importance because of its central role in modern global production networks. an empirical illustration of this phenomenon is shown by fig. , which reports the backward participation of prominent industrialized countries with respect to china for , and . this index is often used to approximate the level of international fragmentation of production, as it measures the percentage share of added value produced abroad (in china in this case) which is contained in the gross exports of each individual country over the total gross exports of the latter. in other words, this indicator measures to what extent the exports of countries under observation depend on the import of intermediate goods from china. as fig. focusing on italy, it is worth mentioning that the industries showing the highest level of backward participation with respect to china report much higher and increasing values over time, which in spanned from . % for machinery and equipment to . % for textiles and wearing apparel sector, up to a level equal to . % for computer and electronic and . % for electrical equipment industry (data from oecd-tiva). unsurprisingly, with the outbreak of the pandemic, this complex global interconnection of production-and the crucial role that the "factory of the world" plays in this context-has prompted part of the economic literature to warn about the need for many countries to promote supplier diversification with the aim of reducing their dependence on imports of intermediate goods from china (javorcik ; monga ). it is worth noting that, on the one hand, this diversification had been to a small extent already implemented by some mncs, which had begun to diversify their supply sources in response to trade restrictions due to the trade war between the united states and china. in particular, some mncs have moved production to other countries mainly in south east asia and/or have undertaken reshoring policies-i.e. return of production back to the country of origin from overseas-a phenomenon we will briefly discuss below. on the other hand, it should not be forgotten that the breadth of the chinese market makes the complete abandonment of the asian giant very risky and unprofitable for many companies. however, the italian backward linkages are not limited to asian-centered production networks. although the latter are gaining ever more prominence over time, the italian industrial fabric shows a higher level of integration in european and north american gvcs (wto ). indeed, italy reported an overall backward participation equal to . % in , of which the chinese share accounted for less than % (data from oecd-tiva). it follows that the success in fighting the contagion by china since march has only alleviated the gvcs-disruption burden on several italian production activities. in fact, after being imposed in china, the lockdown hit even harder european economies and the us, with respect to which italy shows an even stronger vertical integration. in this regard, fig. shows the italian industries with the highest level of backward participation, i.e. those relying the most on foreign production, with the exclusion of the manufacturing industry of coke and refined petroleum products (whose backward index reaches almost % in ). the figure reports clear-cut evidence about the crucial role of imports from foreign suppliers for the production activity of key italian industries such as basic metals, automotive, electrical equipment as well as chemicals and pharmaceutical products; for these sectors the amount of foreign value added the italian firms process to export products to third countries is more than % of their gross exports. in other words, and beyond what happens in china, if the world stops because of covid- pandemic, italy can only suffer highly negative consequences, at least in the short period. notably, the evolution of the backward participation index of the italian automotive industry (sector ) is the one showing the most marked increase from to , which goes from to %. this growing trend mirrors the offshoring of manufacturing stages of production increasingly carried out by a share of italian firms towards central and eastern european countries (ceecs) since the nineties (celi et al. ) . furthermore, the growing vertical fragmentation of the automotive sector and the changing position of italy in the global production network related to motor vehicles represents a good example to illustrate the potential implications of the pandemic on italy's economy. especially after the great recession of , the italian automotive industry has clearly made a shift in gvcs, playing a greater role as a global supplier of components for car industry, while losing positions in the final markets of cars (fig. ) . although the car production is still important in the national economy, employing around , workers in firms in , it exhibits a lower export propensity than the car electrical components industry. in particular, after , the latter largely increased its share of exports, shifting the functional role of the italian automotive industry into gvcs and increasing the country's specialization in the production of some parts of the automotive production chain, namely electrical components and bodies for motor vehicles. the traditional car industry, which used to be highly concentrated and dependent upon the italian domestic market, has thus moved towards a specialization pattern more and more focused on the exports of motor vehicles components, while the national production of cars is stable if not declining (being its volume in equal to the one in ). hence, as the world stops because of pandemic, the italian car industry suffers the interruption of purchases from foreign markets of its components while the home market has diminished its capacity to absorb them, due to the relatively low growth of its assembly activities. quite symmetrically, the lockdown firstly imposed by italian government has represented inevitably a supply chain disruption also for the other countries relying on import of intermediate goods from italy for production, exacerbating the economic crisis. alongside the vulnerabilities of gvcs highlighted by the pandemic, in italy the latter also showed how the austerity policies pursued in the last decade have made the healthcare system unable to successfully deal with emergencies (prante et al. ) . although an ageing population requires larger health expenditures, the public health system shrank. several healthcare profiles, like doctors and nurses, had to deal with a labour market with few professional opportunities and highly relying on temporary positions due to the outsourcing of health services to the private sector, especially in lombardy-i.e. the italian region most affected by covid- . the pandemic has thus revealed the shortage of health personnel, of medical infrastructures like intensive care units (icu) and of individual protection devices (ipds), like face masks for the medical staff (ranney et al. ). in this context, the pandemic caused a further reduction of the available medical staff due to a large amount of contagions because of the individual protection devices shortage, pushing the italian government to set up a call to scale up the medical staff. in addition, the restriction on international trade and the export limits imposed by several governments, aimed at ensuring national self-sufficiency, partially disrupted the supply chain of medical devices. although the medical industry was not included into the lockdown policy due to its relevance for the emergency, offshoring of relevant segments of the manufacturing chain by italian firms might have reduced the sourcing capacity of key medical equipment. the pandemic also highlights how the extension of gvcs impacts on the "knowledge chain". in fact, the covid- emergency has shown that de-specializing in some key segments of healthcare equipment manufacturing is detrimental to the overall efficiency of the system and to its resilience in particular. for example, only a few italian firms still have the knowledge to rapidly set up fully equipped icus if necessary, despite the long lasting italian industrial tradition in the biomedical sector. another example is provided by the united kingdom; the country's long-term deindustrialisation process made it very difficult to cope with the pandemic. the prime minister boris johnson had to struggle to find national companies able to manufacture icus, leading him to negotiate with the automotive industry. at the same time, he faced a shortage of personal protective equipment and of health workers, who are essential for managing any icu. the pandemic has thus highlighted the fragilities of the vertical disintegration of production across national borders, questioning at least partially the merits of fragmenting supply chains occurred in the last three decades. from this perspective, the global diffusion of covid- is likely to contribute to a reduction of trade in gvc, a process that has already been recorded in the aftermath of the great recession of , giving rise to a number of studies on production reshoring and back-shoring phenomena (rodrik ) . there are several motivations underlying such a re-thinking of gvcs. first, a sluggish global demand and market saturation phenomena reduced the (domestic and cross-border) investment opportunities for firms. second, china's involvement in the global production networks appears to have decreased over the past decade-as documented by the sharp fall in backward participation of the country from . % in to . % in (data from oecd-tiva). third, reshoring and back-shoring strategies have become more common for mncs, as a reaction to: fewer arbitrage opportunities on labour cost; higher coordination costs associated to global networking; and, last but not least, the diffusion of digital production technologies, making it convenient to re-internalize production without incurring in higher costs (dachs and zanker ; dachs et al. ; ilo ; seric and winkler ) . alongside these factors, policy-driven anti-globalization forces as brexit, trade wars and protectionist measure by the us have likely played a role (dachs and seric ) . in this scenario, the challenge for a new industrial policy concerns the capability to govern the "landslides" of a world economy constantly pressured by globalization and deglobalization forces. in other terms, leaving aside calls for hyper-globalization on one side and national self-sufficiency on the other, the mission of a modern industrial policy should be the consolidation of resilient gvcs on a reduced, largely regional, scale to pursue common development objectives. the strategy we are suggesting is not so different from what china is promoting. the asian country in the last decade addressed huge public and private resources to diversify and upgrade its production matrix, moving from low-to high-value added segments of production while building an ever more complete value chain within the south-east asian borders (european union chamber of commerce in china ). this allowed china to partially shorten its gvcs and its dependence upon imports-especially for those inputs it strongly needed in the first times of its industrialisation-without giving up the productivity gains from specialization and trade (de backer et al. ; duan et al. ; stollinger ) . in doing this, china has been building brand new production capacity with the aim of expanding and diversifying its productive matrix and enter in global markets from which chinese firms were absent until a couple of decades ago. in front of this competitive challenge, the european economy needs to simultaneously combine greater resilience of the productive system to major disruptive events as pandemics, with a strengthened international competition of its firms on global markets. we suggest that this will inevitably require a vast investment plan at the continental level. more precisely, we argue that a well-targeted european industrial policy aimed at consolidating its continental supply chains could take four birds with one stone. first, aiming to fill in the gaps-even expanded by the pandemic-along its continental value chains, the european economy would enjoy the efficiency gains arising from the division of labour while fundamentally augmenting the resilience of its supply chains. this would be ever more important with regard to strategic productions, reason why europe should act cooperatively sharing resources, knowledge and managing capabilities in industries like health, food supply, infrastructure, environmental and renewable energy and social security (see also pianta et al. ) . moreover, the pandemic-induced reorganization of the global industrial structure could be an opportunity for the selection and evolution of small-and medium-size enterprises (juergensen et al. ) , also through their repositioning into shorter supply chains marked by lower power asymmetries. second, such european industrial plan could favour economic convergence among country members, finally closing the long-term structural asymmetries which triggered the european debt crisis of - and doomed europe to a sluggish recover in the following years (celi et al. ) . it follows that investments needed to pursue such industrial strategy should be firstly addressed towards those european countries which in the last decades have suffered the most the progressive impoverishment of their production matrix (celi et al. ; landesmann and stollinger ; lucchese and pianta a) . from this perspective, the consolidation of a european block of converging and well-integrated economies, effectively participating in regional level gvcs, could constitute a stronger basis to selectively develop intercontinental production networks. the latter could be aimed at gaining access to valuable resources and knowledge assets that might not be available and cost effectively developed within one region, and that can be conveniently sought elsewhere. european economies would then benefit from specialization and trade at a wider scale, while exerting a greater bargaining power in their effort to access such resources. third, while the hyper-specialization of national economies in a few phases of the value chain increases the mutual dependence of economies-up to exposing single countries to whatever shock occurring wherever in the global economy-it also risks to make it more difficult to convert domestic production in case of shortage of critical supplies. the industrial policy design we are advancing should aim to reduce such interdependence while increasing the responsiveness and adaptation of the productive system to both internal and external conditions. finally, an industrial policy aimed at strengthening the position of european national industries in the value chains of strategic productions could help reduce the social and economic costs and risks of hyper-specialization. in fact, a draw-back of the extreme fragmentation of production is that it has reduced the incentives of a part of firms to make investments for the development of those value chain activities-other than core competencies-which appeared too risky, not enough profitable, and largely offshorable (antràs ; dottling et al. ; gutiérrez and philippon ; hummels and klenow ) . this has reduced their knowledge base while decreasing the resilience of european production networks. in fact, during the booming period of the international fragmentation of production and of greater expansion of gvcs, i.e. at the beginning of the xxi century with the entry of china in the global capitalistic market, the european economy witnessed a stagnant dynamics of investment in fixed assets, both tangible and intangible (arrighetti ; pisano and shih ). the hyper-specialization of firms thus produced an incentive to outsourcing (and geographical dispersion of) the other value chain business activities (baldwin and gu ; bernard et al. ) . the consequence has been that also fixed capital investments have been concentrated on a increasingly narrow production matrix, causing a reduction of their productive capacity and of the pool of knowledge on which their ability to compete was built on. such a dynamics points out a structural shortcoming in the process of cross-border vertical disintegration of production: not only key gaps in value chains involving strategic industries emerge, but the whole evolution of the technological matrix of european companies comes out depressed. furthermore, all that goes hand in hand with stagnant productivity, weak employment dynamics and an impoverishment of the technological base of european companies vis-à-vis emerging companies overseas. this further clarifies the necessity and feasibility of a continental industrial policy plan aimed at recovering and expanding the technological matrix of companies and at the same time filling the gaps in the value chains involving strategic productions. these considerations suggest that current tendencies of globalisation and gvcs should not be addressed with a "tina" (there is no alternative) slogan; the same slogan which contributed to the acceleration of globalisation itself over the s. conversely, alternatives to overcome the weaknesses of modern global economy exist, as also mentioned in a world investment report of a few years ago: "are active promotion of gvcs and gvc-led development strategies the only available options or are there alternatives? [the alternative] is an industrial development strategy aimed at building domestic productive capacity, including for exports, in all stages of production […] to develop a vertically integrated industry that remains relatively independent from the key actors of gvcs for its learning and upgrading processes. […] it thus appears that countries can make a strategic choice whether to promote or not to promote gvc participation. to do so, they need to carefully weigh the pros and cons of gvc participation" (unctad , p. ). in this contribution we illustrated the supply-and demand-side mechanisms through which the covid- pandemic has triggered the economic contagion which led to a sharp reduction of production worldwide. we showed that between the pandemic and globalization a close and multi-faceted relationship exists. on the one hand, the outbreak of the pandemic has an impact on global production with a domino effect. on the other hand, global production has played a role in the amplification of the pandemic, especially accelerating its spread up to a rate never experienced before. further, we provided empirical evidence about the crucial role of gvcs in bringing about the leap forward in the level of interdependence of the economies on global scale. most of this empirical evidence concerns the impact of the pandemic in italy, one of the most affected countries. in this context, we underlined the weaknesses of italian participation in gvcs as a pivotal factor in causing the collapse of production. the debate about the future trend of international production and gvcs is open (eiu ; lucchese and pianta b; seric and winkler ; strange ). many economists support the idea of an even increasing expansion of gvcs, as their widespread diversification might help reducing the global impact of national shocks (baldwin and evenett ; baldwin and weder di mauro ; monga ). however, the idea that "firms are eager, willing and able to meet any demand" (baldwin ) proved to be at least over-optimistic. conversely, one can quite convincingly argue that the pandemic has provided further reasons to rethink the pros and cons of gvcs, assessing the merits of cross-border production networks while highlighting the structural shortcomings of extreme specialization. we suggest leaving aside calls for hyper-globalization on one side and national self-sufficiency on the other, arguing for a european industrial policy plan targeted to the consolidation of resilient gvcs on a continental scale to pursue common development goals. such regional/continental level production networks-partially resembling attempts made by china to coordinate production within the south east asian region-could provide a less fragile basis for the development of more selective, strategically oriented, intercontinental networks. at the same time, a european industrial policy which goes in this direction would significantly increase the growth rate of capital formation and expand the matrix of technology used, two tendencies which appear at odds with the productive hyper-specialization induced by the modern global fragmentation of production. crisi di offerta e crescita senza investimenti, siepi.org the impact of trade on plant scale, production-run length, and diversification global supply chains: why they emerged, why they matter, and where they are going the great convergence the supply side matters: guns versus butter, covid-style covid- and trade policy: why turning inward won't work thinking ahead about the trade impact of covid- mitigating the covid economic crisis: act fast and do whatever it takes input linkages and the transmission of shocks: firm-level evidence from the tōhoku earthquake global supply chains in the pandemic the governance of the global factory: location and control of world economic activity a fragile and divided european union meets covid- : further disintegration or 'hamiltonian moment crisis in the european monetary union: a core-periphery perspective bringing it all back home? backshoring of manufacturing activities and the adoption of industry . technologies industry . and the changing topography of global value chains backshoring of production activities in european manufacturing analyzing global and regional value chains is there an investment gap in advanced economies? if so, why? why has china's vertical specialization declined? economic systems research global value chain analysis: a primer investment-less growth: an empirical investigation, nber working paper the variety and quality of a nation's exports robotics and reshoring. geneva: ilo. imf is available here the propagation of the economic impact through supply chains: the case of a mega-city lockdown to contain the spread of covid- global supply chains will not be the same in the post-covid- world european manufacturing smes amidst the covid crisis: assessing impact and policy responses structural change, trade and global production networks: an 'appropriate industrial policy' for peripheral and catching-up economies europe's alternative: a green industrial policy for sustainability and convergence the coming coronavirus crisis: what can we learn? the misguided war on global value chains, project syndicate the policy space for a novel industrial policy in europe producing prosperity: why america needs a manufacturing renaissance decades of tight fiscal policy have left the health care system in italy ill-prepared to fight the covid- outbreak critical supply shortages-the need for ventilators and personal protective equipment during the covid- pandemic new technologies, global value chains, and the developing economies, pathway for prosperity commission background paper series transmission of -ncov infection from an asymptomatic contact in germany -could -spur-autom ation -and-rever se-globa lisat ion-some-exten t. accessed globalization in the time of covid- . covid economics-vetted and real-time papers asian experiences with global and regional value chain integration and structural change the covid- pandemic and global value chains strengthening strategic value chains for a future-ready eu industry world investment report . global value chains: investment and trade for development the coronavirus shock: a story of another global crisis foretold and what policymakers should be doing about it big farms make big flu: dispatches on influenza, agribusiness, and the nature of science world development report : trading for development in the age of global value chains global value chain development report : technological innovation, supply chain trade, and workers in a globalized world fdi patterns and global value chains in the digital economy acknowledgements open access funding provided by università degli studi di urbino carlo bo within key: cord- -hk bzqm authors: cintia, paolo; fadda, daniele; giannotti, fosca; pappalardo, luca; rossetti, giulio; pedreschi, dino; rinzivillo, salvo; bonato, pietro; fabbri, francesco; penone, francesco; savarese, marcello; checchi, daniele; chiaromonte, francesca; vineis, paolo; guzzetta, giorgio; riccardo, flavia; marziano, valentina; poletti, piero; trentini, filippo; bella, antonino; andrianou, xanthi; manso, martina del; fabiani, massimo; bellino, stefania; boros, stefano; urdiales, alberto mateo; vescio, maria fenicia; brusaferro, silvio; rezza, giovanni; pezzotti, patrizio; ajelli, marco; merler, stefano title: the relationship between human mobility and viral transmissibility during the covid- epidemics in italy date: - - journal: nan doi: nan sha: doc_id: cord_uid: hk bzqm we describe in this report our studies to understand the relationship between human mobility and the spreading of covid- , as an aid to manage the restart of the social and economic activities after the lockdown and monitor the epidemics in the coming weeks and months. we compare the evolution (from january to may ) of the daily mobility flows in italy, measured by means of nation-wide mobile phone data, and the evolution of transmissibility, measured by the net reproduction number, i.e., the mean number of secondary infections generated by one primary infector in the presence of control interventions and human behavioural adaptations. we find a striking relationship between the negative variation of mobility flows and the net reproduction number, in all italian regions, between march th and march th, when the country entered the lockdown. this observation allows us to quantify the time needed to"switch off"the country mobility (one week) and the time required to bring the net reproduction number below (one week). a reasonably simple regression model provides evidence that the net reproduction number is correlated with a region's incoming, outgoing and internal mobility. we also find a strong relationship between the number of days above the epidemic threshold before the mobility flows reduce significantly as an effect of lockdowns, and the total number of confirmed sars-cov- infections per k inhabitants, thus indirectly showing the effectiveness of the lockdown and the other non-pharmaceutical interventions in the containment of the contagion. our study demonstrates the value of"big"mobility data to the monitoring of key epidemic indicators to inform choices as the epidemics unfolds in the coming months. understanding the relationship between human mobility patterns and the spreading of covid- is crucial to the restart of social and economic activities, limited or put in "stand-by" during the national lockdown to contain the diffusion of the epidemics, and to monitor the risk of a resurgence during the current phase , or lockdown exit. recent analyses document that, following the national lockdown of march th, the mobility fluxes in italy have significantly decreased by % or more, everywhere in the country, as studied in our previous report [ ] and [ , ] . in this report we study the relation between human mobility and sars-cov- transmissibility before, during and after the national lockdown. we compare the flows of people between and within italian regions with the net reproduction number r t , i.e., the mean number of secondary infections generated by one primary infector in the presence of control interventions and human behavioural adaptations. to pursue this goal, we use mobile phone data at national scale to reconstruct the self-, in-and out-flows of italian regions before and during the national lockdown (initiated on march th, ), after the closure of non-essential productive and economic activities (march th, ), and after the partial restart of economic activities and within-region movements (the "phase ", from may th, ). in this report, we address the following analytical questions: • how does the net reproduction number vary in relation to the variation of mobility flows? • what differences, if any, do we observe across the italian regions? • can we relate the delay in limiting human mobility with the rate of positive covid- cases across the population? the answers to these questions are highlighted in the next sections. an interactive, dynamically updated version of this report is available at http://sobigdata.eu/covid_report/#/report page in this report, we rely on mobile phone data, which have proven to be a useful data source to track the time evolution of human mobility [ , , ] , and thus a tool for monitoring the effectiveness of control measures such as movement restrictions and physical distancing [ , , ] . specifically, the raw data used in this report are the result of normal service operations performed by the mobile operator windtre a : cdrs (call detail records) and xdrs (extended detail records). in both cases, the fundamental geographical unit is the "phone cell" defined as the area covered by a single antenna, i.e., the device that captures mobile radio signals and keeps the user connected with the network. multiple antennas are usually mounted on the same tower, each covering a different direction. the position of the tower (expressed as latitude and longitude) and the direction of the antenna allow inferring the extension of the corresponding phone cell. the position of caller and callee is approximated by the corresponding antenna serving the call, whose extension is relatively small in urban contexts (in the order of m x m) and much larger in rural areas (in the order of km x km or more). based on this configuration, cdrs describe the location of mobile phone users during call activities and xdrs their location during data transmission for internet access. the information content provided by standard cdr and xdr is the following: in both cdrs and xdrs, the identity of the users is replaced by artificial identifiers. the correspondence between such identifiers and the real identities of the users is known only to the mobile phone operator, who might use it in case of necessity. this pseudonymization procedure is a first important step (mentioned in article ( ) and article ( ) of the gdpr, the eu general data protection regulation) to provide anonymity [ , , ] and it will then turn into totally anonymous data for the possible treatment data use. for the analyses in this report, we used aggregated data computed by the mobile operator covering the period january th, to may th, . for each phone call, a tuple <n o , n i , t , a s , a e , d> is recorded, where n o and n i are pseudo-anonymous identifiers, respectively of the "caller" and the "callee"; t is a timestamp saying when the call was placed; a s and a e are the identifiers of the towers/antennas to which the caller was connected at the start and end of the call; finally, d is the call duration (e.g., in minutes). they are similar to cdrs, except that the communication is only between the antenna and the connected mobile phone, and an amount k of kilobytes is downloaded in the process. the format of xdr is, therefore, a tuple <n, t, a, k>. a windtre is one of the main mobile phone operators in italy, covering around % of the residential "human" mobile market. page origin-destination matrices -for a better matching with the available covid- data (number of positive cases and net reproduction number), we aggregated the municipality-to-municipality origin-destination matrices (ods) into province-to-province or region-to-region ods, in which each node represents an italian province or region. in particular, for each day, we compute both the out-flows, indicating the total number of people moving from a province/region to any other province/region, and the in-flows, indicating the total number of people moving to a province/ region from any other province/region. the trips between municipalities of the same province/region are aggregated into a self-flow, which indicates the province/region's internal mobility. for privacy reasons, we eliminate all out-, in-and self-flows with values lower than . as they are calculated by the operator, we store the daily municipality-to-municipality od matrices and the daily region-to-region ones into a relational dbms and access them through calls to a dedicated api. we normalize the self-, in-and out-flows by multiplying them by coefficients provided by the mobile phone operator, which indicate an estimation of market share for every municipality. after this transformation, we have an estimation of the real size of the mobility flow between each origin and destination municipality. for ease of readability, figures , and visualize the od matrix of flows between italian regions on february th (before the initiation of the national lockdown on march th), march th (during the lockdown), and may th (during phase ), respectively. we find that the od matrix becomes significantly more sparse during the lockdown and the phase , denoting a drastic reduction of the routes between italian regions. numerically, we estimate this sparsity through the network density, i.e., the proportion of the potential connections in a network that are actual connections. we find that network density halves during the lockdown: it decreases from d feb = . to d mar = . , indicating that the lockdown erases half of the possible connections between regions compared to the previous period. network density remains almost unchanged between the lockdown and the phase (d may = . ), presumably because movements between regions are still forbidden by law except for specific circumstances (e.g. commuting for work). these results clearly highlight the drastic change in the structure of the human mobility network between regions in the two periods taken into consideration. indeed, most of the regional out-flows go towards adjacent regions. for example, before the lockdown, most of lombardy's out-flow is directed towards veneto, piedmont and emilia-romagna (adjacent regions), and the rest of the out-flows distribute more or less uniformly across all other regions, both in the north and the south (figure ). in contrast, during the lockdown, the number of these more modest out-flows decreases substantially, and most of them disappear at all ( figure ). the width of the arrows is proportional to the flow between the two regions. the density of the flows network is d feb = . . on the left, numbers in parenthesis indicate the out-flow. on the right, numbers in parenthesis indicate the in-flow. the relationship between human mobility and viral transmissibility during the covid- epidemics in italy before lockdown the width of the arrows is proportional to the flow between the two regions. the density of the flows network is d mar = . . on the right, numbers in parenthesis indicate the in-flow. the width of the arrows is proportional to the flow between the two regions. the density of the flows network is d may = . . on the right, numbers in parenthesis indicate the in-flow. the relationship between human mobility and viral transmissibility during the covid- epidemics in italy flows between italian regions -another important aspect of the mobility of a region or a province, complementary to the volumes of incoming and outgoing flows, is the diversification of the provenance and the destination of people. specifically, we define the in-flow diversity of a province a as the shannon entropy of the in-flows to the province [ ] where p in is the number of provinces with non-null flow to province a, p(x) is the probability that the in-flow to province a comes from province x, and log(n) is a normalization factor where n= is the number of italian provinces. the outflow diversity of province a is computed similarly as: where p out is the number of provinces with non-null flow from province a, and p(x) is the probability that the out-flow from province a goes to province x. mobility diversity during the pre-lockdown and lockdown period has been studied in our first report [ ] . another important aspect of the mobility of a region or a province, complementary to the volumes of incoming and outgoing flows, is the diversification of the provenance and the destination of people. specifically, we define the in-flow diversity of a province as the shannon entropy of the in-flows to the a province [ ] : where is the number of provinces with non-null flow to province , is the probability that the p in a (x) p in-flow to province comes from province , and is a normalization factor where is the a x og(n ) l n = number of italian provinces. the out-flow diversity of province is computed similarly as: where is the number of provinces with non-null flow from province , and is the probability p out a (x) p that the out-flow from province goes to province . mobility diversity during the pre-lockdown and a x lockdown period has been studied in our first report [ ] . we compare the evolution of the out-, in-and self-flows with the evolution of the daily disease transmissibility in italian regions, measured in terms of the net reproduction number . the net r t reproduction number represents the mean number of secondary infections generated by one primary infector, in the presence of control interventions and human behavioural adaptations. when decreases r t below the epidemic threshold of , the number of new infections begins to decline. the estimates of were r t computed from the daily time series of new cases by date of symptom onset. case-based surveillance data used for estimating were collected by regional health authorities and collated by the istituto superiore di r t sanità using a secure online platform, according to a progressively harmonized track-record. data include, among other information, the place of residence, the date of symptom onset and the date of first hospital admission for laboratory-confirmed covid- cases [ ] . the distribution of the net reproduction number was estimated by applying a well-established statistical method [ ] [ ] [ ] , which is based on the r t knowledge of the distribution of the generation time and on the time series of cases. in particular, the posterior distribution of for any time point was estimated by applying the metropolis-hastings mcmc r t t sampling to a likelihood function defined as follows: another important aspect of the mobility of a region or a province, complementary to the volumes of incoming and outgoing flows, is the diversification of the provenance and the destination of people. specifically, we define the in-flow diversity of a province as the shannon entropy of the in-flows to the a province [ ] : where is the number of provinces with non-null flow to province , is the probability that the p in a (x) p in-flow to province comes from province , and is a normalization factor where is the a x og(n ) l n = number of italian provinces. the out-flow diversity of province is computed similarly as: a where is the number of provinces with non-null flow from province , and is the probability p out a (x) p that the out-flow from province goes to province . mobility diversity during the pre-lockdown and a x lockdown period has been studied in our first report [ ] . we compare the evolution of the out-, in-and self-flows with the evolution of the daily disease transmissibility in italian regions, measured in terms of the net reproduction number . the net r t reproduction number represents the mean number of secondary infections generated by one primary infector, in the presence of control interventions and human behavioural adaptations. when decreases r t below the epidemic threshold of , the number of new infections begins to decline. the estimates of were r t computed from the daily time series of new cases by date of symptom onset. case-based surveillance data used for estimating were collected by regional health authorities and collated by the istituto superiore di r t sanità using a secure online platform, according to a progressively harmonized track-record. data include, among other information, the place of residence, the date of symptom onset and the date of first hospital admission for laboratory-confirmed covid- cases [ ] . the distribution of the net reproduction number was estimated by applying a well-established statistical method [ ] [ ] [ ] , which is based on the r t knowledge of the distribution of the generation time and on the time series of cases. in particular, the posterior distribution of for any time point was estimated by applying the metropolis-hastings mcmc r t t sampling to a likelihood function defined as follows: we compare the evolution of the out-, in-and self-flows with the evolution of the daily disease transmissibility in italian regions, measured in terms of the net reproduction number r t . the net reproduction number represents the mean number of secondary infections generated by one primary infector, in the presence of control interventions and human behavioural adaptations. when r t decreases below the epidemic threshold of , the number of new infections begins to decline. the estimates of r t were computed from the daily time series of new cases by date of symptom onset. case-based surveillance data used for estimating r t were collected by regional health authorities and collated by the istituto superiore di sanità using a secure online platform, according to a progressively harmonized track-record. data include, among other information, the place of residence, the date of symptom onset and the date of first hospital admission for laboratory-confirmed covid- cases [ ] . the distribution of the net reproduction number r t was estimated by applying a well-established statistical method [ ] [ ] [ ] , which is based on the knowledge of the distribution of the generation time and on the time series of cases. in particular, the posterior distribution of r t for any time point t was estimated by applying the metropolis-hastings mcmc sampling to a likelihood function defined as follows: where p(κ;λ) is the probability mass function of a poisson distribution (i.e., the probability of observing κ events if these events occur with rate λ). c(t) is the daily number of new cases having symptom onset at time t; r t is the net reproduction number at time t to be estimated; φ(s) is the probability distribution density of the generation time evaluated at time s. as a proxy for the distribution of the generation time, we used the distribution of the serial interval, estimated from the analysis of contact tracing data in lombardy [ ] , i.e., a gamma function with shape . and rate . , having a mean of . days. this estimate is within the range of other available estimates for sars-cov- infections, i.e. between and . days [ ] [ ] [ ] . we compare the evolution of the out-, in-and self-flows with the evolution of the daily disease transmissibility in italian regions, measured in terms of the net reproduction number . the net r t reproduction number represents the mean number of secondary infections generated by one primary infector, in the presence of control interventions and human behavioural adaptations. when decreases r t below the epidemic threshold of , the number of new infections begins to decline. the estimates of were r t computed from the daily time series of new cases by date of symptom onset. case-based surveillance data used for estimating were collected by regional health authorities and collated by the istituto superiore di r t sanità using a secure online platform, according to a progressively harmonized track-record. data include, among other information, the place of residence, the date of symptom onset and the date of first hospital admission for laboratory-confirmed covid- cases [ ] . the distribution of the net reproduction number was estimated by applying a well-established statistical method [ ] [ ] [ ] , which is based on the r t knowledge of the distribution of the generation time and on the time series of cases. in particular, the posterior distribution of for any time point was estimated by applying the metropolis-hastings mcmc r t t sampling to a likelihood function defined as follows: is the probability mass function of a poisson distribution (i.e., the probability of observing (κ; ) p λ κ events if these events occur with rate ). λ • is the daily number of new cases having symptom onset at time ; is the net reproduction number at time to be estimated; is the probability distribution density of the generation time evaluated at time . φ (s) s as a proxy for the distribution of the generation time, we used the distribution of the serial interval, estimated from the analysis of contact tracing data in lombardy [ ] , i.e., a gamma function with shape . and rate . , having a mean of . days. this estimate is within the range of other available estimates for sars-cov- infections, i.e. between and . days [ ] [ ] [ ] . the number of covid- positive cases is provided by protezione civile, the italian public institution in charge of monitoring the covid- emergency. they collect data from every italian administrative region and make them available on a public github repository [ ] . for each region, we focus on the number of new positive cases per day. specifically, given a day , we compute the average of values over the four g days before and the four days after . g the relationship between human mobility and viral transmissibility during the covid- epidemics in italy page net reproduction number r t epidemiologic data - figure and show the evolution of the mobility self-flows (blue curves), the net reproduction number (orange curves) and the number of positive cases (grey curves) for the northern regions and central-southern regions, respectively. these curves reveal numerous interesting insights. all regions have a net decrease of the self-flow soon after the first national lockdown (march th [ ] ). the flows stabilized on the new, reduced volume after about one week. subsequent restriction ordinances, such as the closing of non-essential economic activities on march th [ ] , had a minor impact on the reduction of self-flows. for almost all regions, we find an increase of the self-flow since the start of phase on may th. this behaviour is particularly pronounced for emilia-romagna, toscana, puglia, and lazio. further investigation is needed to understand why these regions had such a marked increase. interestingly, we find a slight increase in the self-flows approaching may th, the starting of "phase " during which a wider range of movement within regions has been allowed by the government. we interpret this result as a progressive, although slight, relaxation of compliance with the mobility limitations imposed by the lockdown. the case of molise is different and particularly compelling: it is indeed the only region for which the self-flow decreases since may th ( figure ). the reason for this decrease may be due to news media coverage about a funeral on april th, attended by a large number of people, which resulted in a large local outbreak. this may have induced parts of the molise population to self-restrict movements during the following days . the relationship between human mobility and viral transmissibility during the covid- epidemics in italy page the number of covid- positive cases is provided by protezione civile, the italian public institution in charge of monitoring the covid- emergency. they collect data from every italian administrative region and make them available on a public github repository [ ] . for each region, we focus on the number of new positive cases per day. specifically, given a day g, we compute the average of values over the four days before and the four days after g. https://www.ilfattoquotidiano.it/ / / /coronavirus-nuovo-focolaio-in-molise- -contagi-a-campobasso-legati-a-un-funerale-della-comunita-rom/ / although the date when r t = for the first time varies from region to region, for all regions r t decreases concurrently with the net decrease of self-flows due to the beginning of the national lockdown (figures and ) ., highlighting the importance of the government intervention. note that r t starts taking values lower than since march th, when the self-flows stabilize on the new, reduced volume. from that moment on, self-flows remain stable. still, the r t continues decreasing. this may be due to other ordinances by local and national governments related to the wearing of masks and gloves in public areas, social distancing and ban on gatherings --and possibly to other factors to be further investigated evolution between january th and may th of self-flow (blue curve), net reproduction number r t (orange curve) and moving average of the number of confirmed sars-cov- infections (grey curve) in the northern regions of italy. for each day, we plot the average of the r t values of the three days before and after that day. the orange-shaded area indicates r t > . the value of the grey curve for a given day is computed as the average of the number of confirmed sars-cov- infections over the four days before and the four days after that day. the vertical dashed lines indicate the beginning of the national lockdown (ld, march th, ), the closing of non-essential economic activities (cna, march th, ) and the partial restarting of economic activities and within-region movements ("ph ", may th, ). the area in white indicates the period before mobility reduction (mr) in that region. note that the beginning of mr does not necessarily coincide with the national lockdown (e.g see lombardia) . evolution between january th and may th of self-flows (blue curve), net reproduction numbers (orange curve) and moving average of the number of confirmed sars-cov- infections (grey curve) in the central-southern regions of italy. for each day, we plot the average of the r t values of the three days before and after that day. the orange-shaded area indicates r t > . the value of the grey curve for a given day is computed as the average of the number of confirmed sars-cov- infections over the four days before and the four days after that day. the vertical dashed lines indicate the beginning of the national lockdown (ld, march th, ), the closing of non-essential economic activities (cna, march th, ) and the partial restarting of economic activities and within-region movements ("ph ", may th, ). the area in white indicates the period before mobility reduction (mr) in that region. note that the beginning of mr does not necessarily coincide with the national lockdown. can we estimate the value of r t of italian regions (or provinces) from the mobility of their population? this is a complex question, which we address here only preliminarily --focusing on in-and out-flow diversity, switching from the regional level used for the analysis to the level of provinces. a first cut regression model for estimating the daily r t as a function of mobility diversity is specified as follows: the beginning of the national lockdown (ld, march th, ), the closing of non-necessary economic activities (cna, march th, ) and the partial restarting of economic activities and within-region movements ("ph ", may th, ). the area in white indicates the period in which there is a drastic decrease of both self-flows and net reproduction numbers. can we estimate the value of of italian regions (or provinces) from the mobility of their population? this is r t a complex question, which we address here only preliminarily --focusing on in-and out-flow diversity, switching from the regional level used for the analysis to the level of provinces. a first cut regression model for estimating the daily as a function of mobility diversity is specified as follows: indicates the fixed effect of province (to control for the non-observable heterogeneity between α i i the provinces), and are the daily in-and out-flow diversities and for indiversity it outdiversity it (i) e in (i) e out province on the same day , indicates the fixed effect of day , are stochastic errors residuals, and i t δ t t ε it (the outcome variable) is the net reproduction number estimated for day and province . we considered of italian provinces for which a sufficient number of symptomatic cases had been recorded for a reliable computation of the estimate. as an outcome of regressing the daily on the in-and r t out-flow diversities of the same day, we find that contributes to reduce the and outdiversity it r t indiversity it to increase it, but these effects are not statistically significant. the picture changes substantially if we introduce time lags, e.g., through the model: in which the regressors cover the entire week before the measurement of . in table (column ) , we r t consider the fixed effects of province only, while in column we add the fixed effect for day. for , we j = find that contributes to increase the and to reduce it, with statistical indiversity it− r t outdiversity it− significance at %. if we increase the lagging period past one week ( , we find a stronger statistical significance (table , ≥ ) j columns and ). for out-flow diversity, the closer the day is to the date of the ( ) the stronger the r t j < impact on contagion. conversely, for in-flow diversity, the further away the day is to the date of the ( r t ≥ j ) the stronger the impact on contagion. figure reports the temporal profile of the coefficients estimated in column of where α i indicates the fixed effect of province i (to control for the non-observable heterogeneity between the provinces), indiversity it and outdiversity it are the daily in-and out-flow diversities e in (i) and e out (i) for province i on day t, δ t indicates the fixed effect of day t, ε it are stochastic errors residuals, and r it (the outcome variable) is the net reproduction number estimated for day t and province i. we considered of italian provinces for which a sufficient number of symptomatic cases had been recorded for a reliable computation of the estimate. as an outcome of regressing the daily r t on the in-and out-flow diversities of the same day, we find that outdiversity it contributes to reduce the r t and indiversity it to increase it, but these effects are not statistically significant. the picture changes substantially if we introduce time lags, e.g., through the model: the picture changes substantially if we introduce lags, e.g. through the model: the beginning of the national lockdown (ld, march th, ), the closing of non-necessary economic activities (cna, march th, ) and the partial restarting of economic activities and within-region movements ("ph ", may th, ). the area in white indicates the period in which there is a drastic decrease of both self-flows and net reproduction numbers. can we estimate the value of of italian regions (or provinces) from the mobility of their population? this is r t a complex question, which we address here only preliminarily --focusing on in-and out-flow diversity, switching from the regional level used for the analysis to the level of provinces. a first cut regression model for estimating the daily as a function of mobility diversity is specified as follows: indicates the fixed effect of province (to control for the non-observable heterogeneity between α i i the provinces), and are the daily in-and out-flow diversities and for indiversity it outdiversity it (i) e in (i) e out province on the same day , indicates the fixed effect of day , are stochastic errors residuals, and i t δ t t ε it (the outcome variable) is the net reproduction number estimated for day and province . we considered of italian provinces for which a sufficient number of symptomatic cases had been recorded for a reliable computation of the estimate. as an outcome of regressing the daily on the in-and r t out-flow diversities of the same day, we find that contributes to reduce the and outdiversity it r t indiversity it to increase it, but these effects are not statistically significant. the picture changes substantially if we introduce time lags, e.g., through the model: in which the regressors cover the entire week before the measurement of . in table (column ) , we r t consider the fixed effects of province only, while in column we add the fixed effect for day. for , we j = find that contributes to increase the and to reduce it, with statistical indiversity it− r t outdiversity it− significance at %. if we increase the lagging period past one week ( , we find a stronger statistical significance (table , ≥ ) j columns and ). for out-flow diversity, the closer the day is to the date of the ( ) the stronger the r t j < impact on contagion. conversely, for in-flow diversity, the further away the day is to the date of the ( r t ≥ j ) the stronger the impact on contagion. figure reports the temporal profile of the coefficients estimated in column of in which the regressors cover the entire week before the measurement of r t . in table (column ), we consider the fixed effects of province only, while in column we add the fixed effect for day. for j= , we find that indiversity it- contributes to increase the r it and outdiversity it- to reduce it, with statistical significance at %. if we increase the lagging period past one week (j ≥ ), we find a stronger statistical significance (table , columns and ). for out-flow diversity, the closer the day is to the date of the r t (j< ) the stronger the impact on contagion. conversely, for in-flow diversity, the further away the day is to the date of the r t (j ≥ ) the stronger the impact on contagion. figure reports the temporal profile of the coefficients estimated in column of robust standard errors in brackets -*** p< . , ** p< . , * p< . robust errors clustered by provincecountry fixed effects included. the relationship between human mobility and viral transmissibility during the covid- epidemics in italy influence of variables on disease transmissibility, measured as the net reproduction number r t with entropic measures we cannot control for movements within each province. if we replace mobility diversity with the in-flows, out-folws and self-flows of each province (number of individuals registered as either changing provinces or moving within the province), we find similar statistical significance for mobility in the previous week, confirming that outflow of people contributes to contagion reduction, while arrival of people from outside the province or internal mobility raises it. while a more detailed modeling of the association between contagion and mobility is certainly needed, and may lead to additional insights, these preliminary results --together with the analysis of self-flows presented in the previous section, provide clear evidence for a critical role of human mobility in the spatio-temporal unfolding of the epidemics. we go one step further in our analysis of the relationships between mobility flows and contagion by computing two quantities, for each region: ) the delay in mobility reduction, i.e., the number of days in which r t > before the mobility flows of a region decrease by at least % w.r.t. the usual (pre-epidemics) weekly mobility, observed over january and the first two weeks of february, and ) the total number of reported sars-cov- infections per k inhabitants in the region (as of may th, ). the date of mobility reduction below % for each region is indicated as the mr black vertical line in the time series of figures and . figure shows a scatter plot of the two quantities for all regions, where the size of the circle of each region is proportional to the total number of reported infections in the observation period; the positive correlation between the two quantities is robust (pearson coefficient = . , p < . , r = . ), suggesting that larger delays could have induced heavier spreading of the virus. this is a strong evidence that timely lockdowns are instrumental for better containment of the contagion. two further considerations follow. the mobility reduction in lombardy started around days after the first day in which r t > , leading to the highest number of positive cases per inhabitant in italy. similarly, for other regions severely affected by the virus, such as liguria, emilia-romagna and piedmont, the mobility reduction started around, respectively, and days after the first day in which r t > . the central-western regions in north italy regions lie above the dashed regression line, in the top right part of the plot. the regions below this line, in the bottom right part of the plot, were more effective than the regions above the line in containing the contagion, despite the delay in lockdown of days or more, such as veneto, lazio, and tuscany. this fact may be explained by several factors, including the effectiveness of the epidemic surveillance, the intensity of the testing and tracing strategy adopted, the capacity of outbreak containment, and also the absolute number of cases when r t jumps above . on the other hand, for southern regions the mobility reduction started with around days of delay (molise, basilicata) or around days (campania, puglia, sicily, calabria), which presumably was effective in containing the spread of the virus: all southern regions (with the only exception of molise) are below the regression line in the bottom left part of the plot (low number of infections per k inhabitants). central-southern regions are the ones who benefited the most from the lockdown, presumably because it started more timely. this brings further evidence of the effectiveness of the lockdown. in the scatter plot, the horizontal axis has the number of days between the first time r t > and the beginning of the national lockdown. the vertical axis has the cumulative incidence of confirmed sars-cov- infections per k inhabitants (as of may th, ). the size of the circles is proportional to the total number of positive cases in the period (pearson coefficient= . , p< . , r = . ). the relationship between human mobility and viral transmissibility during the covid- epidemics in italy page our combined analysis of mobility and epidemics highlighted a striking relation between the negative variation of movement fluxes and the negative variation of the net reproduction number, in all italian regions, in the time interval of approximately one week, from march th till march th, during the transition between the two mobility modalities. during this week, the two curves gracefully overlap; at the end of this week, the country has reached a new "stable" mobility regime, approximately at % of the pre-lockdown level. the two curves exhibit the same pattern everywhere, both at regional and provincial level, with minimal temporal lags. we call this phenomenon, represented schematically in figure , the "epi-mob" pattern. mobility, the blue curve, is a "switch" between two very different levels, before and during the lockdown, with an exponential fall from the first to the second; the epidemics is a peaked distribution with an exponential growth and fall, overlapping with the "switch" during the fall. the presumable effectiveness of the lockdown for the containment of the epidemics is further substantiated by the pattern in figure , relating the timeliness of lockdown in every region with the total number of infected individuals per k inhabitants. we can also quantify the time needed to "switch off" the country mobility (approx. week to reach the new lower regime) and the time needed to bring the net reproduction number below (again, approx. week). notice that r t continues to slowly decrease during lockdown, and also that at the beginning of phase (lockdown exit), when mobility begins to rise again, r t does not jump into a new uncontrolled growth, at least until may th, the last day of observation in this report. this is probably due to the non-pharmaceutical interventions in place, including the increased compliance of people to use personal protective equipment and respect social distancing (compared to the pre-lockdown phase). another factor limiting the increase of r t in the phase could also be related to an increased ability to trace, test and isolate infected individuals. we have also shown that a simple regression model provides reasonable estimates of the r t in a given region (or province) as a function of the in-and out-flows. clearly, the accuracy of the estimation is influenced by the changes in containment interventions and in citizens' behaviour towards prevention measures. consequently, learning a regression model for r t during pre-lockdown and early lockdown and applying the model for predicting the r t during phase would probably lead to overestimating it; however, such worstcase scenario might be useful for comparison with the actual r t measured, as a mean to evaluate the effectiveness of the containment policy in act and citizens' social behavior. an interesting point that calls for further study is how to continuously learn an estimation model for the r t , whose accuracy is continuously monitored, to the purpose of nowcasting the r t , shortening as much as possible the lag of time to wait until the r t becomes known. as a conclusion, we believe that this study demonstrated the value of "big" mobility data, a detailed proxy of human behavior available every day in real time, to the purpose of refining our understanding of the dynamics of the epidemics, reasoning on the effectiveness of policy choices for non-pharmaceutical interventions and on citizens' compliance to social distancing measures, and help monitoring key epidemic indicators to inform choices as the epidemics unfolds in the coming months. during the first week of lockdown, the two curves describing mobility flows and net reproduction number gracefully overlap. at the end of this week, the country has reached a new "stable" mobility regime, approximately at % of the pre-lockdown level. the two curves exhibit the same pattern everywhere, both at regional and provincial level, with minimal temporal lags. the relationship between human mobility and viral transmissibility during the covid- epidemics in italy page appendix (mobility inflow) - mar apr may for each day, we plot the average of the r t values of the three days before and after that day. the orange-shaded area indicates r t > . the value of the grey curve for a given day is computed as the average of the number of confirmed sars-cov- infections over the four days before and the four days after that day. the vertical dashed lines indicate the beginning of the national lockdown (ld, march th, ), the closing of non-essential economic activities (cna, march th, ) and the partial restarting of economic activities and within-region movements ("ph ", may th, ). the area in white indicates the period before mobility reduction (mr) in that region. note that the beginning of mr does not necessarily coincide with the national lockdown. evolution between january th and may th of out-flow (blue curve), net reproduction number r t (orange curve) and moving average of the number of confirmed sars-cov- infections (grey curve) in the northern regions of italy. for each day, we plot the average of the r t values of the three days before and after that day. the orange-shaded area indicates r t > . the value of the grey curve for a given day is computed as the average of the number of confirmed sars-cov- infections over the four days before and the four days after that day. the vertical dashed lines indicate the beginning of the national lockdown (ld, march th, ), the closing of non-essential economic activities (cna, march th, ) and the partial restarting of economic activities and within-region movements ("ph ", may th, ). the area in white indicates the period before mobility reduction (mr) in that region. note that the beginning of mr does not necessarily coincide with the national lockdown (e.g. see lombardia). evolution between january th and may th of out-flows (blue curve), net reproduction numbers (orange curve) and moving average of the number of confirmed sars-cov- infections (grey curve) in the central-southern regions of italy. for each day, we plot the average of the r t values of the three days before and after that day. the orange-shaded area indicates r t > . the value of the grey curve for a given day is computed as the average of the number of confirmed sars-cov- infections over the four days before and the four days after that day. the vertical dashed lines indicate the beginning of the national lockdown (ld, march th, ), the closing of non-essential economic activities (cna, march th, ) and the partial restarting of economic activities and within-region movements ("ph ", may th, ). the area in white indicates the period before mobility reduction (mr) in that region. note that the beginning of mr does not necessarily coincide with the national lockdown. mobile phone data and covid- : missing an opportunity aggregated mobility data could help fight covid- measuring levels of activity in a changing city: a study using cellphone data streams on the privacy-conscientious use of mobile phone data a survey of results on mobile phone datasets analysis returners and explorers dichotomy in human mobility primule: privacy risk mitigation for user profiles a data mining approach to assess privacy risk in human mobility data an analytical framework to nowcast well-being using mobile phone data covid- outbreak response: first assessment of mobility changes in italy following lockdown so) big data and the transformation of the city epidemiological characteristics of covid- cases in italy and estimates of the reproductive numbers one month into the epidemic who ebola response team. ebola virus disease in west africa-the first the relationship between human mobility and viral transmissibility during the covid- epidemics in italy page months of the epidemic and forward projections a new framework and software to estimate time-varying reproduction numbers during epidemics measurability of the epidemic reproduction number in data-driven contact networks serial interval of novel coronavirus (covid- ) infections estimating clinical severity of covid- from the transmission dynamics in wuhan, china early transmission dynamics in wuhan, china, of novel coronavirus-infected pneumonia mobile phone data analytics against the covid- epidemics in italy: flow diversity and local job markets during the national lockdown human mobility in response to covid- in france ulteriori disposizioni attuative del decreto-legge febbraio , n. , recante misure urgenti in materia di contenimento e gestione dell'emergenza epidemiologica da covid- , applicabili sull'intero territorio nazionale misure di potenziamento del servizio sanitario nazionale e di sostegno economico per famiglie, lavoratori e imprese connesse all'emergenza epidemiologica da covid- key: cord- - zgwj fa authors: strafella, claudia; caputo, valerio; termine, andrea; barati, shila; gambardella, stefano; borgiani, paola; caltagirone, carlo; novelli, giuseppe; giardina, emiliano; cascella, raffaella title: analysis of ace genetic variability among populations highlights a possible link with covid- -related neurological complications date: - - journal: genes (basel) doi: . /genes sha: doc_id: cord_uid: zgwj fa angiotensin-converting enzyme (ace ) has been recognized as the entry receptor of the novel severe acute respiratory syndrome coronavirus (sars-cov- ). structural and sequence variants in ace gene may affect its expression in different tissues and determine a differential response to sars-cov- infection and the covid- -related phenotype. the present study investigated the genetic variability of ace in terms of single nucleotide variants (snvs), copy number variations (cnvs), and expression quantitative loci (eqtls) in a cohort of individuals representative of the general italian population. the analysis identified five snvs (rs , rs , rs , rs , and rs ) in the italian cohort. of them, rs and rs displayed a significant different frequency distribution in the italian population with respect to worldwide population. the eqtls analysis located in and targeting ace revealed a high distribution of eqtl variants in different brain tissues, suggesting a possible link between ace genetic variability and the neurological complications in patients with covid- . further research is needed to clarify the possible relationship between ace expression and the susceptibility to neurological complications in patients with covid- . in fact, patients at higher risk of neurological involvement may need different monitoring and treatment strategies in order to prevent severe, permanent brain injury. angiotensin-converting enzyme (ace ) has recently caught the attention of the scientific community, since it has been recognized as the entry receptor of the novel pathogenic severe acute respiratory syndrome coronavirus (sars-cov- ) [ ] . ace is a protein encoded by its homologous gene (ace ), which maps on chromosome x (xp . ) and consists of exons. ace is classified as an the study was performed by utilizing dna samples representative of the italian general population, which were partially available at the genomic medicine laboratory of santa lucia foundation hospital and partially derived from international databases. the italian cohort of samples was composed of samples analyzed by a comparative genomic hybridization (acgh) array for assessing the presence of structural genomic variations, and samples utilized for identifying common and rare variants located in the coding or splice site regions of the genome. the genetic data referred to these samples were partially derived by whole exome sequencing (wes), available at the genomic medicine laboratory of irccs santa lucia foundation hospital and partially extracted by the ensembl database [ ] [ ] [ ] . italian patients had an average age of ± years and a female/male ratio of : . the research was approved by the ethics committee of santa lucia foundation of rome (ce/prog. , approved on march ), and was performed according to the declaration of helsinki. the participants provided signed informed consent. the cnv analysis was performed by chromosome analysis suite (chas) . (affymetrix, santa clara, ca, usa) using the cytoscan k_array single sample analysis "na _hg " as reference file and an average resolution of kb. concerning snvs, we decided to analyze snvs (namely snps and indels) located within the coding and splice site dna regions, because they are most likely to affect protein function. we therefore selected the snvs of interest by extracting the variants localized in the exonic and splice site regions of ace , whose frequency data were available from the genomes and gnomad databases [ ] [ ] [ ] [ ] . the frequency cutoff for selecting the genetic variants of interest was set at minor allele frequency (maf) > . . this approach allowed for selecting putative variants located within the ace sequence (table s ). the ensembl database [ ] was also utilized to download the allele frequency data of the snvs of interest in the worldwide populations, in order to compare the frequencies between the italian cohort and the frequencies observed in african, american, asian, and european populations. the presence of the snvs in the italian samples was evaluated by analyzing the output file derived by the wes and ensembl databases. for wes results, a coverage of x was considered for the analysis of ace sequence. the variant caller files (vcfs) obtained by wes analysis were first scanned with vcfr [ ] , and then subjected to analysis by "genomic variants filtering by deep learning models in ngs" (garfield-ngs) [ ] . in particular, vcfr is a package that enables visualization, manipulation, and performing quality control of vcf data [ ] . garfield-ngs is an informatics tool, which relies on deep learning models to dissect false and true variants in exome sequencing experiments [ ] . the allelic frequency distribution of the detected snvs and the existence of significant differences among italian and worldwide populations were calculated by statistical tools. all statistical analyses were performed in an r environment [ ] . a two-sided fisher's exact test and a p-value (p) were calculated in order to assess the different allelic distributions of detected snvs in the italian cohorts with respect to the other populations. the significance threshold was set at p < . . in addition, multiple testing correction (false discovery rate) was performed by calculating the q-value (q) [ ] and setting a significance cutoff at at q < . . considering that ace maps to the x chromosome, statistical analysis (two-sided fisher's exact test) was also performed by stratifying the cohorts according to gender to evaluate sex-related effects. moreover, the snvs detected in the italian population were subjected to bioinformatic predictive analysis to assess their potential impact on ace protein function and splicing mechanisms. to this purpose, varsite [ ] , human splicing finder (hsf) [ ] , and uniprot database [ ] were interrogated. varsite analyzes and predicts the effect of amino acid changes on the protein structure. hsf evaluates the effects of variants on the splicing mechanisms. moreover, uniprot annotation database was utilized to retrieve the topological and functional domains organization of proteins. concerning eqtl analysis, the genotype-tissue expression (gtex) database [ ] was utilized to retrieve the eqtls variants with a significant effect on ace expression in different tissues, and biomart tool [ ] was used to extract the significant eqtls distributed on the basis of the affected tissue. gtex database is a public resource that enables us to study tissue-specific gene expression and regulation, as well as their relationship to genetic variation [ ] . the significance threshold for eqtl analysis was set at p < . . biomart is a web-based tool that allows one to extract data in a uniform way and filter them for different queries [ ] . in this study, biomart was utilized to extract and filter the eqtls significantly distributed in brain tissues. the entire analytical workflow of the study is illustrated in figure . moreover, all raw data utilized for statistical and computational analysis are available at the following link: https://github.com/andreater/data-and-rmd-for-ace -article. the final goal of the study has been the research of variants potentially affecting ace expression and function, which may contribute to sars-cov- spreading among worldwide populations, and may have a clinical significance regarding the clinical variability and outcome displayed by patients with covid- . the analysis of cnvs in ace did not report any significant variation in our study cohort, ruling out that frequent copy number variations could potentially impact ace expression. concerning snvs, though, the screening of the snvs previously selected was performed on italian samples, and revealed the presence of five variants: rs (c/t), rs (t/c), rs (a/-), rs (t/c), and rs (c/t). these variants presented some differences in the frequency distribution in the italian cohort, with respect to worldwide populations (table and figure ). as expected, the italian cohort showed overlapping frequency distributions with the european population, whereas significant differences were observed with respect to african, american, and asian populations ( table ). the statistical significance was also confirmed after correction for multiple testing (q-value) in these cohorts ( table ). the final goal of the study has been the research of variants potentially affecting ace expression and function, which may contribute to sars-cov- spreading among worldwide populations, and may have a clinical significance regarding the clinical variability and outcome displayed by patients with covid- . the analysis of cnvs in ace did not report any significant variation in our study cohort, ruling out that frequent copy number variations could potentially impact ace expression. concerning snvs, though, the screening of the snvs previously selected was performed on italian samples, and revealed the presence of five variants: rs (c/t), rs (t/c), rs (a/-), rs (t/c), and rs (c/t). these variants presented some differences in the frequency distribution in the italian cohort, with respect to worldwide populations (table and figure ). as expected, the italian cohort showed overlapping frequency distributions with the european population, whereas significant differences were observed with respect to african, american, and asian populations ( table ). the statistical significance was also confirmed after correction for multiple testing (q-value) in these cohorts ( table ). in particular, the rs (c/t) is a synonymous variant, whose allelic frequency resulted in being significantly different from the frequencies observed in african and asian populations, whereas it overlapped with frequencies recorded in the european and american groups (tables and ). in fact, the rs appeared to be more frequent in italian, european, and american populations, compared to the very low frequency observed in the african and asian cohorts ( figure ). concerning rs , the italian population showed overlapping frequencies with american, european, and asian populations but not with the african cohort, in which the variant allele was almost absent (figure ). the frequency of rs in the italian population overlaps with all investigated populations except for the american group, which reported a slightly higher frequency ( figure ). the rs did not display significant differences in terms of frequency distribution of wild-type and variant alleles between the italian and the other worldwide populations ( table ). the variant allele was extremely rare among all populations, with a slight increase in the frequency observed in the italian cohort (table , figure ). however, none of these variants (rs , rs , and rs ) reported significantly different distributions among populations after correction for multiple testing (table ) . concerning rs , this is the only variant showing a higher frequency, with respect to other variants previously discussed, that appeared to be very rare among the investigated populations ( figure ). concerning the frequency of rs in the italian population, it was found to be significantly different with respect to the african, american, asian, and even european population (table ) . after multiple tests, the significance was maintained for american and asian populations ( table ). in particular, the variant allele of rs showed the lowest frequency in the italian cohort compared to the other populations (table , figure ). the frequency of the five snvs were similarly distributed among male and female patients in each population, as well as between different populations, indicating that there are no gender effects underlying the frequency distribution of ace variants. altogether, the analysis of the frequency distribution of ace coding variants in the italian population with respect to worldwide populations showed a low rate of coding variants in ace gene in the italian cohort, suggesting that the susceptibility to sars-cov- infection may depend from other genetic variants outside ace , or in other genes. on this subject, it would be interesting to investigate the possible contribution of non-coding variants located in the regulatory regions (such as promoters and enhancers) of ace to the risk for sars-cov- infection. in addition, further research should be performed to assess possible population-specific effects that may explain the variable susceptibility to sars-cov- infection in the italian as well as in the worldwide populations. consistent with our findings, a similar study on the chinese population investigated the genetic variability of ace in their population, finding a different frequency distribution of ace variants with respect to the other populations [ ] . moreover, they found a higher allelic frequency of eqtl variants, which is associated with higher ace expression in tissues, suggesting a different susceptibility or response to sars-cov- infection with respect to other populations under similar conditions [ ] . however, these data are not sufficient to demonstrate direct evidence of a relationship between ace genetic variants and a differential susceptibility to sars-cov- infection among populations. in fact, data obtained on italian and chinese populations should be replicated in larger cohorts and in case/control studies, in order to assess their potential association with susceptibility to sars-cov- infection. moreover, further functional studies should be carried out to demonstrate and explain these eventual associations. concerning the functional analysis of the five snv variants identified in the italian cohort, most results were inconclusive or not significant to predict the functional impact of such variants on the resulting proteins. the rs is a synonymous variant, so it does not result in an amino acid change, and thus it is unlikely to affect the protein function. moreover, it is located in a region coding for the transmembrane portion of the protein, which normally does not interact with the sars-cov- s protein. the rs is a missense variant, resulting in the amino acid change from asparagine (asn) with a neutral side chain to aspartate carrying a negatively charged side chain, which is therefore more hydrophilic. however, interrogation of varsite showed that the change from asn to asp is not a large one, indicating that it may or may not result in a change to the protein's function. the prediction analysis of the impact of rs on protein function was therefore inconclusive, so it is not possible to predict the functional impact of this variant at the moment. however, it is interesting to note that the rs is localized in the region coding for the extracellular portion of ace , which normally interacts with the sars-cov- s protein. it will be therefore interesting to evaluate the impact of this variant by functional experiments in future. the rs variant is a single nucleotide deletion and is located in a splicing intronic region. therefore, it has been subjected to hsf analysis, in order to test the variant for a potential alteration of splicing. the hsf interrogation showed that the variant allele of rs may create an exonic splicing silencer (ess) site, but it is not significant, and therefore it probably does not affect splicing. the rs is a synonymous variant localized in a region encoding the extracellular portion of the protein, which interacts with the s protein of sars-cov- . interestingly, the interrogation of hsf reported that this variant may affect the splicing by activating an exonic cryptic acceptor site or altering an exonic splicing enhancer (ese) site. these findings suggest that rs should be further investigated, together with other variants that have a larger effect on ace function. the rs variant was a variant located in the splice site region of ace . however, the prediction analysis by hsf did not reveal significant splicing alterations. successively, all previously discussed variants were evaluated as potential eqtls in the gtex database, and only rs was classified as a significant eqtl in several brain tissues, namely the amygdala, anterior cingulate cortex, basal ganglia, cortex, cerebellum, hippocampus, and hypothalamus ( figure ). genes , , x for peer review of so it is not possible to predict the functional impact of this variant at the moment. however, it is interesting to note that the rs is localized in the region coding for the extracellular portion of ace , which normally interacts with the sars-cov- s protein. it will be therefore interesting to evaluate the impact of this variant by functional experiments in future. the rs variant is a single nucleotide deletion and is located in a splicing intronic region. therefore, it has been subjected to hsf analysis, in order to test the variant for a potential alteration of splicing. the hsf interrogation showed that the variant allele of rs may create an exonic splicing silencer (ess) site, but it is not significant, and therefore it probably does not affect splicing. the rs is a synonymous variant localized in a region encoding the extracellular portion of the protein, which interacts with the s protein of sars-cov- . interestingly, the interrogation of hsf reported that this variant may affect the splicing by activating an exonic cryptic acceptor site or altering an exonic splicing enhancer (ese) site. these findings suggest that rs should be further investigated, together with other variants that have a larger effect on ace function. the rs variant was a variant located in the splice site region of ace . however, the prediction analysis by hsf did not reveal significant splicing alterations. successively, all previously discussed variants were evaluated as potential eqtls in the gtex database, and only rs was classified as a significant eqtl in several brain tissues, namely the amygdala, anterior cingulate cortex, basal ganglia, cortex, cerebellum, hippocampus, and hypothalamus ( figure ). violin plots represent the correlation between rs genotypes and ace mrna expression in different brain tissues. the reports on the x-axis are the genotypes, with the correspondent counts in brackets. on the y-axis, the normalized expression of ace is reported. moreover, a p-value of for the variant significance is reported. the figure has been obtained from genotype-tissue expression (gtex) [ ] . as shown in figure , the homozygous genotype for the variant allele may increase the expression of ace in multiple brain tissues, and consequently, may affect ace functions in the brain. this finding suggests that the genetic variability of ace may have a greater impact on covid- -related symptoms and sars-cov- tissue tropism, rather than on the susceptibility to sars-cov- infection. on this subject, evidence of genetic variants in tmprss and a large presence of eqtls in the lung may suggest that the genetic variability of tmprss might have a role in figure . violin plots represent the correlation between rs genotypes and ace mrna expression in different brain tissues. the reports on the x-axis are the genotypes, with the correspondent counts in brackets. on the y-axis, the normalized expression of ace is reported. moreover, a p-value of for the variant significance is reported. the figure has been obtained from genotype-tissue expression (gtex) [ ] . as shown in figure , the homozygous genotype for the variant allele may increase the expression of ace in multiple brain tissues, and consequently, may affect ace functions in the brain. this finding suggests that the genetic variability of ace may have a greater impact on covid- -related symptoms and sars-cov- tissue tropism, rather than on the susceptibility to sars-cov- infection. on this subject, evidence of genetic variants in tmprss and a large presence of eqtls in the lung may suggest that the genetic variability of tmprss might have a role in determining the different susceptibility to sars-cov- infection among populations. however, these are preliminary observations, which have to be confirmed by further investigations. considering the broader effect of rs in multiple brain tissues, we decided to look at the distribution of the eqtl variants located in and targeting ace in the different brain tissues, which may thereby affect ace expression, and in turn, contribute to the neurological symptoms and complications observed in patients with covid- . interestingly, literature studies have highlighted the crucial role of ace in brain physiology and pathophysiology, including marked regulatory effects on blood pressure, cardiac hypertrophy, stress response, anxiety, cognition, and brain injury [ , , ] . the eqtl analysis on the gtex portal and biomart allowed identification of significant eqtls, which have been predicted to affect ace expression in the brain at different levels ( table s ). most of them ( eqtls) have a significant effect on multiple brain tissues, suggesting that they may affect ace -related brain functions as a whole. six eqtls instead showed a more tissue-specific effect, indicating that they may be involved in the alteration of brain functions regulated by restricted areas of the brain. interestingly, the mostly enriched tissues with significant ace -associated eqtls were the basal ganglia, cortex, hypothalamus, and substantia nigra, whereas the amygdala and cerebellum appeared to be less affected. these findings suggest that the alteration of ace expression may be involved in different neurological symptoms (seizures, stroke, encephalitis, dizziness, headache, confusion, alteration of body temperature, anosmia, and ataxia) observed in covid- patients, in relation to the brain-affected area. interestingly, none of ace eqtl variants located in and targeting ace were reported in the lung tissue, which, instead, appeared particularly enriched in tmprss-associated eqtl variants. these findings raise the need for further investigation on the role of ace genetic variability in the susceptibility and clinical outcome of patients with covid- , especially concerning neurological symptoms. indeed, these studies will be useful for identifying patients at a higher risk of neurological complications, which may need different monitoring and treatment strategies in order to prevent fatal outcomes or severe, permanent brain injury. supplementary materials: the following supplementary files are available online at http://www.mdpi.com/ - / / / /s , table s : selection of the variants within the coding and splice site regions of ace with frequency data, annotated in genomes. the reported variants were retrieved from ensembl grch assembly. chr: chromosome; bp: base pairs; maf: minor allele frequency; aa: amino acid; table s : list of the significant eqtl variants distributed in relation to brain tissues. author contributions: c.s., v.c., c.c., g.n., e.g., and r.c. conceived the study; c.s., v.c., a.t., and s.b. collected the data and performed the analyses. c.s., v.c., s.g., p.b., c.c., g.n., e.g., and r.c. contributed to the interpretation of the results and supervised the study. c.s., v.c., and a.t. wrote the manuscript; s.g., p.b., c.c., g.n., e.g., and r.c. revised the manuscript. all authors have read and agreed to the published version of the manuscript. structure of the sars-cov- spike receptor-binding domain bound to the ace receptor ace in brain physiology and pathophysiology: evidence from transgenic animal models evidence of the covid- virus targeting the cns: tissue distribution, host-virus interaction, and proposed neurotropic mechanisms sars-cov- cell entry depends on ace and tmprss and is blocked by a clinically proven protease inhibitor tmprss contributes to virus spread and immunopathology in the airways of murine models after coronavirus infection tmprss activates the human coronavirus e for cathepsin-independent host cell entry and is expressed in viral target cells in the respiratory epithelium angiotensin-converting enzyme (ace ) as a sars-cov- receptor: molecular mechanisms and potential therapeutic target master regulator analysis of the sars-cov- /human interactome clinical features of patients infected with novel coronavirus in hlh across speciality collaboration, uk. covid- : consider cytokine storm syndromes and immunosuppression the neuroinvasive potential of sars-cov may play a role in the respiratory failure of covid- patients neurologic features in severe sars-cov- infection limb-girdle muscular dystrophies (lgmds): the clinical application of ngs analysis assessing individual risk for amd with genetic counseling, family history, and genetic testing genomes project consortium a global reference for human genetic variation the mutational constraint spectrum quantified from variation in , humans development of an italian rm y-str haplotype database: results of the gefi collaborative exercise vcfr: a package to manipulate and visualize variant call format data in r genomic variants filtering by deep learning models in ngs r: a language and environment for statistical computing. r foundation for statistical computing qvalue: q-value estimation for false discovery rate control varsite: disease variants and protein structure human splicing finder: an online bioinformatics tool to predict splicing signals uniprot: a worldwide hub of pro-tein knowledge human genomics. the genotype-tissue expression (gtex) pilot analysis: multitissue gene regulation in humans the biomart community portal: an innovative alternative to large, centralized data repositories comparative genetic analysis of the novel coronavirus ( -ncov/sars-cov- ) receptor ace in different populations nervous system involvement after infection with covid- and other coronaviruses does sars-cov- invade the brain? translational lessons from animal models this article is an open access article distributed under the terms and conditions of the creative commons attribution (cc by) license funding: this research received no external funding. the authors declare no conflict of interest. key: cord- -b f wtfn authors: caldarelli, guido; nicola, rocco de; petrocchi, marinella; pratelli, manuel; saracco, fabio title: analysis of online misinformation during the peak of the covid- pandemics in italy date: - - journal: nan doi: nan sha: doc_id: cord_uid: b f wtfn during the covid- pandemics, we also experience another dangerous pandemics based on misinformation. narratives disconnected from fact-checking on the origin and cure of the disease intertwined with pre-existing political fights. we collect a database on twitter posts and analyse the topology of the networks of retweeters (users broadcasting again the same elementary piece of information, or tweet) and validate its structure with methods of statistical physics of networks. furthermore, by using commonly available fact checking software, we assess the reputation of the pieces of news exchanged. by using a combination of theoretical and practical weapons, we are able to track down the flow of misinformation in a snapshot of the twitter ecosystem. thanks to the presence of verified users, we can also assign a polarization to the network nodes (users) and see the impact of low-quality information producers and spreaders in the twitter ecosystem. propaganda and disinformation have a history as long as mankind, and the phenomenon becomes particularly strong in difficult times, such as wars and natural disasters. the advent of the internet and social media has amplified and made faster the spread of biased and false news, and made targeting specific segments of the population possible [ ] . for this reason the vice-president of the european commission with responsibility for policies on values and transparency, vȇra yourová, announced, beginning of june , a european democracy action plan, expected by the end of , in which web platforms admins will be called for greater accountability and transparency, since 'everything cannot be allowed online' [ ] . manufacturers and spreaders of online disinformation have been particularly active also during the covid- pandemic period (e.g., writing about bill gates role in the pandemics or about masks killing children [ , ] ). this, alongside the real pandemics [ ] , has led to the emergence of a new virtual disease: covid- infodemics. in this paper, we shall consider the situation in italy, one of the most affected countries in europe, where the virus struck in a devastating way between the end of february and the end of april [ ] . in such a sad and uncertain time, propaganda [ ] in italy, since the beginning of the pandemics and at time of writing, almost k persons have contracted the covid- virus: of these, more than k have died. source: http://www.protezionecivile.gov.it/. accessed september , . has worked hard: one of the most followed fake news was published by sputnik italia receiving , likes, shares and comments on the most popular social media. 'the article falsely claimed that poland had not allowed a russian plane with humanitarian aid and a team of doctors headed to italy to fly over its airspace', the ec vice-president yourová said. actually, the studies regarding dis/mis/information diffusion on social media seldom analyse its effective impact. in the exchange of messages on online platforms, a great amount of interactions do not carry any relevant information for the understanding of the phenomenon: as an example, randomly retweeting viral posts does not contribute to insights on the sharing activity of the account. for determining dis/misinformation propagation two main weapons can be used, the analysis of the content (semantic approach) and the analysis of the communities sharing the same piece of information (topological approach). while the content of a message can be analysed on its own, the presence of some troublesome structure in the pattern of news producer and spreaders (i.e., in the topology of contacts) can be detected only trough dedicated instruments. indeed, for real in-depth analyses, the properties of the real system should be compared with a proper null model. recently, entropy-based null models have been successfully employed to filter out random noise from complex networks and focus the attention on non trivial contributions [ , ] . essentially, the method consists in defining a 'network benchmark' that has some of the (topological) properties of the real system, but is completely random for all the rest. then, every observation that does not agree with the model, i.e., cannot be explained by the topological properties of the benchmark, carries non trivial information. notably, being based on the shannon entropy, the benchmark is unbiased by definition. in the present paper, using entropy-based null-models, we analyse a tweet corpus related to the italian debate on covid- during the two months of maximum crisis in italy. after cleaning the system from the random noise, by using the entropy-based null-model as a filter, we have been able to highlight different communities. interestingly enough, these groups, beside including several official accounts of ministries, health institutions, and -online and offline -newspapers and newscasts, encompass four main political groups. while at first sight this may sound surprising -the pandemic debate was more on a scientific than on a political ground, at least in the very first phase of its abrupt diffusion -, it might be due to pre-existing echo chambers [ ] . the four political groups are found to perform completely different activities on the platform, to interact differently from each other, and to post and share reputable and non reputable sources of information with great differences in the number of their occurrences. in particular, the accounts from the right wing community interact, mainly in terms of retweets, with the same accounts who interact with the mainstream media. this is probably due to the strong visibility given by the mainstream media to the leaders of that community. moreover, the right wing community is more numerous and more active, even relatively to the number of accounts involved, than the other communities. interestingly enough, newly formed political parties, as the one of the former italian prime minister matteo renzi, quickly imposed their presence on twitter and on the online political debate, with a strong activity. furthermore, the different political parties use different sources for getting information on the spreading on the pandemics. to detect the impact of dis/misinformation in the debate, we consider the news sources shared among the accounts of the various groups. with a hybrid annotation approach, based on independent fact checking organisations and human annotation, we categorised such sources as reputable and non reputable (in terms of credibility of the published news and the transparency of the sources). notably, we experienced that a group of accounts spread information from non reputable sources with a frequency almost times higher than that of the other political groups. and we are afraid that, due to the extent of the online activity of the members of this community, the spreading of such a volume of non reputable news could deceit public opinion. we collected circa . m tweets in italian language, from february st to april th [ ] . details about the political situation in italy during the period of data collection can be found in the supplementary material, section . : 'evolution of the covid- pandemics in italy'. the data collection was keyword-based, with keywords related the covid- pandemics. twitter's streaming api returns any tweet containing the keyword(s) in the text of the tweet, as well as in its metadata. it is worth noting that it is not always necessary to have each permutation of a specific keyword in the tracking list. for example, the keyword 'covid' will return tweets that contain both 'covid ' and 'covid- '. table lists a subset of the considered keywords and hashtags. there are some hashtags that overlap due to the fact that an included keyword is a sub-string of another one, but we included both for completeness. the left panel of fig. shows the network obtained by following the projection procedure described in section . . the network resulting from the projection procedure will be called, in the rest of the paper, validated network. the term validated should not be confused with the term verified, which instead denotes a twitter user who has passed the formal authentication procedure by the social platform. in order to get the community of verified twitter users, we applied the louvain algorithm [ ] to the data in the validated network. such an algorithm, despite being one of the most popular, is also known to be order dependent [ ] . to get rid of this bias, we apply it iteratively n times (n being the number of the nodes) after reshuffling the order of the nodes. finally, we select the partition with the highest modularity. the network presents a strong community structure, composed by four main subgraphs. when analysing the emerging communities, we find that they correspond to right wing parties and media (in steel blue) center left wing (dark red) stars movement (m s ), in dark orange institutional accounts (in sky blue) details about the political situation in italy during the period of data collection can be found in the supplementary material, section . : 'italian political situation during the covid- pandemics'. this partition in four subgroups, once examined in more details, presents a richer substructure, described in the right panel of fig. . starting from the center-left wing, we can find a darker red community, including various ngos and various left oriented journalists, vips and pundits. a slightly lighter red sub-community turns out to be composed by the main politicians of the italian democratic party (pd), as well as by representatives from the european parliament (italian and others) and some eu commissioners. the violet red group is mostly composed by the representatives of italia viva, a new party founded by the former italian prime minister matteo renzi (december -february ). in golden red we can find the subcommunity of catholic and vatican groups. finally the dark violet red and light tomato subcommunities consist mainly of journalists. in turn, also the orange (m s) community shows a clear partition in substructures. in particular, the dark orange subcommunity contains the accounts of politicians, parliament representatives and ministers of the m s and journalists. in aquamarine, we can find the official accounts of some private and public, national and international, health institutes. finally, in the light slate blue subcommunity we can find various italian ministers as well as the italian police and army forces. similar considerations apply to the steel blue community. in steel blue, the subcommunity of center right and right wing parties (as forza italia, lega and fratelli d'italia). in the following, this subcommunity is going to be called as fi-l-fdi, recalling the initials of the political parties contributing to this group. the sky blue subcommunity includes the national federations of various sports, the official accounts of athletes and sport players (mostly soccer) and their teams. the teal subcommunity contains the main italian news agencies. in this subcommunity there are also the accounts of many universities. the firebrick subcommunity contains accounts related to the as roma football club; analogously in dark red official accounts of ac milan and its players. the slate blue subcommunity is mainly composed by the official accounts of radio and tv programs of mediaset, the main private italian broadcasting company. finally, the sky blue community is mainly composed by italian embassies around the world. for the sake of completeness, a more detailed description of the composition of the subcommunities in the right panel of figure is reported in the supplementary material, section . : 'composition of the subcommunities in the validated network of verified twitter users'. here, we report a series of analyses related to the domain names, hereafter simply called domains, that mostly appear in all the tweets of the validated network of verified users. the domains have been tagged according to their degree of credibility and transparency, as indicated by the independent software toolkit newsguard https://www.newsguardtech.com/. the details of this procedure are reported below. as a first step, we considered the network of verified accounts, whose communities and sub-communities are shown in fig. . on this topology, we labelled all domains that had been shared at least times (between tweets and retweets). table shows the tags associated to the domains. in the rest of the paper, we shall be interested in quantifying reliability of news sources publishing during the period of interest. thus, for our analysis, we will not consider those sources corresponding to social networks, marketplaces, search engines, institutional sites, etc. tags r, ∼ r and nr in table are used only for news sites, be them newspapers, magazines, tv or radio social channels, and they stand for reputable, quasi reputable, not reputable, respectively. label unc is assigned to those domains with less than occurrences in ours tweets and rewteets dataset. in fact, the labeling procedure is a hybrid one. as mentioned above, we relied on newsguard, a plugin resulting from the joint effort of journalists and software table tags used for labeling the domains developers aiming at evaluating news sites according to nine criteria concerning credibility and transparency. for evaluating the credibility level, the metrics consider whether the news source regularly publishes false news, does not distinguish between facts and opinions, does not correct a wrongly reported news. for transparency, instead, the tool takes into account whether owners, founders or authors of the news source are publicly known; and whether advertisements are easily recognizable [ ] . after combining the individual scores obtained out of the nine criteria, the plugin associates to a news source a score from to , where is the minimum score for the source to be considered reliable. when reporting the results, the plugin provides details about the criteria which passed the test and those that did not. in order to have a sort of no-man's land and not to be too abrupt in the transition between reputability and non-reputability, when the score was between and , we considered the source to be quasi reputable, ∼r. it is worth noting that not all the domains in the dataset under investigation were evaluated by newsguard at the time of our analysis. for those not evaluated automatically, the annotation was made by three tech-savvy researchers, who assessed the domains by using the same criteria as newsguard. table gives statistics about number and kind of tweets (tw = pure tweet; rt = retweet), the number of url and distinct url (dist url), the number of domains and users in the validated network of verified users. we clarify what we mean by these terms with an example: a domain for us corresponds to the so-called 'second-level domain' name [ ] , i.e., the name directly to the left of .com, .net, and any other top-level domains. for instance, repubblica.it, corriere.it, nytimes.com are considered domains by us. instead, the url maintains here its standard definition [ ] and an example is http://www.example.com/index.html. table shows the outcome of the domains annotation, according to the scores of newsguard or to those assigned by the three annotators, when scores were no available from newsguard. at a first glance, the majority of the news domains belong to the reputable category. the second highest percentage is the one of the untagged domains -unc. in fact, in our dataset there are many domains that occur only few times once. for example, there are domains that appear in the datasets only once. fig. shows the trend of the number of tweets and retweets, containing urls, posted by the verified users of the validated projection during the period of data [ ] newsguard rating process: https://www.newsguardtech.com/ratings/rating-process-criteria/ [ ] https://en.wikipedia.org/wiki/domain_name [ ] table annotation results over all the domains in the whole dataset -validated network of verified users. in [ ] . going on with the analysis, table shows the percentage of the different types of domains for the communities identified in the left plot of fig. . it is worth observing that the steel blue community (both politicians and media) is the most active one, even if it is not the most represented: the number of users is lower than the one of the center left community (the biggest one, in terms of numbers), but the number of their posts containing a valid url is almost the double of that of the second more active community. interestingly, the activity of the verified users of the steel blue community is more focused on content production of (see the only tweets sub-table) than in sharing (see the only retweets sub-table). in fact, retweets represent almost . % of all posts from the media and the right wing community, while in the case of the center-left community it is . %. this effect is observable even in the average only tweets post per verified user: a right-wing user and a media user have an average of . original posts, against . for center-left-wing users. these numbers are probably due to the presence in the former community of the italian most accessed media. they tend to spread their (original) pieces of news on the twitter platform. interestingly, the presence of urls from a non reputable source in the steel blue community is more than times higher than the second score in the same field in the case of original tweets (only tweets). it is worth noting that, for the case of the dark orange and sky blue communities, which are smaller both in terms of users and number of posts, the presence of non classified sources is quite strong (it represents nearly % of retweeted posts for both the communities), as it is the frequency of posts linking to social network contents. interestingly enough, the verified users of both groups seem to focus slightly more on the same domains: there are, on average, . and . posts for each url domain respectively for the dark orange and sky blue communities, and, on average, . and . posts for the steel blue and the dark red communities. the right plot in fig. report a fine grained division of communities: the four largest communities have been further divided into sub-communities, as mentioned in subsection . . here, we focus on the urls shared in the purely political sub-communities in table . broadly speaking, we examine the contribution of the different political parties, as represented on twitter, to the spread of mis/disinformation and propaganda. table clearly shows how the vast majority of the news coming from sources considered scarce or non reputable are tweeted and retweeted by the steel blue political sub-community (fi-l-fdi). notably, the percentage of non reputable sources shared by the fi-l-fdi accounts is more than times the percentage of their community (the steel blue one) and it is more than times the second community in the nr ratio ranking. for all the political sub-communities the incidence of social network links is much higher than in their original communities. looking at table , even if the number of users in each political sub-community is much smaller, some peculiar behaviours can be still be observed. again, the center-right and right wing parties, while representing the least represented ones in terms of users, are much more active than the other groups: each (verified) user is responsible, on average of almost . messages, while the average is . , . and . for m s, iv and pd, respectively. it is worth noticing that italia viva, while being a recently founded party, is very active; moreover, for them the frequency of quasi reputable sources is quite high, especially in the case of only tweets posts. the impact of uncategorized sources is almost constant for all communities in the retweeting activity, while it is particularly strong for the m s. finally, the posts by the center left communities (i.e., italia viva and the democratic party) tend to have more than one url. specifically, every post containing at least a url, has, on average, . and . urls respectively, against the . of movimento stelle and . for the center-right and right wing parties. to conclude the analysis on the validated network of verified users, we report statistics about the most diffused hashtags in the political sub-communities. fig. focuses on wordclouds, while fig. reports the data under an histograms form. actually, from the various hashtags we can derive important information regarding the communications of the various political discursive communities and their position towards the management of the pandemics. first, it has to be noticed that the m s is the greatest user of hashtags: their two most used hashtags have been used almost twice the most used hashtags used by the pd, for instance. this heavy usage is probably due to the presence in this community of journalists and of the official account of il fatto quotidiano, a newspaper explicitly supporting the m s: indeed, the first two hashtags are "#ilfattoquotidiano" and "#edicola" (kiosk, in italian). it is interesting to see the relative importance of hashtags intended to encourage the population during the lockdown: it is the case of "#celafaremo" (we will make it), "#iorestoacasa" (i am staying home), "#fermiamoloinsieme" (let's stop it together ): "#iorestoacasa" is present in every community, but it ranks th in the m s verified user community, th in the fi-l-fdi community, nd in the italia viva community and th in the pd one. remarkably, "#celafaremo" is present only in the m s group, as "#fermiamoloinsieme" can be found in the top hashtags only in the center-right and right wing cluster. the pd, being present in various european institutions, mentions more european related hashtags ("#europeicontrocovid ", europeans against covid- ), in order to ask for a common reaction of the eu. the center-right and right wing community has other hashtags as "#forzalombardia" (go, lombardy! ), ranking the nd, and "#fermiamoloinsieme", ranking th. what is, nevertheless, astonishing, is the presence among the most used hashtags of all communities of the name of politicians from the same group ('interestingly '#salvini" is the first used hashtag in the center right and right wing community, even if he did not perform any duty in the government), tv programs ("#mattino ", "#lavitaindiretta", "#ctcf", "#dimartedì"), as if the main usage of hashtags is to promote the appearance of politicians in tv programs. finally, the hashtags used by fi-l-fdi are mainly used to criticise the actions of the government, e.g., "#contedimettiti" (conte, resign! ). fig. shows the structure of the directed validated projection of the retweet activity network, as outcome of the procedure recalled in section of the supplementary material. as mentioned in section of the supplementary material, the affiliation of unverified users has been determined using the tags obtained by the validated projected network of the verified users, as immutable label for the label propagation of [ ] . after label propagation, the representation of the political communities in the validated retweet network changes dramatically with respect to the case of the network of verified users: the center-right and right wing community is the most represented community in the whole network, with users (representing . % of all the users in the validated network), followed by italia viva users with accounts ( . % of all the accounts in the validated network). the impact of m s and pd is much more limited, with, respectively, and accounts. it is worth noting that this result is unexpected, due to the recent formation of italia viva. as in our previous study targeting the online propaganda [ ] , we observe that the most effective users in term of hub score [ ] are almost exclusively from the center-right and right wing party: considering the first hubs, only are not from this group. interestingly, out of these are verified users: roberto burioni, one of the most famous italian virologists, ranking nd, agenzia ansa, a popular italian news agency, ranking st, and tgcom , the popular newscast of a private tv channel, ranking rd. the fourth account is an online news website, ranking th: this is a not verified account which belongs to a not political community. remarkably, in the top hubs we find of the top hubs already found when considered the online debate on migrations from northern africa to italy [ ] : in particular, a journalist of a neo-fascist online newspaper (non verified user), an extreme right activist (non verified user) and the leader of fratelli d'italia giorgia meloni (verified user), who ranks rd in the hub score. matteo salvini (verified user), who was the first hub in [ ] , ranks th, surpassed by his party partner claudio borghi, ranking th. the first hub in the present network is an extreme right activist, posting videos against african migrants to italy and accusing them to be responsible of the contagion and of violating lockdown measures. table shows the annotation results of all the domains tweeted and retweeted by users in the directed validated network. the numbers are much higher than those shown in table , but the trend confirms the previous results. the majority of urls traceable to news sources are considered reputable. the number of unclassified domains is higher too. in fact, in this case, the annotation was made considering the domains occurring at least times. table annotation results over all the domains -directed validated network table reports statistics about posts, urls, distinct urls, users and verified users in the directed validated network. noticeably, by comparing these numbers with those of table , reporting statistics about the validated network of verified users, we can see that here the number of retweets is much more higher, and the trend is the opposite: verified users tend to tweet more than retweet ( vs ), while users in the directed validated network, which comprehends also non verified users, have a number of retweets . times higher than the number of their tweets. fig. shows the trend of the number of tweets containing urls over the period of data collection. since we are analysing a bigger network than the one considered in section . , we have numbers that are one order of magnitude greater than those shown in fig. ; the highest peak, after the discovery of the first cases in lombardy, corresponds to more than , posts containing urls, whereas the analogous peak in fig. corresponds to , posts. apart from the order of magnitudes, the two plots feature similar trends: higher traffic before the beginning of the italian lockdown, and a settling down as the quarantine went on [ ] . table shows the core of our analysis, that is, the distribution of reputable and non reputable news sources in the direct validated network, consisting of both verified and non-verified users. again, we focus directly on the political sub-communities identified in the previous subsection. two of the sub-communities are part of the center-left wing community, one is associated to the stars movement, the remaining one represents center-right and right wing communities. in line with previous results on the validated network of verified users, the table clearly shows how the vast majority of the news coming from sources considered scarce or non reputable are tweeted and retweeted by the center-right and right wing communities; % of the domains tagged as nr are shared by them. as shown in table , the activity of fi-l-fdi users is again extremely high: on average there are . retweets per account in this community, against the . of m s, the . of iv and the . of pd. the right wing contribution to the debate is extremely high, even in absolute numbers, due to the the large number of users in this community. it is worth mentioning that the frequency of non reputable sources in this community is really high (at about % of the urls in the only tweets) and comparable with that of the reputable ones (see table , only [ ] the low peaks for february and march are due to an interruption in the data collection, caused by a connection breakdown. table domains annotation per political sub-communities -directed validated network tweets). in the other sub-communities, pd users are more focused on un-categorised sources, while users from both italia viva and movimento stelle are mostly tweeting and retweeting reputable news sources. and users, but also in absolute numbers: out of the over m tweets, more than k tweets refer to a nr url. actually, the political competition still shines through the hashtag usage even for the other communities: it is the case, for instance, of italia viva. in the top hashtags we can find '#salvini', '#lega', but also '#papeete' [ ] , '#salvinisciacallo' (salvini jackal ) and '#salvinimmmerda' (salvini asshole). on the other hand, in italia viva hashtags supporting the population during the lockdown are used: '#iorestoacasa', '#restoacasa' (i am staying home), '#restiamoacasa' (let's stay home). criticisms towards the management of lombardy health system during the pandemics can be deduced from the hashtag '#commissariamtelalombardia' (put lombardy under receivership) and '#fontana' (the lega administrator of the lombardy region). movimento stelle has the name of the main leader of the opposition '#salvini', as first hashtag and supports criticisms to the lombardy administration with the hashtags '#fontanadimettiti' (fontana, resign! ) and '#gallera', the health and welfare minister of the lombardy region, considered the main responsible for the bad management of the pandemics. nevertheless, it is possible to highlight even some hashtags encouraging the population during the lock down, as the above mentioned '#iorestoacasa', '#restoacasa' and '#restiamoacasa'. it is worth mentioning that the government measures, and the corresponding m s campaigns, are accompanied specific hashtags: '#curaitalia' is the name of one of the decree of the prime minister to inject liquidity in the italian economy, '#acquistaitaliano' (buy italian products! ), instead, advertise italian products to support the national economy. as a final task, over the whole set of tweets produced or shared by the users in the directed validated network, we counted the number of times a message containing a url was shared by users belonging to different political communities, although without considering the semantics of the tweets. namely, we ignored whether the urls were shared to support or to oppose the presented arguments. table shows the most tweeted (and retweeted) nr domains shared by the political communities presented in table , the number of occurrences is reported next to each domain. the first nr domains for fi-l-fdi in table are related to the right, extreme right and neo-fascist propaganda, as it is the case of imolaoggi.it, ilprimatonazionale.it and voxnews.info, recognised as disinformation websites by newsguard and by the two main italian debunker websites, bufale.net and butac.it. as shown in the table, some domains, although in different number of occurrences, are present under more than one column, thus shared by users close to different political communities. this could mean, for some subgroups of the community, a retweet with the aim of supporting the opinions expressed in the original tweets. however, since the semantics of the posts in which these domains are present were not investigated, the retweets of the links by more than one political community could be due to contrast, and not to support, the opinions present in the original posts. despite the fact that the results were achieved for a specific country, we believe that the applied methodology is of general interest, being able to show trends and peculiarities whenever information is exchanged on social networks. in particular, when analysing the outcome of our investigation, some features attracted our attention: persistence of clusters wrt different discussion topics: in caldarelli et al. [ ] , we focused on tweets concerned with immigration, an issue that has been central in the italian political debate for years. here, we discovered that the clusters and the echo chambers that have been detected when analysing tweets about immigration are almost the same as those singled out when considering discussions concerned with covid- . this may seem surprising, because a discussion about covid- may not be exclusively political, but also medical, social, economic, etc.. from this we can argue that the clusters are political in nature and, even when the topic of discussion changes, users remain in their cluster on twitter. (indeed, journalists and politicians use twitter for information and political propaganda, respectively). the reasons political polarisation and political vision of the world affect so strongly also the analysis of what should be an objective phenomenon is still an intriguing question. persistence of online behavioral characteristics of clusters: we found that the most active, lively and penetrating online communities in the online debate on covid- are the same found in [ ] , formed in a almost purely political debate such as the one represented by the right of migrants to land on the italian territory. (dis)similarities amongst offline and online behaviours of members and voters of parties: maybe less surprisingly, the political habits is also reflected in the degree of participation to the online discussions. in particular, among the parties in the centre-left-wing side, a small party (italia viva) shows a much more effective social presence than the larger party of the italian centre-left-wing (partito democratico), which has many more active members and more parliamentary representation. more generally, there is a significant difference in social presence among the different political parties, and the amount of activity is not at all proportional to the size of the parties in terms of members and voters. spread of non reputable news sources: in the online debate about covid- , many links to non reputable (defined such by newsguard, a toolkit ranking news website based on criteria of transparency and credibility, led by veteran journalists and news entrepreneurs) news sources are posted and shared. kind and occurrences of the urls vary with respect to the corresponding political community. furthermore, some of the communities are characterised by a small number of verified users that corresponds to a very large number of acolytes which are (on their turn) very active, three times as much as the acolytes of the opposite communities in the partition. in particular, when considering the amount of retweets from poorly reputable news sites, one of the communities is by far (one order of magnitude) much more active than the others. as noted already in our previous publication [ ] , this extra activity could be explained by a more skilled use of the systems of propaganda -in that case a massive use of bot accounts and a targeted activity against migrants (as resulted from the analysis of the hub list). our work could help in steering the online political discussion around covid- towards an investigation on reputable information, while providing a clear indication of the political inclination of those participating in the debates. more generally, we hope that our work will contribute to finding appropriate strategies to fight online misinformation. while not completely unexpected, it is striking to see how political polarisation affects also the covid- debate, giving rise to on-line communities of users that, for number and structure, almost closely correspond to their political affiliations. this section recaps the methodology through which we have obtained the communities of verified users (see section . ). this methodology has been designed in saracco et al. [ ] and applied in the field of social networks for the first time in [ , ] . for the sake of completeness, the supplementary material, section , recaps the methodology through which we have obtained the validated retweet activity network shown in section . . in section of the supplementary material, the detection of the affiliation of unverified users is described. in the supplementary material, the interested reader will also find additional details about ) the definition of the null models (section ); ) a comparison among various label propagation for the political affiliation of unverified users (section ); and ) a brief state of the art on fact checking organizations and literature on false news detection (section ). many results in the analysis of online social networks (osn) shows that users are highly clustered in group of opinions [ , - , , , ] ; indeed those groups have some peculiar behaviours, as the echo chamber effects [ , ] . following the example of references [ , ] , we are making use of this users' clustering in order to detect discursive community, i.e. groups of users interacting among themselves by retweeting on the same (covid-related) subjects. remarkably, our procedure does not follow the analysis of the text shared by the various users, but is simply related on the retweeting activity among users. in the present subsection we will examine how the discursive community of verified twitter users can be extracted. on twitter there are two distinct categories of accounts: verified and unverified users. verified users have a thick close to the screen name: the platform itself, upon request from the user, has a procedure to check the authenticity of the account. verified accounts are owned by politicians, journalists or vips in general, as well as the official accounts of ministers, newspapers, newscasts, companies and so on; for those kind of users, the verification procedure guarantees the identity of their account and reduce the risk of malicious accounts tweeting in their name. non verified accounts are for standard users: in this second case, we cannot trust any information provided by the users. the information carried by verified users has been studied extensively in order to have a sort of anchor for the related discussion [ , , , , ] to detect the political orientation we consider the bipartite network represented by verified (on one layer) and unverified (on the other layer) accounts: a link is connecting the verified user v with the unverified one u if at least one time v was retweeted by u, or viceversa. to extract the similarity of users, we compare the commonalities with a bipartite entropy-based null-model, the bipartite configuration model (bicm [ ] ). the rationale is that two verified users that share many links to same unverified accounts probably have similar visions, as perceived by the audience of unverified accounts. we then apply the method of [ ] , graphically depicted in fig. , in order to get a statistically validated projection of the bipartite network of verified and unverified users. in a nutshell, the idea is to compare the amount of common linkage measured on the real network with the expectations of an entropy-based null model fixing (on average) the degree sequence: if the associated p-value is so low that the overlaps cannot be explained by the model, i.e. such that it is not compatible with the degree sequence expectations, they carry non trivial information and we project the related information on the (monopartite) projection of verified users. the interested reader can find the technical details about this validated projection in [ ] and in the supplementary information. the data that support the findings of this study are available from twitter, but restrictions apply to the availability of these data, which were used under license italian socio-political situation during the period of data collection in the present subsection we present some crucial facts for the understanding of the social context in which our analysis is set. this subsection is divided into two parts: the contagion evolution and the political situation. these two aspects are closely related. a first covid- outbreak was detected in codogno, lodi, lombardy region, on february, th [ ] . in the very next day, two cases were detected in vò, padua, veneto region. on february, th, in order to contain the contagions, the national government decided to put in quarantine municipalities, in the area around lodi and vò, near padua [ ] . nevertheless, the number of contagions raised to , hitting different regions; one of the infected person in vò died, representing the first registered italian covid- victim [ ] . on february, th there were already confirmed cases in italy. the first lockdown should have lasted until the th of march, but due to the still increasing number of contagions in northern italy, the italian prime minister giuseppe conte intended to extend the quarantine zone to almost all the northern italy on sunday, march th [ ] : travel to and from the quarantine zone were limited to case of extreme urgency. a draft of the decree announcing the expansion of the quarantine area appeared on the website of the italian newspaper corriere della sera on the late evening of saturday, th, causing some panic in the interested areas [ ] : around people, living in milan, but coming from southern regions, took trains and planes to reach their place of [ ] prima lodi, ""paziente ", il merito della diagnosi va diviso... per due", th june [ ] italian gazzetta ufficiale, "decreto-legge febbraio , n. ". the date is intended to be the very first day of validity of the decree. [ ] il fatto quotidiano, "coronavirus,è morto il enne ricoverato nel padovano. contagiati in lombardia, un altro in veneto", nd february . [ ] bbc news, "coronavirus: northern italy quarantines million people", th march " [ ] the guardian, "leaked coronavirus plan to quarantine m sparks chaos in italy", th march origins [ ] [ ] . in any case, the new quarantine zone covered the entire lombardy and partially other regions. remarkably, close to bergamo, lombardy region, a new outbreak was discovered and the possibility of defining a new quarantine area on march th was considered: this opportunity was later abandoned, due to the new northern italy quarantine zone of the following days. this delay seems to have caused a strong increase in the number of contagions, making the bergamo area the most affected one, in percentage, of the entire country [ ] ; at time of writing, there are investigations regarding the responsibility of this choice. on march, th, the lockdown was extended to the whole country, resulting in the first country in the world to decide for national quarantine [ ] . travels were restricted to emergency reason or to work; all business activities that were not considered as essentials, as pharmacies and supermarkets, had to be closed. until the st of march lockdown measures became progressively stricter all over the country. starting from the th of april, some retails activities as children clothing shops, reopened. a first fall in the number of deaths was observed on the th of april [ ] . a limited reopening started with the so-called "fase " (phase ) on the th of may [ ] . from the very first days of march, the limited capacity of the intensive care departments to take care of covid-infected patients, took to the necessity of a re-organization of italian hospitals, leading, e.g., to the opening of new intensive care departments [ ] . moreover, new communication forms with the relatives of the patients were proposed, new criteria for the intubating patients were developed, and, in the extreme crisis, in the most infected cases, the emergency management took to give priority to the hospitalisation to patients with a higher probability to recover [ ] . outbreaks were mainly present in hospitals [ ] . unfortunately, healthcare workers were contaminated by the covid [ ] . this contagion resulted in a relative high number of fatalities: by the nd of april, covid deaths were registered among doctors. due to the pressure on the intensive care capacity, even the healthcare personnel was subject to extreme stress, especially in the most affected zones [ ] . on august th, , the leader of lega, the main italian right wing party, announced to negate the support to the government of giuseppe conte, which was formed after a post-election coalition between the renzi formed a new center-left party, italia viva (italy alive, iv), due to some discord with pd; despite the scission, italia viva continued to support the actual government, having some of its representatives among the ministers and undersecretaries, but often marking its distance respect to both pd and m s. due to the great impact that matteo salvini and giorgia meloni -leader of fratelli d'italia, a right wing party-have on social media, they started a massive campaign against the government the day after its inauguration. the regions of lombardy, veneto, piedmont and emilia-romagna experienced the highest number of contagions during the pandemics; among those, the former are administrated by the right and center-right wing parties, the fourth one by the pd. the disagreement in the management of the pandemics between regions and the central government was the occasion to exacerbate the political debate (in italy, regions have a quite wide autonomy for healthcare). the regions administrated by the right wing parties criticised the centrality of the decisions regarding the lock down, while the national government criticises the health management (in lombardy the healthcare system has a peculiar organisation, in which the private sector is supported by public funding) and its non effective measure to reduce the number of contagions. the debate was ridden even at a national level: the opposition criticized the financial origin of the support to the various economic sectors. moreover, the role of the european union in providing funding to recover italian economics after the pandemics was debated. here, we detail the composition of the communities shown in figure of the main text. we remind the reader that, after applying the leuven algorithm to the validated network of verified twitter users, we could observe main communities, that correspond to right wing parties and media (in steel blue) center left wing (dark red) stars movement (m s ), in dark orange institutional accounts (in sky blue) starting from the center-left wing, we can find a darker red community, including various ngos (the italian chapters of unicef, medecins sans frontieres, action aid, emergency, save the children, etc.), various left oriented journalists, vips and pundits [ ] . finally, we can find in this group political movements (' sardine') and politicians on the left of pd (as beppe civati, pietro grasso, ignazio marino) or on the left current of the pd (laura boldrini, michele emiliano, stefano bonaccini). a slightly lighter red sub-community turns out to be composed by the main politicians of the italian democratic party (pd), as well as by representatives from the european parliament (italian and others) and some eu commissioners. the violet red group is mostly composed by the representatives of the newly founded italia viva, by the former italian prime minister matteo renzi (december -february ) and former secretary of pd. in golden red we can find the subcommunity of catholic and vatican groups. finally the dark violet red and light tomato subcommunities are composed mainly by journalists. interestingly enough, the dark violet red contains also accounts related to the city of milan (the major, the municipality, the public services account) and to the spoke person of the chinese minister of foreign affair. in turn, also the orange (m s) community shows a clear partition in substructures. in particular, the dark orange subcommunity contains the accounts of politicians, parliament representatives and ministers of the m s and journalists and the official account of il fatto quotidiano, a newspaper supporting the movement stars. interestingly, since one of the main leaders of the movement, luigi di maio, is also the italian minister of foreign affairs, we can find in this subcommunity also the accounts of several italian embassies around the world, as well as the account of the italian representatives at nato, ocse and oas. in aquamarine, we can find the official accounts of some private and public, national and international, health institutes (as the italian istituto superiore di sanità, literally the italian national institute of health, the world health organization, the fondazione veronesi) the minister of health roberto speranza, and some foreign embassies in italy. finally, in the light slate blue subcommunity we can find various italian ministers as well as the italian police and army forces. similar considerations apply to the steel blue community. in steel blue, the subcommunity of center right and right wing parties (as forza italia, lega and fratelli d'italia). the presidents of the regions of lombardy, veneto and liguria, administrated by center right and right wing parties, can be found here. (in the following this subcommunity is going to be called as fi-l-fdi, recalling the initials of the political parties contributing to this group.) the sky blue subcommunity includes the national federations of various sports, the official accounts of athletes and sport players (mostly soccer) and their teams, as well as sport journals, newscasts and journalists. the teal subcommunity contains the main italian news agencies, some of the main national and local newspapers, [ ] as the cartoonists makkox and vauro, the singers marracash, frankiehinrg, ligabue and emphil volo vocal band, and journalists from repubblica (ezio mauro, carlo verdelli, massimo giannini), from la tv channel (ricardo formigli, diego bianchi). newscasts and their journalists. in this subcommunity there are also the accounts of many universities; interestingly enough, it includes also the all the local public service local newscasts. the firebrick subcommunity contains accounts related to the as roma football club; analogously in dark red official accounts of ac milan and its players. the slate blue subcommunity is mainly composed by the official accounts of radio and tv programs of mediaset, the main private italian broadcasting company, together with singers and musicians. other smaller subcommunities includes other sport federations, and sports pundits. finally, the sky blue community is mainly composed by italian embassies around the world. the navy subpartition contains also the official accounts of the president of the republic, the italian minister of defense and the one of the commissioner for economy at eu and former prime minister, paolo gentiloni. in the study of every phenomenon, it is of utmost importance to distinguish the relevant information from the noise. here, we remind a framework to obtain a validated monopartite retweet network of users: the validation accounts the information carried by not only the activity of the users, but also by the virality of their messages. we represented pictorially the method in fig. . we define a directed bipartite network in which one layer is composed by accounts and the other one by the tweets. an arrow connecting a user u to a tweet t represents the u writing the message t. the arrow in the opposite direction means that the user u is retweeting the message t. to filter out the random noise from this network, we make use of the directed version of the bicm, i.e. the bipartite directed configuration model (bidcm [ ] ). the projection procedure is then, analogous to the one presented in the previous subsection: it is pictorially displayed in the fig. . briefly, consider the couple of users u and u and consider the number of message written by u and shared u . then, calculate which is the distribution of the same measure according with the bidcm: if the related p-value is statistically significant, i.e. if the number of u 's tweets shared by u is much more than expected by the bidcm, we project a (directed) link from u to u . summarising, the comparison of the observation on the real network with the bidcm permits to uncover all contributions that cannot originate from the constraints of the null-model. using the technique described in subsection . of the main text, we are able to assign to almost all verified users a community, based on the perception of the unverified users. due to the fact that the identity of verified users are checked by twitter, we have the possibility of controlling our groups. indeed, as we will show in the following, the network obtained via the bipartite projection provides a reliable description regarding the closeness of opinions and role in the social debate. how can we use this information in order to infer the orientation of non verified users? in the reference [ ] we used the tags obtained for both verified and unverified users in the bipartite network described in subsection . of the main real network c) e) figure schematic representation of the projection procedure for bipartite directed network. a) an example of a real directed bipartite network. for the actual application, the two layers represent twitter accounts (turquoise) and posts (gray). a link from a turquoise node to a gray one represents that the post has been written by the user; a link in the opposite direction represents a retweet by the considered account. b) the bipartite directed configuration model (bidcm) ensemble is defined. the ensemble includes all the link realisations, once the number of nodes per layer has been fixed. c) we focus our attention on nodes i and j and count the number of directed common neighbours (in magenta both the nodes and the links to their common neighbours), i.e., the number of posts written by i and retweeted by j. subsequently, d) we compare this measure on the real network with the one on the ensemble: if this overlap is statistically significant with respect to the bidcm, e) we have a link from i to j in the projected network. text and propagated those labels accross the network. in a recent analysis, we observed that other approaches are more stable [ ] : in the present manuscript we make use of the most stable algorithm. we use the label propagation as proposed in [ ] on the directed validated network. indeed, the validated directed network in the present appendix we remind the main steps for the definition of an entropy based null model; the interested reader can refer to the review [ ] . we start by revising the bipartite configuration model [ ] , that has been used for detecting the network of similarities of verified users. we are then going to examine the extension of this model to bipartite directed networks [ ] . finally, we present the general methodology to project the information contained in a -directed or undirected-bipartite network, as developed in [ ] . let us consider a bipartite network g * bi , in which the two layers are l and Γ. define g bi the ensemble of all possible graphs with the same number of nodes per layer as in g * bi . it is possible to define the entropy related to the ensemble as [ ] : where p (g bi ) is the probability associated to the instance g bi . now we want to obtain the maximum entropy configuration, constraining some relevant topological information regarding the system. for the bipartite representation of verified and unverified user, a crucial ingredient is the degree sequence, since it is a proxy of the number of interactions (i.e. tweets and retweets) with the other class of accounts. thus in the present manuscript we focus on the degree sequence. let us then maximise the entropy ( ), constraining the average over the ensemble of the degree sequence. it can be shown, [ ] , that the probability distribution over the ensemble is where m iα represent the entries of the biadjacency matrix describing the bipartite network under consideration and p iα is the probability of observing a link between the nodes i ∈ l and α ∈ Γ. the probability p iα can be expressed in terms of the lagrangian multipliers x and y for nodes on l and Γ layers, respectively, as in order to obtain the values of x and y that maximize the likelihood to observe the real network, we need to impose the following conditions [ , ]        where the * indicates quantities measured on the real network. actually, the real network is sparse: the bipartite network of verified and unverified users has a connectance ρ . × − . in this case the formula ( ) can be safely approximated with the chung-lu configuration model, i.e. where m is the total number of links in the bipartite network. in the present subsection we will consider the case of the extension of the bicm to direct bipartite networks and highlight the peculiarities of the network under analysis in this representation. the adjancency matrix describing a direct bipartite network of layers l and Γ has a peculiar block structure, once nodes are order by layer membership (here the nodes on l layer first): where the o blocks represent null matrices (indeed they describe links connecting nodes inside the same layer: by construction they are exactly zero) and m and n are non zero blocks, describing links connecting nodes on layer l with those on layer Γ and viceversa. in general m = n, otherwise the network is not distinguishable from an undirected one. we can perform the same machinery of the section above, but for the extension of the degree sequence to a directed degree sequence, i.e. considering the in-and out-degrees for nodes on the layer l, (here m iα and n iα represent respectively the entry of matrices m and n) and for nodes on the layer Γ, the definition of the bipartite directed configuration model (bidcm, [ ] ), i.e. the extension of the bicm above, follows closely the same steps described in the previous subsection. interestingly enough, the probabilities relative to the presence of links from l to Γ are independent on the probabilities relative to the presence of links from Γ to l. if q iα is the probability of observing a link from node i to node α and q iα the probability of observing a link in the opposite direction, we have where x out i and x in i are the lagrangian multipliers relative to the node i ∈ l, respectively for the out-and the in-degrees, and y out α and y in α are the analogous for α ∈ Γ. in the present application we have some simplifications: the bipartite directed network representation describes users (on one layer) writing and retweeting posts (on the other layer). if users are on the layer l and posts on the opposite layer and m iα represents the user i writing the post α, then k in α = ∀α ∈ Γ, since each message cannot have more than an author. notice that, since our constraints are conserved on average, we are considering, in the ensemble of all possible realisations even instances in which k in α > or k in α = , or, otherwise stated, non physical; nevertheless the average is constrained to the right value, i.e. . the fact that k in α is the same for every α allows for a great simplification of the probability per link on m: where n Γ is the total number of nodes on the Γ layer. the simplification in ( ) is extremely helpful in the projected validation of the bipartite directed network [ ] . the information contained in a bipartite -directed or undirected-network, can be projected onto one of the two layers. the rationale is to obtain a monopartite network encoding the non trivial interactions among the two layers of the original bipartite network. the method is pretty general, once we have a null model in which probabilities per link are independent, as it is the case of both bicm and bidcm [ ] . the first step is represented by the definition of a bipartite motif that may capture the non trivial similarity (in the case of an undirected bipartite network) or flux of information (in the case of a directed bipartite network). this quantity can be captured by the number of v −motifs between users i and j [ , ] , or by its direct extension (note that v ij = v ji ). we compare the abundance of these motifs with the null models defined above: all motifs that cannot be explained by the null model, i.e. whose p-value are statistically significance, are validated into the projection on one of the layers [ ] . in order to assess the statistically significance of the observed motifs, we calculate the distribution associated to the various motifs. for instance, the expected value for the number of v-motifs connecting i and j in an undirected bipartite network is where p iα s are the probability of the bicm. analogously, where in the last step we use the simplification of ( ) [ ] . in both the direct and the undirect case, the distribution of the v-motifs or of the directed extensions is poisson binomial one, i.e. a binomial distribution in which each event shows a different probability. in the present case, due to the sparsity of the analysed networks, we can safely approximate the poisson-binomial distribution with a poisson one [ ] . in order to state the statistical significance of the observed value, we calculate the related p-values according to the relative null-models. once we have a p-value for every detected v-motif, the related statistical significance can be established through the false discovery rate (fdr) procedure [ ] . respect to other multiple test hypothesis, fdr controls the number of false positives. in our case, all rejected hypotheses identify the amount of v-motifs that cannot be explained only by the ingredients of the null model and thus carry non trivial information regarding the systems. in this sense, the validated projected network includes a link for every rejected hypothesis, connecting the nodes involved in the related motifs. in the main text, we solved the problem of assigning the orientation to all relevant users in the validated retweet network via a label propagation. the approach is similar, but different to the one proposed in [ ] , the differences being in the starting labels, in the label propagation algorithm and in the network used. in this section we will revise the method employed in the present article, as compared it to the one in [ ] and evaluate the deviations from other approaches. first step of our methodology is to extract the polarisation of verified users from the bipartite network, as described in section . of the main text, in order to use it as seed labels in the label propagation. in reference [ ] , a measure of the "adherence" of the unverified users towards the various communities of verified users was used in order to infer their orientation, following the approach in [ ] , in turn based on the polarisation index defined in [ ] . this approach was extremely performing when practically all unverified users interact at least once with verified one, as in [ ] . while still having good performances in a different dataset as the one studied in [ ] , we observed isolated deviations: it was the case of users with frequent interactions with other unverified accounts of the same (political) orientation, randomly retweeting a different discursive community verified user. in this case, focusing just on the interaction with verified accounts, those nodes were assigned a wrong orientation. the labels for the polarisation of the unverified users defined [ ] were subsequently used as seed labels in the label propagation. due to the possibility described above of assigning wrongly labels to unverified accounts, in the present paper, we consider only the tags of verified users, since they pass a strict validation procedure and are more stable. in order to compare the results obtained with the various approaches, we calculated the variation of information (vi, [ ] ). v i considers exactly the different in information contents captured by two different partition, as consider by the shannon entropy. results are reported in the matrix in figure for the th of february (results are similar for other days). even when using the weighted retweet network as "exact" result, the partition found by the label propagation of our approach has a little loss of information, comparable with the one of using an unweighted approach. indeed, the results found by the various community detection algorithms show little agreement with the label propagation ones. nevertheless, we still prefer the label propagation procedure, since the validated projection on the layer of verified users is theoretically sound and has a non trivial interpretation. the main result of this work quantifies the level of diffusion on twitter of news published by sources considered scarcely reputable. academy, governments, and news agencies are working hard to classify information sources according to criteria of credibility and transparency of published news. this is the case, for example, of newsguard, which we used for the tagging of the most frequent domains in the direct validated network obtained according to the methodology presented in the previous sections. as introduced in subsection . of the main text, the newsguard browser extension and mobile app [ ] offers a reliability result for the most popular newspapers in the world, summarizing with a numerical score the level of credibility and journalistic transparency of the newspaper. with the same philosophy, but oriented towards us politics, the fact-checking site politifact.com reports with a 'truth meter' the degree of truthfulness of original claims made by politicians, candidates, their staffs, and, more, in general, protagonists of us politics. one of the eldest fact-checking websites dates back to : snopes.com, in addition to political figures, is a fact-checker for hoaxes and urban legends. generally speaking, a fact-checking site has behind it a multitude of editors and journalists who, with a great deal of energy, manually check the reliability of a news, or of the publisher of that news, by evaluating criteria such as, e.g., the tendency to correct errors, the nature of the newspaper's finances, and if there is a clear differentiation between opinions and facts. thus, it is worth noting that recent attempts tried to automatically find articles worthy of being fact-checked. for example, work in [ ] uses a supervised classifier, based on an ensemble of neural networks and support vector machines, to figure out which politicians' claims need to be debunked, and which have already been debunked. despite the tremendous effort of stakeholders to keep the fact-checking sites up to date and functioning, disinformation resists debunking due to a combination of factors. there are psychological aspects, like the quest for belonging to a community and getting reassuring answers, the adherence to one's viewpoint, a native reluctance to change opinion [ , ] , the formation of echo chambers [ ] , where people polarize their opinions as they are insulated from contrary perspectives: these are key factors for people to contribute to the success of disinformation spreading [ , ] . moreover, researchers demonstrate how the spreading of false news is strategically supported by the massive and organized use of trolls and bots [ ] . despite the need to educate the user to a conscious fruition of online information through means also different from those represented by technological solutions, there are a series of promising works that exploit classifiers based on machine learning or on deep learning to tag a news as credible or not. one interesting approach is based on the analysis of spreading patterns on social platforms. monti et al. recently provide a deep learning framework for detection of fake news cascades [ ] . a ground truth is acquired by following the example by vosoughi et al. [ ] collecting twitter cascades of verified false and true rumors. employing a novel deep learning paradigm for graph-based structures, cascades [ ] https://www.newsguardtech.com/ are classified based on user profile, user activity, network and spreading, and content. the main result of the work is that 'a few hours of propagation are sufficient to distinguish false news from true news with high accuracy'. this result has been confirmed by other studies too. work in [ ] , by zhao et al. examine diffusion cascades on weibo and twitter: focusing on topological properties, such as the number of hops from the source and the heterogeneity of the network, the authors demonstrate that networks in which fake news are diffused feature characteristics really different from those diffusing genuine information. diffusion networks investigation appear to be a definitive path to follow for fake news detection. this is also confirmed by pierri et al. [ ] : also here, the goal is to classifying news articles pertaining to bad and genuine information' by solely inspecting their diffusion mechanisms on twitter'. even in this case, results are impressive: a simple logistic regression model is able to correctly classify news articles with a high accuracy (auroc up to %). the political blogosphere and the u.s. election: divided they blog ) coronavirus: 'deadly masks' claims debunked coronavirus: bill gates 'microchip' conspiracy theory and other vaccine claims fact-checked extracting significant signal of news consumption from social networks: the case of twitter in italian political elections fast unfolding of communities in large networks influence of fake news in twitter during the us presidential election how does junk news spread so quickly across social media? algorithms, advertising and exposure in public life the role of bot squads in the political propaganda on twitter tracking social media discourse about the covid- pandemic: development of a public coronavirus twitter data set the statistical physics of real-world networks political polarization on twitter predicting the political alignment of twitter users partisan asymmetries in online political activity echo chambers: emotional contagion and group polarization on facebook mapping social dynamics on facebook: the brexit debate ) tackling covid- disinformation -getting the facts right ) speech of vice president věra jourová on countering disinformation amid covid- -from pandemic to infodemic filter bubbles, echo chambers, and online news consumption community detection in graphs finding users we trust: scaling up verified twitter users using their communication patterns opinion dynamics on interacting networks: media competition and social influence near linear time algorithm to detect community structures in large-scale networks randomizing bipartite networks: the case of the world trade web inferring monopartite projections of bipartite networks: an entropy-based approach maximum-entropy networks. pattern detection, network reconstruction and graph combinatorics journalists on twitter: self-branding, audiences, and involvement of bots emotional dynamics in the age of misinformation debunking in a world of tribes coronavirus, a milano la fuga dalla "zona rossa": folla alla stazione di porta garibaldi coronavirus, l'illusione della grande fuga da milano. ecco i veri numeri degli spostamenti verso sud coronavirus: italian army called in as crematorium struggles to cope with deaths coronavirus: italy extends emergency measures nationwide italy sees first fall of active coronavirus cases: live updates coronavirus in italia, verso primo ok spostamenti dal / , non tra regioni italy's health care system groans under coronavirus -a warning to the world negli ospedali siamo come in guerra. a tutti dico: state a casa coronavirus: ordini degli infermieri, mila i contagiati automatic fact-checking using context and discourse information extracting significant signal of news consumption from social networks: the case of twitter in italian political elections controlling the false discovery rate: a practical and powerful approach to multiple testing users polarization on facebook and youtube fast unfolding of communities in large networks the role of bot squads in the political propaganda on twitter the psychology behind fake news the statistical physics of real-world networks fake news: incorrect, but hard to correct. the role of cognitive ability on the impact of false information on social impressions echo chambers: emotional contagion and group polarization on facebook graph theory (graduate texts in mathematics) resolution limit in community detection maximum likelihood: extracting unbiased information from complex networks. phys rev e -stat nonlinear on computing the distribution function for the poisson binomial distribution reconstructing mesoscale network structures the contagion of ideas: inferring the political orientations of twitter accounts from their connections comparing clusterings by the variation of information fake news detection on social media using geometric deep learning at the epicenter of the covid- pandemic and humanitarian crises in italy: changing perspectives on preparation and mitigation. catal non-issue content near linear time algorithm to detect community structures in large-scale networks randomizing bipartite networks: the case of the world trade web inferring monopartite projections of bipartite networks: an entropy-based approach the spread of low-credibility content by social bots analytical maximum-likelihood method to detect patterns in real networks a question of belonging: race, social fit, and achievement cognitive and social consequences of the need for cognitive closure fake news propagate differently from real news even at early stages of spreading analysis of online misinformation during the peak of the covid- pandemics in italy supplementary material guido caldarelli , , * † , rocco de nicola † , marinella petrocchi † , manuel pratelli † and fabio saracco † there is another difference in the label propagation used here against the one in [ ] : in the present paper we used the label propagation of [ ] , while the one in [ ] was quite home-made. as in reference [ ] , the seed labels of [ ] are fixed, i.e. are not allowed to change [ ] . the main difference is that, in case of a draw, among the labels of the first neighbours, in [ ] a tie is removed randomly, while in the algorithm of [ ] the label is not assigned and goes into a new run, with the newly assigned labels. moreover, the updated of labels in [ ] is asynchronous, while it is synchronous in [ ] . we opted for the one in [ ] for being actually a standard in the label propagation algorithms, being stable, more studied, and faster [ ] . finally, differently from the procedure in [ ] , we applied the label propagation not to the entire (undirected version of the) retweet network, but on the (undirected version of the) validated one. (the intent of choosing the undirected version is that in both case in which a generic account is significantly retweeting or being retweeted by another one, they do probably share some vision of the phenomena under analysis, thus we are not interested in the direction of the links, in this situation.) the rationale in using the validated network is to reduce the calculation time (due to the dimensions of the dataset), while obtaining an accurate result. while the previous differences from the procedure of [ ] are dictated by conservativeness (the choice of the seed labels) or by the adherence to a standard (the choice of [ ] ), this last one may be debatable: why choosing the validated network should return "better" results than the ones calculated on the entire retweet network? we consider the case of a single day (in order to reduce the calculation time) and studied different approaches: a louvain community detection [ ] on the undirected version of the validated network of retweets; a louvain community detection on the undirected version of the unweighted retweet network; a louvain community detection on the undirected version of the weighted retweet network, in which the weights are the number of retweets from user to user; a label propagation a la raghavan et al. [ ] on the directed validated network of retweets; a label propagation a la raghavan et al. on the (unweighted) retweet network; a label propagation a la raghavan et al. on the weighted retweet network, the weights being the number of retweets from user to user. actually, due to the order dependence of louvain [ ] , we run several times the louvain algorithm after reshuffling the order of the nodes, taking the partition in communities that maximise the modularity. similarly, the label propagation of [ ] has a certain level of randomness: we run it several times and choose the most frequent label assignment for every node. key: cord- - dh wg x authors: patrì, angela; gallo, lucia; annunziata, maria c.; megna, matteo; fabbrocini, gabriella title: covid‐ pandemic: university of naples federico ii dermatology's model of dermatology reorganization date: - - journal: int j dermatol doi: . /ijd. sha: doc_id: cord_uid: dh wg x nan closed. the italian authorities have established fines and penal condemnations for those who do not comply with the restrictive measures. the economic and psychological impact of the epidemic on italian people will be great; however, the efficacy of lockdown is now manifesting with decreasing new infected cases as well as the number of patients hospitalized. we all must contribute to the infection containment. "we urge dermatology departments and practices to show leadership. if not now, when?" with these words kwatra et al. clinic has been infected with sars-cov- to date. the situation of private dermatological offices is different: they are indefinitely closed, except for extremely urgent visits. however, private dermatologists are connected through an online network with the hospital structure: they can send patients who need urgent visits as well as can support the research activity of our university. in southern italy, the epidemic has not reached the numbers of the northern regions, and we hope we don't face the same terrible emergency in the immediate future thanks to the all restrictive measures actually in place. we hope our mistakes as well as our initiatives can serve as a lesson for the other countries. dermatologists, let's start first: it's time to act. italy struck by deadliest day as virus prompts industry shutdown dermatology practices as vectors for covid- transmission: a call for immediate cessation of non-emergent dermatology visits emergency management for preventing and controlling nosocomial infection of novel coronavirus: implications for the dermatology department what we know so far: covid- current clinical knowledge and research global coronavirus pandemic ( -ncov): implication for an italian medium size dermatological clinic of a ii level hospital key: cord- -uit nm authors: arnold, theresa; gulati, mitu; panizza, ugo title: how to restructure euro area sovereign debt in the era of covid- date: - - journal: nan doi: . /cmlj/kmaa sha: doc_id: cord_uid: uit nm nan facing holdout issues on the local-law debt. this was also the case with barbados' debt restructuring in , which followed the greek template. however, both greece and barbados faced significant holdout problems in their foreign-law-governed debt; many of the foreign-law creditors refused to take the same offers that were made to the local-law creditors and ended up receiving higher recoveries. in the analysis that follows, we focus on three strategies a euro area sovereign might utilize in a possible future crisis to solve the primary source of disruption to modern day sovereign debt restructurings: the holdout creditor problem. therefore, in the analysis that follows, we assume that the sovereign and its official sector backers have been able to negotiate a restructuring deal with the majority of creditors, but those creditors do not want to enter into an agreement where either their recoveries are at risk of disruption from litigation by holdout creditors or they are in danger of being embarrassed by having suffered much bigger haircuts than the holdouts. we also put to the side the possibility of legal changes at the broader european level that might ameliorate the holdout problem more directly, and abstract from the possibility of reforms that will create a clear seniority structure within euro area bonds. although the analysis that follows might be applied to other euro area sovereigns, we look specifically at italy for the purposes of our analysis. by using italy as an example, we do not mean to suggest that it needs to restructure its debt pre-emptively. however, italy did face a confidence crisis in late and the coronavirus has hit it particularly hard in . italy's gargantuan debt stock of around e . trillion and a debt to gdp ratio of per cent make italy vulnerable to another confidence crisis. the economic recession and ). an alternative proposal which is fiscally neutral but could reduce the likelihood of rollover risk is the padre plan. see pierre pâris and charles wyplosz, padre: politically acceptable debt restructuring in the eurozone, geneva special report on the world economy , icmb and cepr ( ) . for a discussion of the risks associated with a pre-emptive italian debt restructuring, see giampaolo galli, why restructuring of the italian public debt should be avoided ( ) (osservatorio cpi <https://osservatoriocpi.unicatt.it/cpi-why_a_restructuring_ should_be_avoidced_gg_ocpi.pdf>); ugo panizza, 'come risolvere il problema del debito pubblico italiano: un'analisi critica delle soluzioni facili' (forthcoming ) rivista di politica economica. for a prior discussion by one of us, see ugo paniza, public debt in italy: myths, facts and policies ( draft) <http://repec. graduateinstitute.ch/pdfs/working_papers/heidwp - .pdf> ferdinando giugliano, the euro's $ . trillion dollar problem, bloomberg, april ( ) <https://www.bloomberg.com/ news/articles/ - - /the-euro-s- - -trillion-italy-problem> accessed july ; peter s goodman, 'italy's biggest economic problem? it's still italy', ny times, august ( ), <https://www.nytimes.com/ / / /business/italy-economydebt.html> theresa arnold, mitu gulati and ugo panizza • euro area sovereign debt in the era of ballooning expenditure needs associated with covid- are expected to bring the debt-to-gdp ratio well above per cent. quite possibly, such a crisis could be dealt with through some combination of fiscal adjustment, economic reform and official (esm/ omt) financial support. this said, it is also conceivable that such a crisis might lead to a situation in which italy might have to restructure its debts. we thus ask two related questions: in such a situation, what restructuring options might italy have? and could italy do something now to put itself in a better position to deal with such a situation in the future? two considerations for any government in undertaking a debt restructuring are: (i) how to minimize the impact of that restructuring on its costs of future borrowing in situations where the sovereign needs to be able to tap those markets regularly; and (ii) how to minimize the domestic economic and political costs in situations where a large portion of the debt is held domestically. the balance of the foregoing considerations, along with the interests of official sector sponsors, will determine how the choice among restructuring options is made. we do not attempt the exercise of prescribing a specific restructuring plan for italy. instead, we put what we see as the most plausible options on the table. the options to prepare for a possible future debt restructuring fall into three categories: do nothing; do a little; or do a lot. one can think of these options in terms of the choice of what type of crisis insurance to buy: none at all; cheap, partial-coverage insurance; or expensive, full-coverage insurance. the article proceeds as follows. section describes the basic data on italian sovereign debt. section works through the three options noted above. section concludes. according to italian treasury and bank of italy data, as of mid- , italian public debt amounts to approximately e . trillion (table ) . the main holders of italian public debt are domestic banks and other domestic financial institutions (e . trillion, corresponding to per cent of the total), non-resident entities (e billion, corresponding to per cent of the total) and the bank of italy (e billion, corresponding to per cent of the total). italian households, which in the early s held about per cent of the italian public debt, now hold less than per cent of the total. about per cent of the total stock of italian debt (e trillion) consists of tradable instruments (government bonds and bills), which we refer to as "bonded debt". the data show that per cent of bonded debt is held by domestic banks and other domestic financial institutions, per cent by non-residents, per cent by the bank of italy and the for an early discussion of the economic impact of the covid- crisis on italy, see ugo panizza, 'europe's ground zero', in richard baldwin and beatrice weder di mauro, mitigating the covid- economic crisis: act fast and do whatever it takes (cepr press ). the most recent bank of italy data are available at <https://www.bancaditalia.it/pubblicazioni/finanza-pubblica/index.html>. treasury data on the composition of bonded debt are available at: <http://www.dt.tesoro.it/en/debito_pubblico/dati_statistici/sca denze_titoli_suddivise_anno/ . bank of italy data> used in this paper refer to the end of july , while italian treasury data refer to the end of september . there is a small discrepancy between bank of italy data, which report a total for bonded debt of e , billion, and treasury data which report a total for bonded debt of e , billion. remaining per cent by households. the remaining debt is held by italian banks (e billion, corresponding to per cent of the non-bonded debt), resident households (e billion, corresponding to per cent of the non-bonded debt) and non-residents (e billion, corresponding to per cent of the non-bonded debt). more than per cent of italian bonded debt is denominated in euro (issues in us dollars amount to e . billion, issues in british pounds amount to e . billion and issues in japanese yen amount to e million; table ). most bonded debt consists of treasury bonds (btp, per cent of the total), treasury certificates (cct, per cent of the total) and treasury bills (bot, per cent of the total), all of which are governed by local italian law. a small share of bonds are listed on the luxembourg exchange, under a medium term note program (emtn). most of these bonds are denominated in euro, but a small amount are in british pounds and japanese yen. all of these luxembourg-listed bonds, as best we can tell, are governed by local italian law, although as we detail later, these bonds do have different (and better) legal protections for creditors than the aforementioned treasury bonds, treasury certificates and treasury bills. finally there is a small amount of debt in the form of global bonds, issued under a us programme and governed by new york law (roughly e billion, denominated in us dollars). in sum, about per cent of italian debt is governed by local law and, therefore, benefits from the local-law advantage ( . per cent if we include emtn and . per cent excluding emtn; figure ). about one-third (e billion) of outstanding btps were issued before and, hence, do not include collective action clauses (cacs) that are now required by the eu in order to provide an orderly path to restructuring. other bonds that do not include cacs include foreign-law bonds issued before , ispa bonds (these were bonds issued before to finance infrastructure projects), and bot. in total, bonds and bills without government bonds of the euro zone with maturities in excess of one year, the treasury will albeit continue to issue, with its usual regularity, bonds without the above clauses, these being outstanding bonds at / / and off-the-run bonds, so as to guarantee the necessary liquidity.' hence, the share of bonds without cacs is likely to be slightly higher than what is reported here. cacs amount to approximately e billion (ie, per cent of total outstanding bonded debt; per cent of the total if we exclude bot, which amount to approximately e billion). figure shows the maturity profile of outstanding italian bonds. during - , the italian treasury will need to roll over about e billion per quarter, with large amounts of btp without cacs that need to be rolled over in q and q . figure shows the residual stock of bonds without cacs. while the curve is steep, starting with e billion in q and dropping to e billion in q , this is mostly due to maturing bot. after the bot effect ends, the curve flattens and by , italy will still have more than e billion of bonds without cacs. the top panel of figure plots the expected composition of italian bonded debt over the period - . it shows that in , more than per cent of bonded debt will still consist of long-term bonds without cac, if we add bots and hard-to-restructure foreign law and emtn bonds, we find that in , almost one-quarter of italian bonds are hard to restructure. if to this debt we add loans (ie non-bond debt), we find that only two-thirds of italian public debt consists of bonded debt with cacs (bottom panel of figure ). the do nothing approach is for the italian authorities to literally do nothing. that is, to take no steps to utilize the local-law advantage in advance of a potential crisis. this is perhaps the most likely outcome, since governments are loath to admit that a debt crisis might be in the offing. the main reason for a country to take the do nothing approach is the concern that taking steps to prepare for a crisis will send a negative signal to the market and either lead to an increase in sovereign yields or, worse, cause the crisis to happen. in most cases, this head-in-the-sand approach is far from ideal. but, as explained below, it may not be a terrible approach in the italian case, given the advantages the legal terms governing the current italian debt stock provide. as a starting point, italy will have first-generation euro cacs in over per cent of its bonded debt stock by mid- ( per cent of its total debt stock, see figure ). these clauses were put in place starting in january , mandatorily, for all euro area sovereign debt issuances of over a year in maturity. simplifying somewhat, the presence of these clauses means that all of the bonds containing them can be modified in a single transaction, so long as the approval of . per cent of the holders of each bond and per cent of the holders in all the bonds put together (in principal amount) is obtained. however, there are issues. the requirement that the approval of . per cent of the creditors in each bond be obtained to effect a consensual debt restructuring gives rise to the risk of holdout creditors targeting a subset of the smaller bonds and accumulating blocking positions in them. in the greek restructuring of , where the majority of foreign-law bonds were governed by english law and had cacs that could only be activated with super-majority votes, there were holdouts in separate sovereign and sovereign guaranteed bonds (amounting to about e . billion in face amount of debt). the holdouts, after blocking the votes in their bonds, got paid in full. and this was in large part because greece did not want to risk the type of litigation that argentina was suffering at that same time. for the italian local-law bonds with cacs, the vote requirement is lower than what was required in the greek foreign-law bonds ( per cent, with a quorum requirement in the context of a bondholder meeting). but this is unlikely to be a significant barrier for a wellcapitalized hedge fund seeking to utilize the holdout strategy. in addition, if a creditor can figure out what the holdings of the european central bank (ecb) are in particular bonds (and they are substantial in many bonds), he/she will be in an even stronger position to hold out. the reason is that the ecb has asserted a , , , , source: own elaboration based on imf data and italian treasury data. in order to build the debt composition for - , we started with imf forecasts for the total stock of debt and then assumed that the shares of loans and bot remain constant. data for external debt, and domestic bonds are instead taken from the flow of maturing bonds data used for figure . the graph is built under the conservative assumption that no external bonds will be issued in the next three years; while italy issued dollar bonds in october , this was a small amount ($ billion) and these issuances-and possible future external issuances-are unlikely to substantially change the figure given their relatively small amount. commitment to not voting in favour of any debt restructuring plan. in a cac that calls for an affirmative vote of the creditors of a certain percentage of principal amount, abstention is scored as a no vote. the ecb's commitment, therefore, effectively lowers the percentage the holdout would be required to purchase in order to block a modification. the foregoing has already caused concern in policy circles-and has resulted in proposals for the euro area to reform its cacs starting in . that said, an argument could be made that the potential holdout problem is less serious than it has been made out to be. one way to see this is by detailing the steps a holdout creditor seeking to litigate against a distressed sovereign generally needs to take. assuming that the bond in question has not yet come due but the sovereign is in default on other indebtedness, the holdout will typically utilize her bond's cross default provision to accelerate the payments owed on her bond, demanding immediate payment of principal and accrued interest. the ability to declare a default and accelerate is key in the litigation context because suing the sovereign on small unpaid coupon payments is unlikely to be worthwhile when measured against legal costs. italian local-law bonds though, except for the sliver listed in luxembourg under the emtn program, have neither cross default nor acceleration provisions. in other words, the only bonds worth holding out on will be the ones where a substantial principal amount is due. worse, from the holdout creditor's perspective, those bonds (except again for the luxembourg-listed ones) lack explicit waivers of sovereign immunity from either suit or execution, which means that the italian government lawyers can potentially use that as a defence. put that together with the dysfunction of the italian legal system, and it is hard to imagine one of the specialist hedge funds that won large recoveries against argentina in the new york courts wanting to invest in holding out on the non-luxembourg listed italian local-law bonds. in contrast to italian creditors, the foreign-law holdout creditors in the greek restructuring in had the rights to declare cross defaults, accelerate the debt and sue for recovery in english, swiss or japanese courts under the explicit consents to jurisdiction and waivers of immunities from suit and execution in the contracts. absent those rights to a key question will be whether there is an implicit waiver of sovereign immunity under italian law. ordinarily, we would presume that an italian court would find such an implicit waiver as a matter of customary international law since sovereign bond issuances are considered a commercial activity. however, italian courts have shown a willingness to reinstate immunity for actions of the state that are necessary to tackle an extreme crisis (and covid- appears to have created just such a situation). - . in the most recent doing business ranking from the world bank, italy ranks nd out of countries in terms of ease of contract enforcement. average enforcement time is , days and the cost of litigation, as a fraction of the claim is . %. those do not strike us as numbers that would attract litigation oriented holdout creditors. for details, see <https://www.doingbusiness. org/content/dam/doingbusiness/country/i/italy/ita.pdf> theresa arnold, mitu gulati and ugo panizza • euro area sovereign debt in the era of declare defaults, accelerate, bring suit and execute against the state's assets in a jurisdiction where the legal system moves quickly and efficiently, a sovereign such as italy might hope that it would simply be unattractive for any investors in its bonds to hold out. if so, doing nothing might work. that said, the costs of default for a country that needs to regularly tap the borrowing markets can be substantial. and potential holdout creditors, especially if they watched what happened in greece a few years prior, might recognize their potential to impose costs on italy even in the absence of substantial litigation rights. this is because during the restructuring period the country is likely to be given a selective default rating by the credit rating agencies, which will make new borrowing hard. plus, its bonds are unlikely to be eligible for delivery at the central bank's discount window, making them an unattractive investment. the end result will probably be an increase in sovereign borrowing costs beyond the usual default penalty. all in all, a country like italy may not want to have a number of defaulted bonds outstanding after its restructuring. let us assume that the prospect of being in a long-term payment default vis-à-vis a substantial number of creditors, even if their litigation options are not attractive, is a prospect that the italian government does not relish. the question then is whether there are some non-obvious options available under existing italian law. we see two: (a) an extension of maturities of those bonds that were issued at low rates in prior years and that are coming due; and (b) the imposition of a withholding tax equivalent to the haircut taken voluntarily by non-holdouts. as we discuss further below, these options pose some litigation risk, and italy might choose to 'do a little' by taking steps to mitigate that risk. as a threshold question to examining these options, we must ask whether euro area countries contemplating a restructuring are constrained to utilizing only the current euro cacs and no other technique. the answer here, we think, is no for a number of reasons, including the explicit language of the cacs and the policy intent behind them. to clarify the question, let us say that the italian government decides that it wants to reduce its debt burden by imposing a withholding tax (more on this later). do the euro cacs in its existing bonds bar the use of this strategy? no. there is nothing in the mandate requiring the inclusion of euro cacs that constrains any individual euro area country from using its tax authority. indeed, there could not be, as italy has retained the authority to levy taxes, even as it has delegated other aspects of sovereignty to the euro area. italy is constrained in terms of the contract provisions it has agreed to with bondholders, but the vast majority of its local-law bonds (except those pesky luxembourg-listed ones) do not contain the protections against the sovereign imposing withholding taxes that are standard in foreign-law bond issuances (including in those issued by italy). a similar analysis applies to other routes to enable an easier restructuring, such as making it harder to sue in italian courts; increasing the costs of litigating a contracts case, for example. that would not be particularly good for the italian legal system and economy generally, but it would deter holdout creditors with debt instruments whose holders were required to litigate in italian courts under italian law. and there is nothing in the body of european law respecting cacs that constrains italy from taking such actions. the language of the cacs themselves signals the possibility of other restructuring techniques being used. in relevant part, they say: in the case of a cross-series modification, the terms and conditions of the bonds and debt securities of any other series . . . may be modified in relation to a reserved matter with the consent of the issuer and [the requisite proportion of bondholders]. the word 'may' is crucial; it tells us that the cacs are a debt restructuring option that the sovereign may use. but just one option. if these cacs were meant to say that sovereigns were constrained from using other restructuring options, they would have said 'must' instead of 'may'. one might ask whether there is anything in the esm treaty, whose article mandated that all euro area nations utilize cacs, suggesting that future debt restructurings of the bonds with cacs could only proceed by utilizing the cacs. there is not. nor is there anything in the history of why these cacs were put in place by the european authorities that suggests so. the history tells us that, starting with the infamous deauville beach walk on october , the utilization of cacs was meant to set up an orderly mechanism for future debt restructurings in europe and to signal to investors that the purchasers of euro area sovereign bonds that those bonds could be restructured in the future. mark weidemaier, in his paper, sums this up: [a euro area] sovereign that has issued local-law debt [with cacs] remains free to alter its law to facilitate restructuring, although it will encounter various legal constraints in doing so. these constraints, however, are not absolute; there is room for the prudent exercise of the local law advantage. having determined that the existence of the cacs does not preclude the sovereign from using the local-law advantage to facilitate other paths to a restructuring, we turn to a closer examination of the two options mentioned above: a maturity extension and the imposition of a withholding tax. a paper from (edelen et al.) , and two follow-ups from (cervantes et al. and cramer et al.) , suggest that there might be a strategy to unilaterally extend maturities of debt with especially low coupon rates. assuming this strategy is workable, the idea would be to ask the holders of local-law bonds with especially low coupons to voluntarily extend maturities. the addendum to that polite request would be that those who refused would have their maturities unilaterally extended by law. the provision that the aforementioned papers hone in on is article of the italian public debt consolidated act (the "debt act"). under article : [ the ministry] has the authority, within the annual limits established by the budgetary law, to issue framework decrees that allow the treasury to . . . proceed, in order to restructure the national and external public debt, to the reimbursement before maturity of bonds, to the transformation of maturities. that language seems clear on its face. it says that the ministry of economy and finance (the 'ministry') has the authority to issue decrees to allow the treasury to either redeem bonds at par (maybe because interest rates have dropped considerably since issuance) or transform (extend) maturities (maybe because interest rates have increased considerably since issuance). in both cases, investors lose and would presumably impose a reputational cost on the issuer. but the options to either reduce or extend the duration of a loan are familiar contractual provisions in the debt world (particularly the former ). one might question whether, if italy does hit turbulent times, this power to extend maturities will be of much value. after all, countries in deep sovereign debt crises typically need to impose principal haircuts as well as maturity extensions. the objection is a fair one. but, given the large amount of the italian debt that is held domestically and the negative impact on the economy that will likely be caused by imposing brutal principal haircuts to the holdings of domestic financial institutions, maturity extensions may be an attractive option for the government. this is especially so if a significant portion of the italian debt stock was incurred when borrowing rates were especially low (as is the case). under the structure of article , the treasury department receives the power to call bonds or extend maturities only if a framework decree authorizing such actions is issued by the ministry. these decrees are not law; they are administrative actions implementing legal authority. they are also of limited duration. each decree lasts for a year, no more. in prior years, they have been used to authorize the treasury to do both debt exchanges and redemptions. it was also under article 's structure that a framework decree was issued by the ministry to impose euro cacs starting in . let us say therefore that the italian government decides to issue a framework decree in to authorize the treasury, if need be, to extend the maturities of all italian local-law bonds by five years. there are at least two possible barriers to the treasury utilizing this authority. we take the potential barriers in turn. first, we consider article of the debt act. this is perhaps the most significant barrier to the use of article . it requires that 'payments of public debt [cannot be] reduced, paid late or subject to any special levy, not even in case of public necessity'. the question then is whether extending maturities or reducing them (as article seems to allow the treasury to do) violates the '[may not be] paid late' requirement. at first cut, it does not appear so, since that reading would reduce article to meaninglessness. that said, the argument could also be made that our reading of article reduces article to meaninglessness as well. and courts generally do not like to negate statutory provisions. the reality, therefore, is that if italy decides to go down this path, a court will likely have to decide which of the clauses to favour, while putting a gloss on the other. and italy's hope would have to be that the court chooses to interpret article in a manner that weakens it, such as by reading it to require simply that whatever the maturity date iswhether shortened, lengthened or kept the same-it has to be complied with. the foregoing is not implausible in the context of the euro cacs. the reason being that if article were to be interpreted to bar extensions of maturity, that would also negate crucial provisions of the cacs that, by italian law, explicitly allow for the sovereign to extend maturities of the bonds with euro cacs with a super majority approval of the creditors. the question for the italian government will be whether it wishes to go down this path of litigation. earlier we discussed the question of whether the existence of the euro cacs as a restructuring mechanism constrains the use of article 's power generally. now, we address the question of whether the treasury might use its article power in a specific bond that contains a euro cac as part of its contract terms. after all, if a crisis were to occur at any time over the next few years, there would be more italian local-law debt with cacs than without. the euro cacs have an explicit mechanism by which the maturities of the bonds can be extended, requiring a pre-specified vote of the creditors. our view is that, for those bonds, the right of the sovereign to unilaterally extend the maturities is dubious. the reason being that these bonds already have a term explicitly governing how maturities may be extended. but for the remaining debt-which we estimate in the years - to range between per cent and per cent of the total debt stock ( figure )-there are no cacs. for that latter subset, there is a strong argument that italy can use the power that it already has by law to unilaterally extend maturities. we should caution though that such a power is unlikely to be unlimited. if the sovereign, for example, tried to extend the maturities of its one-year bonds by years, there would likely be an immediate expropriation challenge. because the option of unilaterally extending maturities is likely to only work for locallaw bonds that do not contain cacs and the relief that italy would obtain would be limited compared to a more significant restructuring that reduces principal amounts as well as extending maturities, italy may wish to pursue other options presented in this article. however, if italy wanted to make this tool a more feasible option for a future restructuring, it could reduce the litigation risk associated with this option by proactively clarifying the apparent discrepancy between article and article . indeed, repealing article might be even better. another objection that we have heard informally from some in the italian government has to do with the ministry's annual decrees issued between and , which say that restructurings in those years will occur via 'mutual consent'. this guidance, we suspect, was a signal to the markets that restructurings would be done via a consultative process. there is no specification of what that consultative process would be, nor have we been unable to uncover evidence of the intent behind the words 'mutual consent'. but we have reason to think that 'mutual consent' does not mean that unanimous approval of creditors is required. there are two points worth considering here. given that bonds issued after contain euro cacs, "mutual consent" cannot possibly mean a requirement of unanimous creditor approval because that would contradict the explicit super majority requirements of the euro cacs for maturity extensions. further, the mutual consent requirement, whatever it means, comes out of an annual decree. it is up to the ministry, every year, to decide whether it wishes to use that language in its decrees. if it happens to be a year where the ministry wishes to give the treasury to power to unilaterally extend maturities, it can delete the 'mutual consent' language from that year's decree, as it has in many prior years. for discussions of the argument that the power of unilateral extension might apply to the bonds with cacs as well, see cervantes et al (n ); cramer et al (n ). we do not, however, think that an expropriation objection would work generally against all uses of the art power to unilaterally extend maturities. expropriation objections apply to a state taking away property rights that creditors have been granted (eg, the greek retrofit of ). however, art has been in place since at least. see cervantes et al (n ). that said, italy would be on stronger ground in the future if it issued a clarification of art 's powers earlier rather than later. inherent to sovereignty is the power to tax. bondholders who are owed moneys by a sovereign have, therefore, since time immemorial, worried about the sovereign deciding to impose a new tax on the payments it owes as a way to avoid payment. as a result, almost every international sovereign bond governed by the laws of new york or england contains what is called the tax gross up clause. this clause protects the bondholder against the sovereign deciding to levy an extra tax on the payments it owes on the bond as a way of doing a back door restructuring. most local-law-governed bonds, however, do not contain such a protection. and that is the case for all of italy's local-law bonds, except the small fraction of luxembourg-listed ones mentioned earlier. a simple strategy to deal with holdouts, therefore, is for the minister of finance to give a speech to local bankers in which she says that holdout bonds will be taxed so that the net (after tax) return will equal the payout on the new bonds. this was tried in jamaica's restructuring of local-law bonds in . the minister's threat was successful and he never in fact had to introduce the tax. the fly in the ointment is if the country has bilateral investment treaties and double taxation treaties that bar such actions against the citizens of other countries. an italian government considering such a strategy would need to consider the risk of future litigation under these treaties. indeed, it could reduce the risk of this litigation by being careful about the provisions it agrees to in any new treaties. taking steps today to reduce the future litigation risk associated with an exercise of italy's power to unilaterally extend maturities or impose a withholding tax on holdouts would improve italy's future restructuring options. but neither of these options provides italy with as much flexibility for significant debt relief as the option of retrofitting more effective cacs, described in the next section. governments, as we noted at the start, are generally reluctant to do anything that might signal to the outside world that they are concerned about the possibility of a financial crisis; even if everyone recognizes that it is a real possibility. so, 'do a lot', is not a strategy we expect to be embraced. nevertheless, let us assume for the purposes of this section that a sovereign such as italy would want to minimize the future costs of potential holdouts in a possible restructuring. buchheit explains how the clause works: 'any deduction as a result of the imposition of withholding taxes thus becomes the borrower's responsibility because the borrower's payment to the lender must [by virtue of the tax gross up clause] be topped up to compensate for the deduction.' lee c buchheit, how to negotiate eurocurrency loan agreements ( st edn ). we have found no mention of this technique in the sovereign debt literature. unsurprisingly, sovereign debt restructuring guru, lee buchheit, advised jamaica at the time. we have not delved into the complexities of italy's tax agreements with other countries. the aforementioned art could pose a barrier here as well, since it bars any 'special levy'. as mentioned earlier, clarifying that art was not meant to pose such a barrier in advance of a crisis might be advisable. theresa arnold, mitu gulati and ugo panizza • euro area sovereign debt in the era of operating under that assumption, we think that the strategy of enhancing the existing cacs in bonds that have them and retrofitting them in the debt that does not already have them is a good option for a number of reasons that we explain below. as noted earlier, all euro area sovereign bonds issued since with a maturity greater than one year already have cacs in them. but these cacs contain the requirement that a . per cent vote of the holders, in principal amount, be obtained for each bond in order to use the aggregation feature to modify all of the bonds containing the cacs. and this creates a vulnerability to holdout creditors. further, there is a whole set of debt instruments that are going to be even more vulnerable to holdouts, which are the bills with maturity under a year, the long-term bonds issued before , government guaranteed instruments, non-bond debt (eg, syndicated loans) and so on. the reason being that cacs were not made mandatory for these sovereign debt instruments. euro area sovereigns could have inserted them voluntarily; but, as far as we know, none did. realistically though, if the crisis is an especially serious one the government is going to need to restructure some of these instruments as well. thus far, we have tackled the easy question of whether the sovereign might use one of its other powers-such as its taxing authority-to engineer a debt reduction without running afoul of the mandatory cacs. the answer was yes. the trickier, albeit closely related, question is whether the sovereign can enhance the existing cacs in its bonds by using its local-law authority. by enhancing, we mean making changes to the existing model cacs provided by the european authorities to make the bonds even less vulnerable to holdout problems than they are now. in particular, two modifications that have already been the subject of debate among policy makers are: (a) introducing a mechanism by which the entire debt stock of a country can be modified at a single shot; for example, via a vote of per cent of all the holders across all the bonds, so long as some non-discrimination criterion is satisfied (the so-called single shot or single-limbed feature); and (b) an addendum to the voting rules for modifications that disenfranchises the ecb from having its votes counted (assuming that the ecb continues to hold to the position that it is legally constrained from voting). conventional wisdom in sovereign debt restructurings is that short maturity instruments (under a year), trade credits, and guaranteed debt, generally get paid in full. but that assumption only holds where the amount of debt in these instruments is small. in the recent barbados restructuring of - , the amount of short-maturity debt was so large that it had to be restructured. and, since those instruments had no cacs, they had to be retrofit into them in the same manner as was both changes to the existing euro cac structure correct for design flaws in the original version and enable the cacs to better achieve their goals of deterring holdouts while providing a clear and predictable mechanism by which creditors can participate in a burden sharing exercise. at the time the original euro cacs were designed, it was thought that a bond-by-bond vote requirement of . per cent would be enough to deter holdouts. and the implications of the ecb not being able to vote in a restructuring (or having to always vote no, regardless of the restructuring proposal) had not been fully considered. so, the question is whether an individual nation, such as italy, concerned about the possibility of a crisis, can act on its own via its 'local law' advantage to legislate improvements in its cacs consistent with the original goals of those cacs. there are two potential barriers here. first, for new bonds issued with the improved cacs, there is the question of whether deviating from the form mandated in violates european law. secondly, for already-issued bonds where these provisions are retrofit via legislative action, there is the potential for creditor lawsuits claiming an interference with property rights. we take the possibilities in turn. neither poses a significant barrier. a country including enhanced anti-holdout provisions in its new bonds potentially violates the instruction in the esm treaty that cacs be introduced in a manner such that their 'legal impact is identical'. the argument against italy being able to fix the design flaws in its cac would be that all european countries are required to have the same flaws until the european authorities decide it is time to fix those flaws for everyone. there is a certain formalist appeal to the foregoing. 'identical legal impact' arguably means the same in every way. but, the very manner in which these euro cacs were designed tells us that the creators could not have meant to use the term 'identical legal impact' in such a broad fashion. the more likely meaning of 'identical legal impact' is much narrower. that is, that every country would have an identical baseline of the same basic cac framework that would signal to holders of sovereign bonds that a debt restructuring was possible. and that one way (not the only way) by which this restructuring could be implemented would be via the basic euro cac. and everyone had to have that identical baseline. there was no prohibition, however, on the sovereign enhancing the operation of the cacs by including additional provisions to its debt contracts. one way to see the foregoing is to compare the operation of cacs in the italian foreign-law (new york law) versus local-law bonds. art ( ) of the esm treaty says: 'collective action clauses shall be included, as of january , in all new euro area government securities, with maturity above one year, in a way which ensures that their legal impact is identical'. some commentators do take this view. for example, klaus-albert bauer writes: 'a natural understanding of [esm art ] this clause would seem to prohibit not only the issue of bonds with non-conforming collective action clauses but also the later unilateral amendment of euro area cacs to suit a particular issuer's needs in times of crisis.' klaus-albert bauer, 'the euro area collective action clause-some questions and answers at ', in collective action clauses and the restructuring of sovereign debt (n ). theresa arnold, mitu gulati and ugo panizza • euro area sovereign debt in the era of the cacs in the italian foreign-law bonds, such as the one issued in october , are buttressed with all sorts of potent investor protections including acceleration clauses, cross default provisions, waivers of immunity and consents to jurisdiction. what this means is that cacs in the foreign-law bonds, if the bondholder has the votes to block a restructuring attempt, can be turned into a potent litigation weapon. assets can be seized, payments to other bondholders can be blocked and so on. conversely, in the local-law bonds that lack these protections, there is no weaponry for the holdout creditor other than the hope that the sovereign does not want to suffer the reputational cost of default. the point being that the issuer, by including a stand-alone cac in the local law bonds and not adding extra creditor protections, has an effective anti-holdout weapon, rather than the possible creditor weapon that the cac tends to be in the foreign law bonds. both the localand foreign-law bonds contain identical cacs, but they can potentially have radically different effects as a function of the other contractual bells and whistles that the issuer adds (or chooses not to add) to the baseline. and this was explicitly envisioned right from the start by the european authorities, who understood that sovereign bonds would be issued under both local and foreign law and that the sovereigns may want to include enhancements such as acceleration clauses or trustee structures. plus, the actual cacs for foreign-law bonds were designed differently in one crucial fashion right from the start-the governing law is a reserved matter that cannot easily be changed in the foreign-law bonds whereas it is not a reserved matter at all in the local-law bonds. that is a significant difference, as weidemaier ( ) points out. in other words, the cacs were designed such that they could not have identical impact (in its most literal sense) right from the start. given the foregoing, we think that the sensible interpretation of article ( )'s dictates is that every euro area nation needs to implement identical model cacs. but those serve as a baseline; ensuring that creditors all are forewarned that a restructuring of sovereign debt in the future is possible and that one method that might be used is the basic euro cac mechanism. nothing, however, stops the sovereign from either adding in more investor protections (making the cac a weaker tool from the sovereign's perspective) or adding improvements to the voting mechanism (making the cac a stronger tool from the sovereign's perspective) as long as the basic model cacs are kept in place. to reiterate, the euro cacs serve as an identical baseline for all euro area sovereigns. from a policy perspective, it is easy to see why the foregoing makes sense. euro area sovereigns vary vastly in terms of their economic strength and the types of markets they are able to tap. that means that they will likely need to offer different contractual rights to investors in order to borrow effectively; especially in a world where the european authorities have made it clear that they are not guaranteeing the payment of each individual sovereign's claims. to the extent a sovereign wants to prepare for a future crisis by experimenting with using enhanced restructuring tools that build on what the european authorities have already provided, and particularly if this is done in consultation with the european authorities, it is hard to see why those authorities would not wish that. the experimentation by different sovereigns with various enhancements can only serve to provide the authorities with helpful information. last but not least, there is the question of whether, even if there was a hypothetical violation of the article ( ) 'identical legal impact' dictate, there would be any practical impact. with the caveat that the matter of international treaty enforcement is beyond our expertise, our understanding is that violations of treaty obligations generally only give the signatories to the treaty the right to bring legal actions. the presumption is that there are no private rights of action. here, if the european authorities are happy with the actions of the individual state, there is not going to be any complaining; just the clinking of champagne glasses. if faced with a debt crisis situation and the need to protect against holdouts, italy will likely wish to retroactively add the two enhancements mentioned above-the option of using single-shot cacs and the disenfranchisement of the ecb-to all of its already-issued and outstanding local-law-governed debt. the legal barrier to making such a retroactive change, we believe, is non-trivial. the starting point in analysing this question has to be the recognition that most modern legal systems disfavour retroactive changes to contract rights. moreover, scepticism about the government's need to make such changes is likely to be especially heightened in contexts where the government is seeking to reduce its own debt obligations. to illustrate, let us say that the current euro cacs allow the issuer the following options: (a) if holders of per cent or more of an individual series of bonds (in principal amount) approve of a restructuring offer, that offer becomes binding on all holders. (b) in a cross-series modification, if holders of . per cent or more (in principal amount) approve of the restructuring offer, and per cent or more (in principal amount) of all the holders of the various series aggregated also approve of the offer, the offer is binding on all holders. to these existing options, let us now say that italy wishes to add a third and different option, using its local-law power. that option enhances the existing cacs by permitting, in the same fashion that is standard for bonds on the international (foreign-law) market, a single-shot restructuring option. (c) in a cross-series modification, if the restructuring offer satisfies the condition that the terms are uniformly applicable to all holders, and per cent or more of the holders of all the series approve the offer, it will be binding on all holders. let us say that italy wishes also to add in a provision to disenfranchise the ecb or any other supra-national institutions that might be mandatorily required to vote one way or the other on a restructuring plan without consideration of the merits of the situation. for bonds that were already issued, prior to the hypothetical italian debt crisis, with either no cacs or the first-generation euro cacs, these modifications will likely face legal challenge as a violation of property rights. both the italian constitution and european law provide protections for expropriations of property rights. article of the italian constitution protects against interference with property rights unless the action is for reasons of 'general interest' and the government makes 'provisions for compensation'. and article of the european convention on human rights is similarly protective with protections for the peaceful enjoyment of property except under conditions of 'public necessity'. in such situations, the interference with property rights needs to be proportional to the public need. the two questions to ask, therefore, are (i) whether the inclusion by italy of an additional option to the existing model euro cac and a provision disenfranchising the ecb, in order to better tackle a financial crisis, would be ruled as the kind of action that was in the general public interest and proportional to what was needed, and (ii) what compensation might be required, if there was a violation. the question of whether the retrofit inclusion of a single-shot cac would survive the kind of proportionality analysis that was done by courts in the aftermath of the greek restructuring has already been the subject of two excellent papers by manuelides ( ) and weidemaier ( ) . both suggest that a retrofit of a single-shot cac option would likely pass muster in the european courts. and although they do not analyse the this condition that the offer must be 'uniformly applicable' is basically a non-discrimination condition that was put in place in international bonds in when the single-shot cac restructuring option was added to the standard terms. question of adding in a provision that disenfranchises the ecb, we think that that would similarly be ruled to not be an undue interference with property rights. the key case that helps predict how the retrofit would fare is the european court of human rights' decision in mamatas and others v greece. others such as manuelides ( ) have parsed the implications of the various portions of the court's discussion and we will not repeat that analysis here. however, it is helpful to boil down the core of the court's analysis in the mamatas case and indeed a variety of other cases from the corporate context where corporations have put in place contractual modifications to help ameliorate a holdout problem. as we see it, the crucial question for the court will boil down to whether the sovereign has, in its use of the local-law advantage, acted opportunistically to take value away from the creditors and benefit itself. or, alternatively, whether it acted in the public interest in trying to fix flaws in the existing mechanism for restructurings so as to enable an orderly restructuring mechanism that benefits both the general public and the majority of creditors. assuming that the steps that italy takes are in the vein of correcting flaws in the existing restructuring mechanism because of the possibility of an imminent crisis and, moreover, that they are similar to those that the european authorities have indicated are in the works for implementation across europe anyway, it is hard to see how or why a court would rule the changes invalid. this is essentially what greece did with its cac retrofit in march . at the time of the greek restructuring in , european policy makers had already given the green light for a committee to design a cac suitable for inclusion in all euro area bonds. but that committee process took time, given that agreement from all the member states had to be obtained, and was not ready as of march , by which time greece was unable to wait any longer. therefore, greece's restructuring lawyers designed a cac that basically was an advance version of what would subsequently be designed at the european level. indeed, the experience with the use of the greek designed clause has been useful in informing discussions of what kind of clause was appropriate for euro-wide inclusion. more importantly for our purposes, it was important to the european court of human rights-in the context of its approval of greece's retrofit-that greece had retrofit the kind of provision that was standard. finally, there is the matter of damages. if it so happens that courts rule that italy has violated property rights protections, the question to ask is what the monetary damages would be. after all, if it turns out that the impact on the prices of the bonds from adding in features that improve their operation is an increase in the price of the bonds, then damages should be small (zero). and if damages are negligible, it should not matter if the courts find a technical legal violation. different legal systems analyse contract and expropriation damages differently. if viewed as a pure contract violation, the legal remedy in both civil and common law systems is what is called expectation damages; which is to give the aggrieved party the monetary amount that puts her back in the position she would have been had there not been a violation. and one way to measure that would be to look at the prices of italian sovereign bonds before and after the retrofit. based on the existing empirical work on the pricing impact of including cacs in sovereign bonds, it is safe to predict that the impact of the retrofit will either be to increase the market price of the bonds (after all, they are less subject to the cost of holdouts now) or to have no price impact at all. that is, damages would be zero. an alternative possibility would be for the court to ask what the value to the creditor would have been to be a holdout from the restructuring. we think it unlikely a court would go down this path, given how speculative the calculations would have to be. that is, unless the crisis has hit and the market begins pricing the holdout-friendly bonds at a premium to bonds that are easier to restructure. this may happen for two reasons. first, clauses that have to do with restructuring are likely to be more salient at a time of crisis and hence may be taken more seriously and even misinterpreted during these periods. in tranquil times, however, markets may end up barely noting such an innovation. secondly, if the sovereign takes action such as retrofitting cacs in the midst of a crisis, it is likely to be viewed as a signal that the sovereign's situation is worse than the market had anticipated. furthermore, there may be a price differential simply because turbulent times are characterized by high price volatility. assume that the price of a group of bonds happens to drop the day a retrofit is included in these bonds, a court may interpret this drop in price as linked to the inclusion of the retrofit even if this drop was purely driven by the high price volatility that characterizes turbulent periods. put simply, the greater the amount of daylight between when the retrofit is done and when the crisis hits, the smaller the likelihood of significant damages. the choice our article poses is between doing something and doing nothing. this choice can be better understood if we frame the problem in the same way as we would frame the decision of whether we should buy insurance to protect ourselves from a particular event that may or may not occur in the future. in this case, the uncertain event is the need to restructure one's sovereign debt. from the point of view of the borrowing country, the expected cost of a restructuring is p  c; where p is the probability that a debt restructuring will be needed and c is the political and economic cost of the restructuring. p and c, respectively, measure the probability and the cost of restructuring if the country does not do anything to protect itself. if the country buys insurance (ie, if it does something along the lines described above), it will pay an insurance premium s, it will need to restructure with probability ap, and, in case of restructuring, it will pay a cost c (with c <c, for the reasons explained above). hence, the expected costs if the country does something to protect itself will be: s þ apÂc . the insurance premium s captures the possible increase in the sovereign spread associated with doing something. this increase in spread could be due to either signalling or to moral hazard (or both). in the case of signalling, spreads may increase because, by doing something, the country reveals private information that signals that its debt situation is worse than what market participants thought. in the case of moral hazard, spreads may increase because the country now faces lower restructuring costs (remember c <c) and it is thus more likely to restructure its debt. moral hazard would hence be reflected in a higher probability of a debt restructuring: ap> p, where a captures the increase in the probability of restructuring associated with doing something. doing something will be convenient if the expected cost of implementing a reform is lower than the expected cost of doing nothing. that is, if: (s þ apÂc) < (pÂc). this condition can be written as: s < pðc À ac Þ theresa arnold, mitu gulati and ugo panizza • euro area sovereign debt in the era of we already discussed that c is likely to be substantially smaller than c. therefore, if the probability of a future restructuring in non-zero (ie, p > ), and if moral hazard is not very important (ie if a is not much larger than one) p(cÀac ), will be positive. but what do we know about s? the existing evidence suggests that s is likely to be very small, or even negative. for instance, multiple papers find that that the introduction of cac in european bonds was associated with a reduction (and not an increase) in yields: pointing to the idea that s could be negative (or, for sure, not positive and large). note that this is view also seems to be shared by the italian authorities. italy recently issued a dollar bond which included all sorts of creditor protections, but according to the director general for public debt of the italian treasury if the bond had been issued under local terms and priced in euros the yield would have been broadly in line with the current yield on italy's benchmark -year btp bonds. in fact, the spread over us treasuries of the italian dollar bonds ( basis points) was basically identical to the spread of italian btps over german bunds. taken together the evidence on bonds with cacs indicates that the market does not penalize bonds that are easier to restructure and anecdotal evidence on the recent italian bond issuance suggests that the market does not seem to reward bonds that have clauses that make them especially creditor friendly. this all suggests that s is likely to be very close to zero. but if s is close to zero, then the moral hazard problem (which contributes to s) is unlikely to be very important, making a close to one. hence, as long as the probability of restructuring is non-zero the condition that s< p(cÀac ) is likely to hold and the option of doing something now dominates that of doing nothing. in theory, that is. see giuseppe fonte, italy plans to issue more foreign currency bonds next year, reuters, october ( ) <https://www. cnbc.com/ / / /reuters-america-update- -italy-plans-to-issue-more-foreign-currency-bonds-next-year.html> the economics and law of sovereign debt and default we draw from mark weidemaier and yannis manuelides on this point. see mc weidemaier, restructuring italian (or other euro area) debt: do euro cacs constrain or expand the options road to nowhere: the legal and remedial futility of challenges to a restricted single-limb retrofit of local-law italian debt stock on private rights of actions under international treaties more generally, see paul b stephan for discussions, see buchheit and gulati restructuring sovereign debt under local law: are retrofit collective action clauses expropriatory' ( ) harv bus l rev this is one of the lessons of the famous gold clause cases of the s, where the us government used the local law advantage to abrogate the gold clauses in its bonds and in the bonds of private parties. see boudreau, ibid. for additional detail, see sebastian edwards, american default: the untold story of fdr, the supreme court and the battle over gold financial crises and constitutional compromise in the actual euro cacs, there are additional options and different voting thresholds and quorum requirements for when the votes are taken at a physical meeting of the holders versus in writing mitu gulati and ugo panizza • euro area sovereign debt in the era of covid- manuelides also analyses the important precedent of germany retrofitting cacs into its local corporate instruments some years prior. see manuelides (n ). numerous others have also analysed the implications of the litigation that followed the restructuring government debt under local law: the greek experience and implications for investor protection under european law the future of involuntary sovereign debt restructurings, mamatas and others v. greece and the protection of holdings of sovereign debt under the echr' ( ) cap mkts international arbitration and greek sovereign debt' ( ) oregon rev int'l l the greek bond haircut: public and private international law and european law limits to unilateral sovereign debt restructuring exit consents in sovereign bond exchanges for reflections on the design of the greek restructuring from its primary architect, see lee c buchheit, the greek debt restructuring of ', global restructuring rev mitu gulati and ugo panizza • euro area sovereign debt in the era of covid- the fact that the voting thresholds that greece saw in its retrofit cacs in were lower than those used in international bonds at the time did not trouble the courts (it had a simple % quorum and . % aggregate vote requirement). nor, for that matter, did the fact that greece used the single-shot cac that aggregated votes across all the series-a type of provision that is now standard we are assuming a scenario where the litigation gets resolved some years after the restructuring has been conducted. that puts the remedy of specific performance out of the realm of consideration since a court is unlikely to be willing this was the outcome in the gold clause case of the s (perry v united states) that involved a constitutional challenge to the us government changing the terms of its own bonds retroactively. see vanberg and gulati for a discussion of comparative contract law remedies, see, eg, luca ficetola, comparing remedies for contract violations in italian and english law but it is a possibility. see, eg, paolo colla et al pricing terms in sovereign debt contracts: a greek case study' ( ) cap mkts l hidden holdouts: contract arbitrageurs and the pricing of collective rights for the argument that whatever may make a debt restructuring process smoother is optimal ex post but not ex ante, see key: cord- -ny lj authors: vese, donato title: managing the pandemic: the italian strategy for fighting covid- and the challenge of sharing administrative powers date: - - journal: nan doi: . /err. . sha: doc_id: cord_uid: ny lj this article analyses the administrative measures and, more specifically, the administrative strategy implemented in the immediacy of the emergency by the italian government in order to determine whether it was effective in managing the covid- pandemic throughout the country. in analysing the administrative strategy, the article emphasises the role that the current system of constitutional separation of powers plays in emergency management and how this system can impact health risk assessment. an explanation of the risk management system in italian and european union (eu) law is provided and the following key legal issues are addressed: ( ) the notion and features of emergency risk regulation from a pandemic perspective, distinguishing between risk and emergency; ( ) the potential and limits of the precautionary principle in eu law; and ( ) the italian constitutional scenario with respect to the main provisions regulating central government, regional and local powers. specifically, this article argues that the administrative strategy for effectively implementing emergency risk regulation based on an adequate and correct risk assessment requires “power sharing” across the different levels of government with the participation of all of the institutional actors involved in the decision-making process: government, regions and local authorities. “and the flames of the tripods expired. and darkness and decay and the red death held illimitable dominion over all”. edgar allan poe, the mask of the red death, complete tales and poems (new york, vintage books ) p international concern" (pheic). in the light of its later levels of spread and severity worldwide, the who then assessed covid- as a "pandemic". the pandemic has spread rapidly in several european union (eu) member states. italy, however, is a special case: here, the covid- outbreak spiralled upwards earlier and more severely than elsewhere in europe, reaching a high mortality rate and creating the conditions for the public healthcare system's collapse. in this scenario, the italian government (from now on the government) declared a nationwide state of emergency, followed by increasingly restrictive measures aimed at slowing and containing the spread of the virus and mitigating the pandemic's effects under the by now well-known "flatten the curve" imperative. the last of these measures established the national lockdown, extending the emergency rules to the entire country for six months and, more generally, providing what has been called the "italian model to fight covid- ", namely "diminish viral contagions through quarantine; increase the capacity of medical facilities; and adopt social and financial recovery packages to address the pandemic-induced economic crisis". in this article, starting from the main regulatory acts and considering recent scientific knowledge and epidemiological data on covid- , we will examine the administrative measures the government has taken and the strategy it has implemented to deal with the pandemic in the immediacy of the emergency. after this initial analysis, we might legitimately wonder whether those measures and that strategy have proven effective in containing the pandemic. more generally, by analysing the administrative strategy, the article emphasises the role that the current system of constitutional separation of powers plays in emergency management and how this system can impact health risk assessment. an explanation of the risk-management system in italian and eu law will be provided and the following key legal issues will be analysed: ( ) the notion and features of emergency risk regulation from a pandemic perspective, distinguishing between risk and emergency; ( ) the potential and limits of the precautionary principle in eu law; who, "statement on the second meeting of the international health regulations ( ) emergency committee regarding the outbreak of novel coronavirus ( -ncov)", geneva, switzerland, january <https://www. who.int/news-room/detail/ - - -statement-on-the-second-meeting-of-the-international-health-regulations-( )emergency-committee-regarding-the-outbreak-of-novel-coronavirus-( -ncov)>. pheic has been defined in the international health regulations (ihr) of as an extraordinary event which can: ( ) constitute a public health risk to other states through the international spread of disease; and ( ) potentially require a coordinated international response. furthermore, this definition implies a situation that is: ( ) serious, unusual or unexpected; ( ) carries implications for public health beyond the affected state's national borders; and ( ) and may require immediate international action. who, "director-general's opening remarks at the media briefing on covid- ", march <https://www. who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid- - -march- >. resolution of the council of ministers of january , adopted pursuant to legislative decree / (civil protection code) <https://www.gazzettaufficiale.it/eli/id/ / / / a /sg>. on the declaration of emergency rule, see european commission for democracy through law (venice commission) <https://www.venice.coe.int/ webforms/documents/default.aspx?pdffile=cdl-std( ) -e>. dpcm of march <www.gazzettaufficiale.it/eli/gu/ / / / /sg/pdf>. for the general framework of all measures adopted by the italian state during the covid- emergency, see <https://www.camera.it/temiap/documentazione/temi/pdf/ .pdf?_ >. fg nicola, "exporting the italian model to fight covid- " (the regulatory review, april ) <https:// www.theregreview.org/ / / /nicola-exporting-italian-model-fight-covid- />. and ( ) the italian constitutional scenario with respect to the main provisions regulating central government, regional and local powers. specifically, the article argues that the administrative strategy for effectively implementing emergency risk regulation based on an adequate and correct risk assessment requires "power sharing" across the different levels of government with the participation of all of the institutional actors involved in the decision-making process: government, regions and local authorities. following the declaration of the state of emergency, the government approved decree-law no. of february vesting the president of the council of ministers with wide ordinance powers to handle the emergency by issuing his own administrative decrees. in particular, decree-law / gave the prime minister the power to issue typical emergency administrative measures in order to ensure social distancing, impose lockdown areas, close offices and public services and suspend economic activities. in addition, it allowed him to adopt atypical administrative powers whereby "further containment and emergency management measures" could be established. in a matter of days, the government approved three important regulatory acts based on the implementation of decree-law / : first with the decree of the president of the council of ministers (dpcm) of march , second with the dpcm of march and third with the dpcm of march, the government established stringent emergency administrative measures to curb the pandemic's spread throughout the country. in the first instance, these measures were gradual and concerned specific municipalities, provinces or regionsespecially in northern italythat were hardest hit by the virus and therefore classified as "red zones" subject to government-imposed local lockdowns. later on, the government established the national lockdown, and emergency measures were extended to the entire country for six months. in particular, pursuant to article ( ) of the dpcm of march , the government imposed a lockdown in lombardy and another fourteen provinces of northern italy. in doing so, the government introduced several legal prohibitions, such as the ban on people travelling to and from places in the red zones. with the subsequent national lockdown, the government imposed a travel ban in the entire country according to article ( ), dpcm of march , and prevented all forms of social gathering in public places or places open to the public across the country, according to article ( ), dpcm of march . furthermore, pursuant to articles ( ), ( ) and ( ), dpcm of march , retail businesses and personal services were suspended. as a consequence of the national lockdown, the ministry of health's order of march provided several stringent measures that prohibited many activities, such as the ban on accessing all public places, on exercising in public places and on going to holiday homes. in addition, with its order of march , the ministry of health, in agreement with the ministry of transport, established that people entering italy by plane, boat, rail or road must declare their reason for travel, the address where they plan to self-isolate, how they intend to travel there and their phone number so that authorities can contact them throughout an obligatory fourteen-day quarantine. moreover, several administrative sanctions were gradually established in the various regulatory acts. the last of these acts introduced rigorous sanctions for people who leave home without valid reasons and for undertakings that do not comply with the order to close. in the meantime, the regions and local authorities also adopted several ordinances establishing emergency administrative measures for the pandemic in their area. lastly, the government issued decree-law no. of march , with the aim of rationalising and coordinating emergency powers among the different levels of government. *** in the following pages, emphasising the role that the current structure of constitutional separation of powers plays in risk assessment, i will argue that the main problems of the italian administrative strategy for the covid- pandemic are due to the lack of effective "sharing of powers", and more specifically to the failure to share administrative in particular, art ( ) of decree-law / did not affect the effects produced and acts adopted on the basis of decrees and ordinances issued pursuant to decree-law / or art of law / , and established that the measures previously adopted by the dpcms of march , march , march and march as still in force on the date of entry into force of the said decree-law shall continue to apply within the original terms. regulatory powers among the different levels of government with the participation and cooperation of all institutional actors involved in the emergency decision-making process: the government, regions and local authorities. from this point of view, as i will attempt to explain, the failure to share administrative regulatory powers can have a decisive impact on risk assessment at the national level in terms of the effectiveness/ineffectiveness of the strategies adopted by the various institutional actors called upon to manage the emergency in their own areas. here, by "sharing powers", i mean the idea that the institutional actors involved in the decision-making process cooperate in the exercise of their powers by adopting consistent measures in the public interest; that is to say, with the aim of maximising the rights of individuals as required by the italian constitution. power sharing does not mean homologation. indeed, adopting different administrative strategies at different levels of government might increase the effectiveness of the response to a pandemic, but these measures must be shared among all of the actors involved in emergency management. sharing powers, measures and local strategies will be useful for an effective policy for containing the virus's nationwide spread based on an overall risk assessment. hence, the idea of shared powers emphasises the role of cooperation in specific institutional contexts, such as italy's, where competences are allocated across the different levels of government. the sense, more generally, is that sharing powers in multi-level systems enables states to perform better in terms of democracy, as powers are balanced between state and local levels. as we will see, however, the absence of effective power sharing at all levels of government in a pandemic can produce serious problems in correctly assessing risk and consequently in the emergency management strategy. in particular, i will discuss the problem of the lack of effective power sharing in italian policies from two key points of view: the government's administrative strategy in addressing the virus's spread by means of an "incremental approach" (section iv. .a); and the government's administrative strategy in implementing a national pandemic health plan (section iv. .b). before doing so, i will outline some key legal issues for the topics examined in this article. in particular, to put the administrative strategy devised by the government in the covid- emergency into context, i will analyse: ( ) the notion and features of emergency risk regulation from a pandemic perspective, distinguishing between risk and emergency; ( ) the potential and limits of the precautionary principle in eu law; and ( ) the italian constitutional scenario with respect to the main provisions governing government's, regions' and local authorities' powers. this preliminary analysis of key legal issues is useful for understanding why the administrative strategy has proven ineffective in managing the pandemic (sections iv. .a and iv. .b). placing the notion and its main features in the context of a pandemic, we could define emergency risk regulation as the action undertaken in the immediacy of a pandemic in order to mitigate its impact. from this perspective, we should bear in mind the distinction between risk and emergency. generally speaking, the traditional approach of administrative law refers to the notion of emergency and not also to the notion of risk, which legal doctrine touches on only marginally. with regards to the emergency, as a safeguard clause to deal flexibly with pandemic risks, governments and other public authorities may invoke the use of extraordinary powers to restore the normal course of legal relations. what is more, regulators have used emergency tools to act in the expectation of a risk for many years, although there is no denying that a risk is a potential danger, whereas an emergency is an actual danger. indeed, it should be sufficiently clear that emergency power is ineffective when applied in a situation that is only potentially dangerous. in this connection, it has been argued that the methods of exercising administrative powers can be better regulated by putting the administrative regulation in the category of risk rather than that of emergency. we might observe that if the notion of "risk" characterises a peculiar, intermediate state between security and destruction, in "emergency risk" the balance between these two clearly tilts towards the latter. in fact, as it is triggered by a pandemic, emergency risk regulation presupposes the existence, or the mere threat, of a pandemic. the pandemic, as a alemanno (ed.), governing disasters: the challenges of emergency risk regulation (cheltenham, edward elgar ) p xix. however, the notion of risk in italian administrative law is analysed by m simoncini, la regolazione del rischio e il sistema degli standard. elementi per una teoria dell'azione amministrativa attraverso i casi del terrorismo e dell'ambiente [risk regulation and the standards system. elements for a theory of administrative action through the cases of terrorism and the environment] (napoli, editoriale scientifica ) chs and , where the author postulating the notion of risk argues and suggests, in an innovative approach, the transition from the "emergency" perspective to the "risk regulation" perspective. . beck is responsible for analysing the sociopolitical dimension of risk management and in particular the problem of the relationship between science and society through the criticism of the monopoly that scientific rationality currently holds. alemanno, supra, note , xxii. a possible cause of disaster for humans, is an event of substantial extent causing significant physical damage or destruction, loss of life or drastic change to the natural environment. typically, one speaks of a pandemic when a threat to people's health is perceived that calls for urgent remedial action under conditions of uncertainty. fundamentally, emergency risk regulation in a pandemic event, as in other disasters, finds its natural regulatory space in two stages: mitigation and emergency response. in principle, mitigation efforts attempt to reduce the potential impact of a pandemic before it strikes, while a pandemic response tends to do so after the event. however, the distinction between emergency mitigation and emergency response is not always very sharp. when called upon to act under the menace of a pandemic, governments must both mitigate and respond to the threat in a situation characterised by suddenness (emergency) and significance. in a pandemic, emergency risk regulation is clearly called on to operate in the initial phase of the disease's spread, when the mere threat overshadows the regulatory context by virtue of its status as an emergency. accordingly, the most cost-effective strategies for increasing pandemic preparedness with administrative regulation, especially in resource-constrained settings, may consist of: ( ) investing to reinforce the main public health infrastructure; ( ) increasing situational awareness; and ( ) quickly containing further outbreaks that could extend the pandemic. in addition, especially once the pandemic has begun, a coordinated response should be implemented where the public regulator focuses on: ( ) maintaining situational awareness; ( ) public health messaging; ( ) reducing disease transmission; and ( ) care and treatment of the ill. successful contingency planning and an administrative strategy using the emergency risk regulation approach call for surge capacity, or in other words the ability to scale up the delivery of health interventions in proportion to the severity of the event, the pathogen and the population at risk. the pandemic may produce significant impact on the regulatory context by justifying the partial or total suspension of the ordinary decision-making process. departures from the rule of law, or simply from established procedures, are generally perceived as necessary if the event has met the significance threshold. however, the use of emergency administrative measures, such as temporary and exceptional measures, should be considered legitimate only for the period in which the pandemic ibid, xxii-xxiii. see also dd caron, "addressing catastrophes: conflicting images of solidarity and self interest" in dd caron and ch leben (eds), lasts. by contrast, prolonging exceptional order beyond the time of the pandemic means that any powers and measures designed to be temporary will be made permanent, intensifying the controlling authority's capacity, even though this might limit the enjoyment of individual rights. in addition, if the general need to prevent a pandemic cannot be ignored, it should be well thought out as an opportunity for risk regulation to prevent not only the sudden impact of a pandemic situation, but also any distorting effects or mishandling of the necessary recourse to emergency powers. consequently, it might now be inferred that emergency risk regulation in the context of a pandemic is a relevant regulatory methodology that combines the risk approach with the possibility of resorting to extraordinary measures in case a pandemic occurs. this methodology is essential for an effective administrative strategy for dealing with a pandemic because it permits constant monitoring and management of risks that can have serious consequences for society. by assessing the risks and taking proportionate measures, the negative effects of the emergency can be reduced and the use of emergency powers can be limited. indeed, it should be pointed out that the principle of reasonableness, which is generally invoked in the exercise of emergency powers against immediate danger, does not operate in emergency risk regulation. instead, as i will claim later, it will be the precautionary principle that matters (section iii. ). furthermore, it must be said that emergency risk regulation entails an accurate assessment of the factual situation based on scientific evidence. to apply this methodology correctly, a variety of factors must be consideredincluding the real level of the threat as well as how people perceive itin a step-by-step analysis based on the available scientific knowledge. in particular, as i will claim in analysing the italian policies (sections iv. .a and iv. .b), the administrative strategy for effectively implementing emergency risk regulation in a pandemic requires power sharing across the different levels of government with the participation of all of the institutional actors involved in the decision-making process in order to adopt consistent measures based on the constant monitoring and updating of the nationwide epidemiological risk assessment. hence, effective sharing of administrative powersand more specifically the administrative regulatory powers for emergenciesbetween the government, regions and local authorities would optimise the adoption of proportionate measures for controlling and containing the virus throughout the country, avoiding or at least delaying the application of stringent measures such as the lockdown of municipalities, provinces, regions or entire states. g martinico and m simoncini, "emergency and risk in comparative public law" (verfassungsblog, may ) <https://verfassungsblog.de/emergency-and-risk-in-comparative-public-law/>. according the authors, it is the facts and not the law that indicate the conclusion of an emergency. thus, the risks posed by the use of extraordinary administrative measures should be considered, especially at the end of the emergency when the government's powers should be subject to legal control in order to avoid departures from original objectives. in the same sense, see also simoncini, supra, note , . on the state of exception, see c schmitt, die diktatur: von den anfängen des modernen souveränitätsgedankens bis zum proletarischen klassenkampf (berlin, duncker & humblot ). schmitt's jurisprudential thinking placed the state of exception at the very centre of analysis, beginning with his work on the roman dictatorship. martinico and simoncini, supra, note . in managing the pandemic, the government's administrative strategy should take the emergency risk regulation methodology we have just outlined into account. in the eu legal system, the precautionary principle is described in article ( ) tfeu on environmental policy. the jurisprudence of the european court of justice (ecj) played a prominent role in elevating the precautionary principle to the status of a general principle of eu law. some ecj judgments in health matters are seminal in this regard. according to the ecj's jurisprudence, the precautionary principle requires that competent authorities adopt appropriate administrative measures to prevent specific potential health risks. the ecj's approach maintains that an appropriate application of the precautionary principle presupposes the identification of hypothetically harmful effects for health flowing from the contested administrative measure, combined with comprehensive assessment of the risks to health based on the most reliable scientific data available. in like manner, the european commission (ec) has contributed significantly to outlining the features of the precautionary principle in the eu legal system. in the communication of , the ec sought to establish a common understanding of the factors leading to recourse to the precautionary principle and its place in decisionmaking. according to the ec communication, the principle covers those circumstances where scientific evidence is insufficient, inconclusive or uncertain, but where preliminary scientific evaluation provides reasonable grounds for concern that the potentially dangerous effects on human health might be inconsistent with the chosen level of protection. various factors can trigger the adoption of precautionary measures. these factors inform the decision on whether to act or not, this being an eminently political decision, a function of the risk level that is "acceptable" to the society on which the risk is imposed. the ec has also established guidelines for those situations where action based on the precautionary principle is deemed necessary in order to manage risk. in these situations, a cost-benefit analysis to compare the likely positive and negative effects of the envisaged action and of inaction is recommended, and it should also include non-economic considerations. however, risk management in accordance with the precautionary principle should be proportionate, meaning that administrative measures should be proportional to the desired level of protection. in some cases, an administrative response that imposes a total ban may not be proportional to a potential risk; in others, it may be the only possible response. in any case, such measures should be reassessed in the light of recent scientific data and changed if necessary. in eu law, therefore, the precautionary principle has been widely recognised as a defining principle of risk regulation alongside the regulatory aim of a high level of protection. nevertheless, this principle might prove ineffective or even harmful if applied in a "strong" form. the strong form of the principle has been authoritatively criticised on the grounds that it suggests that regulation is required whenever there is a potential risk to health, even if the supporting evidence is conjectural and the economic costs of administrative regulation are high. in particular, if governments adopt the strong form of the principle, it would always require regulating activitiesconsequently imposing a burden of proof each timeeven if it cannot be demonstrated that those activities are likely to cause harms. in addition, as the need for selectivity of precautions is not simply an empirical fact but is a conceptual inevitability, no society can be highly precautionary with respect to all risks. hence, in this strong form, the precautionary principle proves ineffective and even harmful by requiring stringent administrative measures that can be paralysing, in that they prohibit regulation and all courses of action, including inaction. thus conceived, this principle may not lead in any direction or provide precise guidance for governments and regulators. recently, the limits of the precautionary principle have been analysed in the field of administrative and constitutional law. an interesting recent work proposes that precautionary and optimising constitutionalism are a dichotomy. in summary, the theory advances two distinct propositions. the first is that constitutions should be viewed as devices for regulating political risks. those political risks are referred to as "second-order risks", as opposed to "first-order risks" such as wars, diseases and other social ills. many of these risks are described as "fat-tail risks" that are exceedingly unlikely to materialise, but more likely than in a normal distribution, and are exceedingly damaging if they do materialise, as in the case of a pandemic. under "maximin constitutional" approaches, it is suggested that precautionary rules can overcompensate for these low-likelihood risks and even cause the very dangers that they seek to prevent. hence, precautionary constitutionalism is myopic in focusing on certain risks, and the notion of unappreciated or unaccommodated risks is central. on the basis of this hypothesis, the best way to regulate risk is thus to avoid obsessive views on risk avoidance or precautions and instead to allow greater flexibility in addressing the full array of risks inherent in government. what vermeule calls "optimising constitutionalism" is an answer to those who frame their understanding of the constitution along more rigid precautionary principles. vermeule's approach has been criticised. following these criticisms, i believe that this approach also reveals some critical points about the notion of risk. unless one adopts a more fungible notion of risk, i do not believe that "precautionary constitutionalism" is suboptimal for risk. it depends on how one weighs the risks involved in governing, even if one accepts risk analysis as the best measure for the success of a constitutional system. i claim, more generally, that correctly applying the precautionary principle, although it works better in a context of risk rather than one of emergency, is nonetheless important in managing a pandemic because it makes it possible to delay the implementation of stringent emergency measures. we have emphasised that administrative precautionary measures, unlike emergency ones, do not suspend the rule of law, since they activate soft government regulation that does not jeopardise fundamental rights concurrent with those threatened by imminent danger. hence, in my opinion, precautionary measures, where they are effectively shared across the different levels of government through appropriate risk assessment, would serve to avoid or at least delay governments' activation of a state of emergency. activating a state of emergency, consequently, would trigger hard government regulation through emergency measures that suspend the rule of law and therefore jeopardise fundamental rights. in a particular context such as the covid- pandemic, the precautionary principle could also be invokedand the implementation of precautionary administrative measures would be usefulin the presence of an emergency declaration issued by governments. in this sense, i argue that the declaration of a state of emergency for a pandemic is based on a technical risk assessment (ie technical discretion ) by the administration (eg government). in a pandemic, then, the emergency relates essentially to the capacity of administrations (eg governments, health authorities) to manage cases requiring healthcare (eg intensive care for respiratory support, hospitalisations for advanced pharmacological treatments and so on). thus, the subject of the technical assessment of the fact (the pandemic) is be provided by the evaluation relating to the administration's capacity to fulfil the tasks established by the legal system to protect the right to health enshrined in article of the italian constitution (section iv. ). furthermore, to be effective in emergencies such as a pandemic, the notion of the principle to which i refer should not entail the activation of precautionary measures typical of its strong version (which is exemplified in the well-known phrase "better safe than sorry"). in its strong version, in fact, the precautionary principle would be both paralysing and uneconomical, since it requires that any and all risks be prevented, even those that are least likely to occur or have been created artificially for italian legal doctrine distinguishes between "administrative discretion" and "technical discretion" under the influence of ms giannini, il potere discrezionale della pubblica amministrazione political reasons (i am thinking here of george w. bush's preventative war doctrine) in order to justify stringent administrative measures issued by governments for purposes not necessarily related to the alleged risk. by contrast, balancing costs against benefits might provide the basis of a principled approach for making decisions in complex contexts, such as the italian legal system, where the current constitutional separation of powers can lead to an inadequate and incorrect assessment of risks and therefore to ineffective emergency management by the different levels of government. in any case, scientific evidence is an essential prerequisite for better regulation by acting on the precautionary principle. to be cost effective, governments should take precautionary administrative measures based on scientific knowledge and thus carefully assess the risks they intend to manage. taking the potential and limits of the precautionary principle from the perspective we have outlined above into account might have an impact on governments' ability to deal effectively with pandemic emergencies. this matters in the case of italy, where the current structure of the constitutional separation of powers between the government, regions and autonomous local authorities plays a crucial role in effectively managing the pandemic emergency. analysing the italian constitutional scenario can provide substantial guidance for understanding the legal structure of powers and competences of government, regions and local authorities and explain why assessing pandemic risk can be impacted by a given separation of powers. such an analysis can shed light on the administrative strategy implemented by the government in the pandemic and enable us to evaluate its effectiveness in managing covid- across the country. first of all, we should bear in mind that the italian constitution (from now on the constitution) does not explicitly refer to emergency power, except for a state of war (article ). however, this power has traditionally been included in the typical powers that the constitution assigns to the government. in the constitutional system, the main rules governing the government's powers are established by articles and . indeed, parliament does not have a monopoly on legislative power, and the government may also issue laws by two legal instruments that should be understood as extraordinary: legislative decree and decree-law. in particular, article allows parliament to delegate its legislative power to the government, which in turn is given the power to issue legislative decrees. hence, the legislative decree is a form of delegated law-making power, where parliament may pass an enabling act entrusting the government to adopt one or more acts that have legal force. generally, the legislative decree is a legislative tool that is often deployed in all matters where a strong technical content is present. the second extraordinary instrument, the decree-law, is provided for by article . this is a form of law-making through emergency powers that the government may exercise in "exceptional cases of necessity and urgency" and under "its own responsibility". the government can thus issuewithout an enabling act from parliament as required by the provisions of article administrative measures that have the force of ordinary laws. however, such administrative measures will lose their effects as of the date of issue if parliament does not transpose them into an ordinary law within sixty days of their publication. with the major reform on "administrative federalism" enacted by law no. of october , which amended title v of the constitution, italy rapidly devolved legislative and regulatory powers to the regions. fundamentally, the constitutional amendment provided a new framework for the distribution of powers and competences between the national and local levels. it established a new institutional structure by dividing legislative and administrative competences and powers across the different levels of government. the amended articles of the constitution are the basis for the fundamental reform of administrative federalism. article recognises local authorities (municipalities, provinces, metropolitan cities) and regions as autonomous entities of the state with their own statutes, powers and functions in accordance with the principles laid down in the constitution. article establishes the role and legislative powers of the state and regions, indicating those matters for which the state has exclusive legislative power and those for which concurrent legislation of both the state and the regions is possible. the regions have exclusive power in all matters not expressly covered by state law. municipalities, provinces and metropolitan cities also have regulatory powers for the organisation and implementation of the functions attributed to them. specifically, article ( ) establishes that the state and regions have concurrent power, and the regions have regulatory powers, in matters of public health. in this connection, at the national level, parliament and government are called upon to: ( ) adopt fundamental health principles by means of framework laws and guidelines; and ( ) establish essential levels of healthcare. at the regional level, the regions implement: ( ) general legislative and administrative activity; ( ) the organisation of health facilities and services; and ( ) the provision of healthcare based on specific local needs. article provides for the subsidiarity principle, according to which all functions are exerted by municipalities, while the possibility remains to confer them to higher levels of government in order to guarantee the uniform implementation of spending functions across the country. article guarantees national unity and the unitary nature of the constitutional system by providing for the government's substitution power. according to article ( ), the government can act for the regions and other local authorities if: ( ) the latter fail to comply with international rules and treaties or eu legislation; ( ) in the case of grave danger for public safety and security; or ( ) whenever such action is necessary to preserve legal or economic unity and in particular to guarantee the basic level of benefits relating to civil and social entitlements, regardless of the geographical borders of local authorities. to this end, the law shall lay down the procedures to ensure that ( ) subsidiary powers (ie the government's substitution power) are exercised in compliance with the principles of "subsidiarity" and "loyal cooperation". lastly, with regards to powers and competences in emergencies, it should be noted that in the italian legal system several authorities can introduce specific regulatory acts establishing administrative measures needed to deal with emergencies in accordance with the constitution. the power of ordinance has a particular role in managing emergencies, as it can be exercised in situations of necessity and urgency. in particular, the legal system provides for: ( ) as we will see, the structure of power just described highlights the problem of risk assessment among the institutional actors involved in the administrative decisionmaking process. though the current system of allocation of powers and competences to the regions and other local authorities might be an advantage in terms of correctly assessing and managing risk in their areas, at the national level, this system requires an effective sharing of powers and strategies between the centre and the periphery, where the measures of the regions and local authorities must be adopted in accordance with the measures advanced by the government, and vice versa. since correct risk assessment by an authority must take the characteristics of its area into accountdata on the epidemiological situation, for example, or on the average age of the legal nature of the "state's substitution power" in italian legal doctrine has been extensively discussed. in particular, some scholars argue that art provides a form of "administrative" substitution of the state over the regions, and that art ( ) concerns "legislative" substitution. other scholars agree on the idea that art provides the genus of substitution powers, whereas art ( ) refers to one species of the genus, being a mere specification of art . however, the constitution seems clear on this point. as we have seen, the provisions of art speak of the "government", while the provisions of art ( ) speak of the "state". the population, and the capacity of the health system with regards especially to the availability of intensive care bedsit might be assumed that in the italian legal system's effective risk assessment could be facilitated by the specific competences established by the constitution for the regions and other local authorities in health matters. however, as i will argue, this is a theoretical advantage that works only if power is effectively shared between the different levels of government. in fact, in order to provide an adequate and correct risk assessment at the national level and take effective measures to contain and manage the pandemic, the current system needs powers and strategies to be shared between local authorities, regions and the government. sharing administrative powers at all levels of government is an important part of the task of states. indeed, enhancing multi-level regulatory governance has become a priority in many eu states. for this reason, the eu supports sharing of administrative regulatory powers by encouraging better regulation at all levels of government, calling on the member states to improve coordination and avoid overlapping responsibilities among regulatory authorities. in italy, until the adoption of constitutional law / , regulatory reform had been promoted, designed and implemented mainly at the national level. with the reform, as we have seen (section iii. ), such a centralised approach lost legal and political ground. at the same time, responsibilities for developing and implementing administrative regulation policies have not been explicitly allocated to either the state, the regions or the local authorities. hence, the responsibility for administrative regulation and regulatory reform lies with each of the levels of government in the matters where they exert legislative powers. in like manner, there is no overall competence at the central level to monitor and control regulatory reform programmes at the local level. accordingly, the new constitutional structure calls for effective sharing of administrative powers across the different levels of government. on the basis of the analysis carried out so far, i will now argue that the main problems of the italian administrative strategy for the covid- pandemic are due to the lack of effective sharing of administrative powers and, more specifically, to the failure to share regulatory powers across the different levels of government with the participation and cooperation of all institutional actors involved in the emergency decision-making process: the government, regions and local authorities. in particular, this problem oecd, "the territorial impact of covid- : managing the crisis across levels of government" (last updated june ) <https://read.oecd-ilibrary.org/view/?ref= _ - agkkojaaa&title=the-territorial-impact-of-covid- managing-the-crisis-across-levels-of-government>. the european committee of the regions (cor), "division of powers between the european union, the member states and regional and local authorities" (december ) <https://cor.europa.eu/en/engage/studies/documents/ study-division-powers/study-division-powers-en.pdf>. see also, oecd-puma, "managing across levels of government" ( ) <http://www.oecd.org/governance/budgeting/ .pdf>. has impacted the risk assessment of the various authorities called upon to manage the health emergency. as a result, the problem has impacted nationwide risk assessment and, consequently, the management of the emergency at the national level, leading to the adoption of inconsistent measures by the various institutional actors involved in the administrative decision-making process. in particular, i discuss this problem in italian policies from two key points of view: the government's administrative strategy for managing the virus's spread by means of the "incremental approach" (section iv. .a) and the government's administrative strategy for implementing the nationwide pandemic health plan (section iv. .b). in doing so, i shall take into account the considerations presented above concerning emergency risk regulation (section iii. ), the precautionary principle (section iii. ) and the rules governing powers in the constitutional scenario (section iii. ). one of italy's main problems in relation to the ineffective sharing of administrative powers for managing the pandemic is clearly displayed in what i will call the "incremental approach". this approach is essentially based on the "progressive" application of emergency measures by the government in order to manage the "exponential" spread of the virus. the italian administrative strategy for the pandemic is fundamentally founded on such an approach. in fact, as we have seen (section ii), the government addressed the pandemic by enacting several decrees (dpcms) that "progressively increased" restrictions in lockdown areas (red zones), which were then extended from time to time until they finally applied to the entire country in the national lockdown. in my opinion, although the incremental approach may be a correct application of the principle of proportionality, given the government's proportionate use of emergency powers in dealing with the pandemic, it is the result of an ineffective sharing of administrative regulatory powers between the government, regions and local authorities. indeed, the progressive enforcement of lockdown areas, which from time to time increased the extent and severity of the emergency measures, demonstrates the difficulty of governing the spread of the virus in the red zones rather than the effective implementation of a proportionate administrative strategy. and this is mainly due to the lack of effective cooperation between the government and the regions in exercising their respective emergency powers. from a general point of view, the incremental approach reveals the limited effectiveness of the national and local measures and strategies for managing and containing the pandemic when those measures and strategies are not shared. i argue that even the stringent national lockdown is essentially the result of the ineffective sharing and planning of administrative measures and strategies for managing the pandemic across the different levels of government and especially, in on this approach, see g pisano, r sadum and m zanini, "lessons from italy's response to coronavirus" (harvard business law review, march ) <https://hbr.org/ / /lessons-from-italys-response-to-coronavirus#comment-section>. dpcm of march <www.gazzettaufficiale.it/eli/gu/ / / / /sg/pdf>. this case, between the government and the regions. one can legitimately wonder whether the government can adopt an effective administrative strategy for managing the emergency without sharing and planning their measures with those of the regions. from this perspective, we can say that the government's incremental approach has proven ineffective in coping with the pandemic. i will now explain why in the following points. ( ) regarding risk assessment for pandemics, the science shows that the spread of covid- is rapid and exponential. consequently, the incremental approach does not work if it is not properly implemented with the effective participation of all institutional actors involved in managing the pandemic. scientific data and statistics on the spread of the virus were not predictive of what the situation would have been in the short and medium term. hence, a correct risk assessment of the virus's nationwide spread would have suggested that the administrative measures and, more generally, the strategies should have been shared among all players involved in the main strategy. very often, however, the government's strategy has not been in line with those of the regions, revealing an inadequate assessment of the risk that the virus would spread throughout the country, and thus the ineffective sharing of emergency powers. in fact, some important emergency measures implemented by the regions clearly contradict the government's main strategy. to take a few examples, marche region ordinance no. of february , issued pursuant to decree-law no. of , established measures that were more stringent than the government's, disregarding the latter's strategy. for this reason, the government contested the order before the court. although a judgment in favour of the government was handed down and the challenged ordinance was suspended, the marche region legitimately adopted a new ordinance establishing emergency measures based on the same decree-law no. / , once again disregarding the government's strategy. another paradigmatic case is provided by a series of ordinances by the campania region aimed at imposing a more stringent lockdown at the local level than the lockdown established by the government at the national level. unlike the marche case, the ordinances of the campania region, although contested before the administrative judge, were not suspended, thus making the government's strategy ineffective. consequently, in the absence of effective sharing and planning of the main strategy with the regions, the government had to 'increase' the emergency measures from time to time until finally imposing the stringent national lockdown. ( ) in the absence of power sharing and strategies based on correct risk assessment at the national level, the government's incremental approach seems to have played a considerable role in people's behaviour, inducing them to make "bad choices". as the data show, the government's incremental lockdown of municipalities, provinces and regions in northern italy induced masses of people to move towards the southern regions, spreading the virus to parts of italy that had not yet been affected. an emblematic case of this kind took place immediately after the dpcm of march (see section ii) locked down lombardy and another fourteen provinces in northern italy, spurring thousands of people to flee to the south. such potential negative externalities, as well as other negative spill-overs or distortions, should have suggested that the government share its regulatory acts with those of the "target" regions (ie the northern regions), as well as with the other regions that could be indirectly jeopardised by the lockdown measures (ie the southern regions). alternatively, the government should have undertaken to coordinate the strategies of the regions and local authorities in order to enhance the adoption of effective control measures for people exiting the red zones and entering less affected regions. more generally, in applying lockdown measures, the government should have shared and planned its strategy with the regions on the basis of a common risk assessment that took into account not only the regional territories, but the entire country. accordingly, the government should have established effective countermeasures together with all of the regions potentially involved in lockdown decisions to prevent the virus from spreading from high-risk to low-risk areas. an effective emergency response must be coordinated as a consistent system of actions taken simultaneously by the different actors involved in the decision-making process. ( ) the government's incremental approach also revealed the problem of effectively sharing and planning precautionary measures (see section iii. ) across the different levels of government. the critical situation that arose because of the epidemic's severity called for effective testing of symptomatic and asymptomatic cases, as well as proactive tracing of potential positives across the country. on this point, these precautionary measures were supported by scientific data on the transmission of covid- by asymptomatic people. the absence of a shared strategy for the adoption and implementation of precautionary measures proved particularly harmful in regions where the epidemic risk is higher. indeed, it is no coincidence that the outbreak spread so quickly in northern italy and especially in lombardy. in this region, the efficient public rail transport network connecting urban areas, large numbers of commuters and high levels of air pollution are thought to have increased the incidence of infection. from this point of view, it is clear that risk assessment has been inadequate, and strategies have thus been ineffectively shared between lombardy and the government. the government should have promoted an effective precautionary strategy for health checks by sharing it with the strategies of the regions and ensuring efficient nationwide implementation on the basis of a global risk assessment. conversely, data on infections and deaths reveal that strategies were not shared effectively with the hardest-hit regions. ( ) the incremental approach shows that most of the problems of administrative strategy are also motivated by political issues between parties governing regions and belonging to the coalition now governing the country. from the time when the virus began to spread, the multi-level management of the emergency has triggered competition and institutional division between the government and regions due to policymakers' political differences. the management of the pandemic, in fact, has thrown light on the deep political division between the government, led by the coalition of left-wing parties such as the democratic party and the five star movement, and the hardest-hit regions -lombardy and venetoled by traditionally right-wing populist parties such as the league and brothers of italy. in particular, many of the administrative measures taken by the regions were in contrast with the government's strategy, largely for political reasons. from this standpoint, it can be seen that there has been an "institutional clash" between the regional governments and the national government on the political and administrative actions to be taken to effectively manage the emergency. it is no coincidence that the government's minister of health is a member of one of the opposition parties in lombardy and veneto, and that the governors of lombardy and veneto belong to the coalition opposing the government. to give a few specific examples, a bitter dispute has occurred between prime minister giuseppe conte and attilio fontana, governor of lombardy and member of the rightwing populist party league, with regards to the ineffective management of the emergency in the region most affected by the virus. similarly, as we have seen, luca ceriscioli, governor of the marche region and member of the centre-left party in the majority coalition, opposed the government's decision to declare a state of emergency only in the northern regions. in essence, these strong political divisions have impacted effective power sharing among the different levels of government, causing problems for the government's incremental administrative strategy. ( ) the incremental approach also shows the important role that scientific competence plays in emergency management. in this regard, one of the main goals of scientific expertise is to inform and legitimise governments' decisions, especially in high-uncertainty situations relating to public health. during the covid- outbreak, scientific and technical experts have assisted central and regional governments by contributing to the content of decisions and, more generally, of administrative emergency management strategies. as scientific evidence is the basis for sound political choices, scientific and technical experts have become part of the rationale of governments' decisions and have been useful in reassuring the public with concrete solutions. indeed, in the immediacy of a pandemic, as is logical to assume, the demand for scientific expertise increases as governments search for certainty in understanding problems and choosing effective measures for managing the emergency. especially in the most delicate phases of an emergency, scientific expertise is useful in informing, legitimising and justifying government evaluations and responses to problems, even as political and administrative considerations continue to govern such choices. the result is an increased reliance on scientific expertise and politicisation of scientific and technical information. by invoking scientific expertise, policymakers create the need for what is perceived as evidence-based policymaking, which suggests to the public that political and administrative decisions are based on reasoned and informed judgments aimed at ensuring the public interest and guaranteeing individual rights. however, a major problem is that scientific expertise might obscure the accountability of decisions. as scientific and technical experts serve to inform and legitimise political and administrative decisions, they may also obscure responsibility for policy responses and outcomes. scientific expertise helps to establish the severity of a pandemic in a population, to understand the epidemiological trend over time and to evaluate the effects of political and administrative measures, from mitigation to suppression. nonetheless, undertaking policy actions is the responsibility of government leaders. as scientific expertise becomes more prominent in the policy process, who is accountable for policymaking becomes more obscure. to work better in emergencies, scientific expertise also requires effective sharing of administrative powers based on accurate risk assessment, as i will now explain. in italy, since the beginning of the virus's spread, the various institutional actors, especially the government and the regions, have established their own scientific task forces to support administrative measures and strategies in managing the pandemic. the main problem is that, by doing so, risk assessment at the national level is fragmented. conflicts can also arise between institutional actors involved in the decision-making process. in this scenario, indeed, the government and the regions have adopted administrative decisions and strategies based on the risk assessments provided by their own central and regional task forces. it should be noted that this situation, like others discussed here, derives from the current constitutional architecture of separation of powers where the decision-making process is assigned to the different levels of government. however, managing a pandemic requires a comprehensive risk assessment. italian policies matter, as they show how, at the beginning of the pandemic, some regions' task forces underestimated covid- , while other regions gave it a certain importance. this behaviour on the part of policymakers was not led by the government, which, on the contrary, criticised the regional governments' solutions. the outcome, as i claimed for the incremental approach, is that the government's measures and strategies are not shared with those of the regions and vice versa, and policymakers' accountability is obscured by invoking scientific expertise for pandemic management decisions. b. implementing the national pandemic health plan there is no doubt that a pandemic affects the whole of society. no single organisation can effectively prepare for a pandemic in isolation, and uncoordinated preparedness of interdependent public organizations will reduce the ability of the health sector to respond. a comprehensive, shared, coordinated, whole-of-government approach to pandemic preparedness is required. the government's strategy, as we have seen in the incremental approach to dealing with the emergency, proved particularly ineffective due to the failure to share administrative powers with the other institutional actors involved in the pandemic decision-making process, particularly the regions. but this, as we shall see now, was not the only weak point. i will argue here that another of the major problems was the lack of effective implementation of the national pandemic health plan. in particular, we will see how and why the ineffective implementation of the plan by the government, regions and local authorities posed serious problems for containing the spread of the virus and, more specifically, for avoiding the collapse of the public healthcare system. on this point, one of the main problems for public health posed by the novel coronavirus is its ability to spread with exceptional ease and speed, threatening to overwhelm the healthcare system. in particular, what should be especially clear from the data is the critical situation of the intensive care system in italy, which has been severely weakened by the pandemic. intensive care system at the national level, cooperating with the regions and local authorities to ensure that critical care bed availability is efficiently managed. in this case, effective actions shared among all institutional actors and based on an adequate and accurate risk assessment at the national level would avoid saturating the intensive care system in the medium and long term, while the government should be able to increase capacity in the short term. yet, the data on the intensive care system show that the situation was inefficiently managed in the regions hardest hit by covid- , especially in lombardy, which paid a high price at the local level for the ineffective implementation of the pandemic health plan at the national level. more generally, it should be emphasised that this point also demonstrates the importance of sharing administrative powers between government, regions and local authorities to implement the pandemic management plan effectively throughout the country. in this connection, many elements based on scientific and epidemiological data demonstrate that the covid- pandemic called for effective cooperation and coordination across all levels of government. in addition, it must be borne in mind that fighting a pandemic hinges on many factors, most of which are time consuming or in any case cannot be accomplished quickly. preparing a candidate vaccine, for example, takes a long time in terms of both preclinical and clinical development. likewise, developing and testing an effective drug involves complex multi-stage clinical trials. such considerations might be sufficient on their own to justify taking effective actions to mitigate the pandemic emergency's impact on the public healthcare system. in this phase, as we have seen, emergency risk regulation requires that regulatory action be taken in the immediacy of an emergency in order to mitigate its impact (section iii. ). to avoid the collapse of the public health system, the government should thus have contained the spread of the virus by effectively implementing the nationwide pandemic management plan with the participation of all institutional actors. the who has recognised the importance of sharing administrative powers through the participation and cooperation of the various institutional actors involved in the strategy against pandemics. in this regard, the who has drawn up specific guidelines for implementing a pandemic influenza preparedness plan that states should apply in order to manage the spread of the virus throughout their territories. in particular, the who's guidelines encourage states to develop efficient plans, based on national risk assessments, with the effective participation of institutional actors at all levels of government. in italy, the most serious problem is that the government, although it had already developed its own national plan, foster its effective adoption by the regions and local authorities, disregarding a crucial point of the who's guidelines. consequently, the failure to implement the national pandemic plan, as we have seen, created the conditions for the collapse of the public health system, with the overcrowding of intensive care units and the consequent loss of life. *** in conclusion, the italian policies regarding the covid- outbreak can demonstrate the importance of: ( ) rethinking the incremental approach; and ( ) implementing a national health plan for pandemics by sharing powers, and more specifically administrative regulatory powers for emergencies based on an adequate and accurate risk assessment at the national level, among the different levels of government with the participation, cooperation and coordination of all institutional actors involved in the pandemic decision-making process. as we have seen, sharing administrative powers at the different levels of government plays a particularly important role in managing emergencies in the constitutional scenario, where competences are distributed between government, regions and local authorities, and several institutional actors are allowed to adopt regulatory acts (see section iii. ). the major changes that the constitutional amendments have brought to policymaking in the italian legal system require that constant support be provided to the regions and local authorities, especially in emergencies. despite significant decentralisation, the government still has a fundamental role to play in sharing and coordinating administrative powers at the different levels of government and in ensuring loyal cooperation among all of the institutional actors involved in emergency decisionmaking processes. indeed, the government is tasked with promoting and coordinating "action with the regions" (article of law / ), as well as with advancing cooperation "between the state, regions and local authorities" (article of legislative decree / ). similarly, the government must promote "the necessary actions for the development of relations between the state, regions and local authorities" and ensure the "consistent and coordinated exercise of the powers and remedies provided for cases of inaction and negligence" (article of legislative decree / ). looking at the constitutional perspective, some possible solutions might be proposed. ( ) in the italian constitutional scenario, although concurrent power to legislate on matters of public health is vested in the state (ie the government) and the regions pursuant to article ( ), the state (ie the government and the regions together), on the basis of the principle established by article ( ), "safeguards health as a fundamental right of the individual and as a collective interest". i argue, more specifically, that safeguarding health is a task of the state based on the fundamental principle of the constitution referred to in article ( ) , where the duty of the state is to "remove those obstacles of an economic or social nature" that, by constraining legislative decree / <http://www.parlamento.it/parlam/leggi/deleghe/ dl.htm>. the "freedom and equality of citizens", impede the "full development of the human person and the effective participation of all workers in the political, economic, and social organisation of the country". thus, i believe that under the joint interpretation of article ( ) and article of the constitution, as well as the principle of loyal cooperation, the government and the regions must act by sharing administrative powers (and strategies) among them in order to protect the fundamental right to health. in so doing, the government can play an essential role in promoting institutional balance and cooperation between the national and local levels, maximising loyal cooperation and implementing vertical and horizontal subsidiarity. ( ) sharing administrative powers for emergencies can also be encouraged and enhanced through the effective implementation of constitutional tools, such as the system of conferences based on the principle of loyal cooperation. (a) the conference on the relationships between government, the regions and the self-governing provinces is the key legal tool for multi-level political negotiation and collaboration. it serves in an advisory, normative and planning capacity and acts as a platform facilitating power sharing. (b) the conference on the relationships between government and the municipalities coordinates relations between the government and local authorities through studies, information and discussion of issues affecting local authorities. (c) the permanent conference on the relationships between government, the regions and the municipalities deals with areas of shared competence. ( ) in order to "safeguard health as a fundamental right of the individual and as a collective interest", article ( ) of the constitution could be applied whenever it is necessary to guarantee "the national unity and the unitary nature of the constitutional system". i claim that this provision, which establishes the government's administrative substitution power, provides for the centralisation of administrative powers in specific cases contemplated by the constitution. in this sense, article ( ) lays down that the government can act for the regions and/or local authorities in cases of "grave danger for public safety and security". in the light of this definition, the government's substitution for the regions and/or local authorities might be invoked as a result of the "grave danger for public safety", as well as in order to preserve "economic unity" and guarantee the "basic level of benefits relating to civil and social entitlements". in my view, however, the government should exercise its power of substitution as an extrema ratio whenever effective sharing among all of the institutional actors has not been implemented. article ( ) is clear in this regard, requiring that the substitution power be exercised in compliance with the principles of "subsidiarity" and "loyal cooperation". italy's national pandemic plan was adopted through the permanent conference on the relationships between central government, the regions, municipalities and other local authorities <http://www.regioni.it/sanita/ / / /accordo-stato-regioni- - - -su-pandemia-influenzale- />. administrative powers"and more specifically the administrative regulatory powers for emergenciesbased on an adequate and accurate risk assessment, across the different levels of government with the participation, cooperation and coordination of all institutional actors involved in the emergency decision-making process: the government, regions and local authorities. fundamentally, i emphasised that the italian case reveals the importance of sharing administrative powers from two main points of view. first, i argued that the "incremental approach" to dealing with the emergency, although based on the proportionate use of powers, is largely ineffective or even harmful in the absence of cooperation among all actorsthe regions and local authoritiesinvolved in the main strategy implemented by the government (section iv. .a) . second, i discussed the importance of cooperation between the government, regions and local authorities for the effective and efficient implementation of a nationwide pandemic health plan (section iv. .b). i suggested that these points be viewed from a constitutional perspective in order to propose some possible solutions. from this perspective, the problems of effective sharing of administrative powers across the different levels of government could be resolved by systematically interpreting the constitution and implementing specific constitutional tools provided by the legal system (section iv. ). in conclusion, more generally, i argue thatand this is the main thrust of the articleadministrative powers should be shared across the different levels of government based on an adequate and accurate risk assessment with the participation and cooperation of all of the institutional actors involved in the emergency decision-making process in order to safeguard the fundamental rights enshrined in the constitution as well as in eu and international law. in pandemics, this aim must be achieved not only to guarantee the right to health, but also to safeguard all of the rights that might be jeopardised by the exercise of administrative powers and, more specifically, the exercise of emergency powers in dealing with the pandemic. the strong measure of "lockdown", for example, should be the extrema ratio of administrative powers because it suspends the rule of law and jeopardises rights. indeed, as i have claimed in analysing the italian policies, sharing powers with effective cooperation between government, regions and local authorities in managing the pandemic would optimise the adoption of nationwide virus containment measures, avoiding or at least delaying the application of stringent emergency measures such as the lockdown of municipalities, provinces, regions or even the entire country. taking into consideration the correct application of emergency risk regulation (section iii. ) and the precautionary principle (section iii. ), although lockdowns aim to contain specific areas that are most affected by the virus, they must be proportional to the risk that they intend to curtail. when such measures are adopted to protect the right to health, as is the case in a pandemic, this right must be balanced with other rights. yet, if administrative powers are not shared effectively across the different levels of government, the balancing principle might be disregarded by jeopardising one or more rights without legitimate justification (eg the right to freedom of movement enshrined in article of the constitution). this is the problem that the italian policies bring to light: a problem that i believe that the government must take into account in the near future as it strives to manage covid- and other similar pandemics. perspectives on the precautionary principle les avatars du principe de precaution en droit public le principe de précaution en droit communautaire: stratégie de gestion des risques ou risque d'atteinte au marché intérieur? the legal origins of the precautionary principle are to be found in the vorsorgeprinzip established by german environmental legislation in the mid- s; see there is a close relationship between the two principles that has led some to argue that they may be used "interchangeably". however, other authors contend that the prevention principle applies in situations where the relevant risk is "quantifiable" or "known" and there is a certainty that damage will occur. in this sense, see, respectively, wt douma principio di prevenzione e novità normative in materia di rifiuti dal pericolo al rischio: l'anticipazione dell'intervento pubblico" [from danger to risk: the anticipation of public intervention] ( ) diritto amministravio . ecj case t- / pfizer animal health sa v council in the same sense, see also case c- / national farmers' union case c- / united kingdom v. commission [ ], ecr i- case c- / monsanto agricoltura italia art of the italian constitution <https italian constitutional justice in global context il sistema amministrativo dopo la riforma del titolo v della costituzione amministrazione e costituzione sussidiarietà e diritto alla salute substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (sars-cov ) a novel coronavirus emerging in china, key questions for impact assessment mathematical biosciences and engineering - . for daily reports around the world, cf. who of the marche regional administrative court suspended the marche region ordinance issued on the campania regional administrative court in particular, i maintain that this power would have been useful as a remedy for inaction or delay on the part of the regions and local authorities in implementing the national pandemic health plan or adopting emergency measures. in such cases, substitution powers may be granted to the central level of government (the appropriate ministry or the council of ministers), which intervenes instead of the local level (the regions and/or local authorities).the same considerations on the government's administrative substitution power can be applied to the following solutions as "ex post remedies" for the failure to share administrative powers effectively across the different levels of government.( ) the government has extraordinary administrative power to annul illegitimate measures (eg ordinances) issued by local authorities in order to "safeguard the unity of the national legal system" (article of law / and article of law / ). the power, generally applied to measures issued by the municipalities, can be extended to regional measures on the basis of article ( ) of the constitution. ( ) the government has the right of action to challenge illegitimate measures issued by the regions and other local authorities before the court, in accordance with the constitutional provisions governing judicial review in the italian administrative justice system (articles , , , , and of the constitution). in this article, starting from the main regulatory acts and considering scientific knowledge and epidemiological data on covid- , i have analysed the administrative measures and strategies for fighting the pandemic implemented by the government in the immediacy of the emergency.specifically, i carried out this analysis to verify this strategy's effectiveness at managing the pandemic throughout the country.in analysing the administrative strategy, i emphasised the role that italy's current system of constitutional separation of powers plays in emergency management and how this system can impact health risk assessment.to do so, i first illustrated the risk management system in italian and eu law and discussed some key legal issues that are useful for clarifying the main points of the matter and for understanding the weaknesses of the administrative strategy with a view to proposing solutions that the government might take into account in adopting its strategy.in particular, i briefly examined the notion and features of emergency risk regulation (section iii. ), the potential and limits of the precautionary principle (section iii. ) and the rules governing powers and competences in the italian constitutional scenario (section iii. ).to summarise the main points of this analysis, i stressed that the italian policies for managing the pandemic demonstrate the importance of effective "sharing of key: cord- - wf yhyz authors: d'aloja, ernesto; finco, gabriele; demontis, roberto; napoli, pietro emanuele; fossarello, maurizio; nioi, matteo title: covid- and medical liability: italy denies the shield to its heroes date: - - journal: eclinicalmedicine doi: . /j.eclinm. . sha: doc_id: cord_uid: wf yhyz nan since early june , several public prosecutors at the head of the most important italian judicial districts started to conduct penal investigations on physician behaviors concerning the prevention, diagnosis, and treatment of citizens affected by sars-cov- . the main issue under scrutiny is the actual possibility to avoid or, to better say, to mitigate the pandemic risk in both public and private healthcare institutions. as well known, italy is one of the countries in a worldwide context more severely affected by the sars-cov- pandemic and some of the northern regions paid the highest price in terms of deaths among health care workers (hcws).the 'istituto nazionale assicurazione infortuni sul lavoro' (inail), the italian public insurance body that protects workers in the event of accidents and occupational diseases, reported that % of the filedfatal cases involved hcws [ ] . on the other hand, the statistics of the 'federazione nazionale degli ordini dei medici chirurghi e degli odontoiatri (fnomceo)', the national federation of italian medical doctors and dentists, counted higher figures ( deaths among those registered at . . ), regardless of whether or not a complaint was filed with inail [ ] . in a few weekstime frame, hcws went from heroes to potentially 'negligent' health professionals, while lawyers and prosecutors want to put their decisions and treatments under examination [ , ] . while several liability protection or pandemic-related immunity schemes have been proposed and adopted worldwide, in italy ordinary laws still rule this extraordinary event. the legal cause of action is dual: multiple manslaughter and negligent epidemic. while in the first case prosecutors have to prove the causal relationship between a putative professional misconduct and the cause of death of the plaintiff, in the other scenario it could be sufficient,to convict the attending physicians and/or the ones in charge of hospital clinical governance, to prove that among the huge amount of the sick ( . infected À . % of world and , % of europe -and . deaths À , % world and . % europe) a relative number of infection (even if a residual percentage) could be somehow prevented [ , ] . the prompt adoption of general measures such as universal use of protective masks, social distancing in the wards, the forbidding to assisting relatives in the hospitals and nurse facilities, a generalized testing for sars-cov- , and covid antibodies testing for all hcws, are today considered as mandatory actions to be implemented in everyday clinical activities. although necessary, none of these precautions À alone or in combination À are sufficient to set the risk to zero. from a negligent pandemic point of view, this may mean that if the hospital À even a no-covid one -does not provide for all these measures, and one or more cases of sars-cov- positive patients are detected in the healthcare facility, a presumption of liability may be enough to pursuing a negligent pandemic crime (article , italian penal code). this crime requests as its main features the existence of the risk of a pandemic diffusion of an 'infectious agent' to be avoided ('risk offence') and, as itsconsequence, the event of one or more affected individuals ('event crime'). the starting point of view of medicine and legal system does not always overlap and the final goal may differ. the actionable goal of medicine, although unattainable in concrete terms, is an in-depth knowledge of causes of a disease, its treatment and/or its prevention, being the acceptance of the illiteracy anessential part of our scientific method (error scientiae). the judicial system may not wait for the best clinical explanation of a 'new' phenomenon. it needs in each circumstance the identification of a 'crisis standard of care'. and sometimes, as it is now happening in italy, the easiest path to follow as prosecutor is to contest the crime on the basis of a breach of the duty to protect. the assumption is that, even in time of uncertainty, hcws always have an answer, knowing how to behave to protect the health of the other citizens, using universal measures of prevention or treasuring previous knowledge. but the sars-cov- taught us that it is not always so. we are facing a biological threat whose boundaries change day by day. all the scientific efforts put into play by international community seem to be ineffective, due to sars-cov- infectivity and transmissibility, to knowledge gaps in the comprehension of the basic biology of the infection, to the achilles' heel represented by a-and/or contents lists available at sciencedirect eclinicalmedicine journal homepage: https://www.journals.elsevier.com/eclinicalmedicine pre-symptomatic individuals. no clear procedures are at the moment available to eradicate the risk of infection and in all the italian hospitals or in the nurse facilities a certain number of infections, sometimes even relevant in magnitude, may have been related to a source of viral contagion potentially avoidable. concluding, italian physicians and managers of health facilities are likely to face a new 'tsunami-like' event, i.e. the judicial one. the latter, unless an intervention by the legislator, will be without the protection of a "penal shield". waiting for it, healthcare workers are 'sitting on the dock of the bay, watching the tide roll away'. eda, mn, gf and pen conceived of the presented idea. eda, mn and rd developed the theory. eda wrote the manuscript in consultation with mn, pen, gf and mf scheda nazionale infortuni covid- : i dati sulle denunce da covid- (monitoraggio al giugno ) elenco dei medici caduti nel corso dell'epidemia di safety recommendations and medical liability in ocular surgery during the covid- pandemic: an unsolved dilemma covid- : the wrong target for healthcare liability claims covid- situation update for the eu/eea and the uk, as of the authors declare that they have no conflict of interest. the authors declare no funding for this research key: cord- -gbi g n authors: fioriti, v.; roselli, i.; chinnici, m.; arbore, a.; sigismondi, n. title: estimating the epidemic growth dynamics within the first week date: - - journal: nan doi: . / . . . sha: doc_id: cord_uid: gbi g n information about the early growth of infectious outbreaks are indispensable to estimate the epidemic spreading. a large number of mathematical tools have been developed to this end, facing as much large number of different dynamic evolutions, ranging from sub-linear to super-exponential growth. of course, the crucial point is that we do not have enough data during the initial outbreak phase to make reliable inferences. here we propose a methodology to estimate the epidemic growth dynamics from the infected cumulative data of just a week, provided a surveillance system is available over the whole territory. the methodology, based on the newcomb-benford law, is applied to italian covid case-study. results show that it is possible to discriminate the epidemic dynamics using the first seven data points collected over fifty italian cities. moreover, the form of the most probable approximating function of the growth, within a six weeks epidemic scenario, is identified. during an outbreak, a major issue to stop or mitigate the virus diffusion is gathering information as soon as possible about the nature of the epidemic from a mathematical point of view. pandemic outbreaks allow a reasonable amount of data to be available only ex-post the event, therefore the analysis is severely limited. on the other hand, well-timed information about the epidemic growth is extremely precious, and justify any effort in this direction. a panoply of tools are available, but their accuracies are subject to limitations, due to the small number of data points available, which also restricts the choice of models for the epidemic curve that is a time-series of the cumulative number of cases per day [ , ] . these curves are produced by different dynamics, ranging from sub-linear to super-exponential, giving rise to a diversity of the early growth profiles that has deep implications for the estimation of the disease transmission and for the countermeasures implementation [ ] . therefore, a fast detection and estimation at least of the basic outbreak characteristics, would be extremely useful, but the mathematical tools able to deal with those very few data are rare. moreover, epidemic data gathered on the field are always polluted by human errors, different collection methods, limited territorial coverage, irregular or random sampling. even the most recent sophisticated signal processing techniques such as the graph spectral analysis, the compressive sensing, the signal on graphs method are affected by these problems [ - ] . however, the newcomb-benford law (nbl) seems able to reveal the epidemic dynamics using only a week of infection data. nbl is the statistic of the first (actually also of the second, third and so on) digit for a set of numbers, discovered and rediscovered independently by s. newcomb and f. benford, today known as the benford law [ , , ] . its wide popularity is due to the apparent ubiquity of the benford distribution and to the extreme simplicity of the calculation procedure involved. in recent years, the nbl has been used to discover fiscal frauds and to confirm scientific data reliability, including epidemic data [ ] . in our work, we are interested to study the capabilities of the nbl to predict the outbreak growth dynamics using very few initial data. it is only necessary to collect the cumulative data of the daily infected over a week in some of the most important cities involved in the outbreak, to form a unique sequence of these numbers and then to calculate the first digit distribution. given the limited amount of data points, the calculated and the actual benford distribution will not coincide exactly, thus to characterize the accuracy an appropriate goodness-of-fit (gof) parameter is used. at this point, by trial-and-error or by any numerical technique, an approximating function is chosen: if its first digit distribution is congruent with that of the italian cities, we can consider the approximating function as an accurate approximation to the real cumulative curve. in the following sections we will show how this is possible invoking the theorems of berger & hill and the ergodicity of the epidemic sis process during the initial expansion phase [ ] . before presenting the essential points of our proposal, we have to introduce the main statistical tool, namely the newcomb-benford law. since an extensive treatment can be found in the fundamental work of hill and berger [ , , ] , we will give just a brief introduction to the nbl from a practical point of view. s. newcomb and f. benford independently observed that the leading digits in many real-life numerical data sets such as macroeconomic, census, financial, fiscal data, were not distributed uniformly, as the common sense would suggest, instead they follow the logarithmic distribution of figure a : more formally, the benford distribution is a logarithmic curve: pr (d = d ) = log ( + /d ), for all d = , , . . . , ( ) where d is the first significant decimal digit. first significant digit means in . , in . , . in . * , . in . * - . when a set of numbers follows exactly the nbl, the digit " " will appear about the % of the times, the digit " " the %, " " the % etc.. the generalized nb law considers also the second, third etc. digit, but here we are interested only to the first one. the compliance to the benford distribution may be found also in some natural data-sets, such as molecular weight tables, sport statistics, drainage areas of rivers, that taken individually do not follow the nbl: what satisfies completely the nbl is the union of all those data-sets. the powers of follow the benford distribution (we say "is benford", for short), but to verify it numerically, we would need of a large set of numbers. this constitutes one first difficulty, because it is not easy to determine the minimum cardinality of the set that guarantees a priori to reveal exactly the benford distribution. moreover, most of the time we do not have enough data to satisfy correctly the nbl, and, as a consequence, an error is introduced. in figure b , c the effect of a limited data set is clearly illustrated: using only data points we obtain a poor fit to the actual benford distribution, although simply doubling the datapoints reduces greatly the error. hence, it is convenient to use a goodness-of-fit parameter to estimate the error committed; here, to this end, we consider the standard measure: but other statistical tests may be used as well. of course, generally the first digit distribution of a data-set may or may not be a benford (first digit) distribution; if this is the case, it will be clearly specified. in any case, the gof indicates the distance between the calculated firs digit distribution and the benford distribution. the main idea is to estimate an approximating function for the epidemic growth curve within a time horizon of tf days, using only the first seven epidemic data points of fifty italian cities, accounting for about the % of the population, considered as a unique sequence formed of x data-points, called _cities sequence. we show that the first digit distribution of the _cities sequence converges to the first digit distribution of the cumulative daily infected italian national sequence, formed summing the infected over all the national territory each day during the first tf days of the epidemic ascending phase. therefore, if the convergence exists, it is possible to know the compliance to benford for the cumulative italian curve in advance of tf - days. in turn, the level of compliance is used as a criterion to predict the accuracy of an approximating function to the real epidemic national curve during the initial phase of tf days (starting from the th february ). now we will sketch the theoretical justification of the above method. first of all, we have to make sure that the italian national cumulative data first digit distribution is benford during the initial spreading period. based on the berger & hill theorems [ , ] , if the sequence is a power law, exponential or super-exponential, its first digit distribution is almost always benford. thus, we have only a sufficient condition. below some of the main results of berger & hill to support our method (formal demonstrations can be found in [ , , ] the next theorem states that the solutions of an ordinary or differential equation system such as many of the classical epidemic models, under general conditions are benford. thus, the nbl is not restricted to discrete dynamics; on the other hand, general results for partial differential, delay or integro-differential equations are not known. theorem (berger & hill ) . consider the dynamic system: all rights reserved. no reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in the copyright holder for this this version posted august , . . https://doi.org/ . / . . . doi: medrxiv preprint where f: ℝ → ℝ is continuously differentiable with f( ) = , and x ∈ ℝ. let f: ℝ → ℝ be c with f( ) = and assume f'( ) < . then, for every x ≠ sufficiently close to , the solution of the system is benford. related to theorem we have the so-called shadowing lemma. the lemma describes the behavior of the pseudo-trajectories (sequences) near a locally structurally stable hyperbolic invariant set [ ] . shadowing lemma. let t: ℝ → ℝ be a map, and β a real number with |β| > . if supx∈ ℝ |t (x) − βx| < +∞ then there exists, for every x ∈ ℝ , one and only one point x° such that the sequence (t n (x)−β n x°) is bounded. this means that every hyperbolic set has the shadowing property, thus every pseudo-trajectory or sequence stays uniformly close to some true trajectory, i.e. a pseudo-trajectory is "shadowed" by a true one. considering the epidemic curve as the solution of a dynamic system, the shadowing lemma allows to believe that very close to it a pseudo-sequence exists, is related to the dynamic system and is structurally stable. in other words, finding an approximating function to the cumulative epidemic italian curve would not be a mere accident, at least locally. moreover, small perturbations of the system initial conditions do not change the approximation: therefore, errors during the initial data collections do not alter the result. theorem (berger & hill ) . let x be exponential with mean , that is fx(t) = max( , − e −t ) , t ∈ ℝ. even though x is not exactly benford, it is close to being benford for all t ∈ [ , ). theorem (berger & hill ) . the sequences : n , n are benford; n, n + , n!, √ n , n , * n are not. in general, a*x b , with a > and b > is benford almost always, but not always, therefore x is almost always benford. moreover, every mixture of n with a random unbiased sequence, is benford. berger and hill also state that apart from some particular cases, processes with linear growth are not benford. this allows to identify the slow epidemic growths, that are a phenomenon more common than previously though [ ] . by slow we mean a linear, sub-linear or a polynomial growth. theorem (berger & hill ) . if x and y are benford sequences, also their sum x + y is benford. if the sequence z is not benford, x + y + z is benford. hence, if the cumulative infected sequence of the italian cities are benford, also the national cumulative sequence is benford as well, during the weeks of the increasing phase. note that an epidemic cumulative sequence cannot be random, being non-decreasing, therefore by theorems berger & hill , , , the fast cumulative epidemic curves are all benford, but could exist also fast non-benford curves in particular circumstances. thus, we cannot rule out the possibility of non-benford growth curves to have a fast dynamics, thought this would be seldom the case, see table . now we can discuss the convergence of the _cities distribution to the distribution of the cumulative daily infected of the italian national sequence after tf days. the _cities sequence is the union of the first days data for each city, but to fix ideas, let us consider only three cities, a, b, c, whose sequences of seven elements are: indicating b(…) as the operator of the first digit distribution calculation, results: all rights reserved. no reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in the copyright holder for this this version posted august , . . https://doi.org/ . / . . . doi: medrxiv preprint b( ui,j{ x j (i) }) → b( ∑k,h ( x j (h)) ) ( where i = , , … m. j = , , … n. k = , , … nit . h = , , … tf . in the trivial case, when the sequences x j are (almost all) benford with m = tf and n = nit, berger & hill guarantees that the summation sequence is benford. on the other hand, the union of the sequences is benford too, thus the ( ) is true. when m = , since it is not known an analytical method to determine a priori the values of m that guarantees a small gof, one can only calculate the first digit distribution b(ui,j{ x j (i) }) and compare it to the benford distribution of figure a . if the gof stays high, the simplest heuristics would be to increase m, yet reducing the forecasting time-span of tfm. of course, the number of city-sequences n may well be expanded to all the cities nit : here we have restricted it to fifty cities only for demonstration purposes. therefore, we do know that ( ) is true, but cannot determine the minimum m necessary. actually, we have chosen m = because it is well below the prediction thresholds often suggested in the literature [ , ] . instead, to determine tf , we consider that during the initial phase of the outbreak if: (where is the birth rate, is the cure rate, and is the contact rate), the overall epidemic process is ergodic [ ] . hence, all the local realizations have similar statistics, are non random non decreasing sequences, and by the shadowing lemma there exists an approximant function f to ∑k,h ( x j (h): as a consequence, the first digit distribution of f(x) approximates that of ∑k,h ( x j (h) , but by the ( ) also that of b( ui,j{ x j (i) }). therefore tf can be determined heuristically or numerically as the value that get b(f(x)) closer to b(ui,j{ x j (i) }) in terms of benford goodness-of-fit. from table it is readily seen that both the cubic and the logistic curve approximate the it_real gof very well for tf = . in addition, tf = is very close to the inflection point of the real italian cumulative curve, which indicates the end of the initial phase for the outbreak. basically, we have three dynamics, very fast, fast, and slow (see figure ); we want to determine which one of them is prevailing by means of the benford' gof, and possibly to find an approximating function. as said in the above section, each city provides a sequence of positive integers, and putting them together the fifty sequences make up an unique sequence of x integers called _cities. note that sequences such as in the section above we have indicated that: where i = , , … . j = , , … k = , , … . nit h = , , … , and tf = are now specified to the actual case-study scenario. to classify various possible approximant curves we have calculated their benford gof, showed in the table , together with the gof of the real italian epidemic data, the logistic curve, of the cubic curve and of the _cities. all rights reserved. no reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in the copyright holder for this this version posted august , . . table ; dotted magenta and red: very fast dynamics; black: the italian cumulative curve and in blue the cubic and logistic approximations; green: slow dynamics. b) days scenario. the cubic function (blue-yellow) approximate well the actual cumulative curve (black) also in this scenario, confirming the same result suggested in [ ] , while the logistic in this scenario performs poorly, and is not in the figure. note the red dotted curve ( * . n ), that seems slow, actually is a fast one; instead, the green black dotted curve ( n ) seems fast, but is very slow. keeping in mind that the union of the sequences of seven data points from some of the most important italian cities, converges to the benford distribution of the sum of all the week sequences for all the cities ( ), theorem berger&hill [ ] guarantee that almost always the exponential growth provides numbers according to the nb distribution. therefore, as explained in other sections, a good benford gof obtained from the epidemic cumulative data of the _cities should be able to identify the type of outbreak dynamic. in table the cities gof is . , very close to the gof for the real italian data of the first six weeks, gof_it = . , while the cubic curve gof_i(ti)*n = . has the minimal distance from the _cities gof. therefore n is the most probable approximating function, whereas in figure a the italy' real epidemic curve is almost coincident with the cubic growth, confirming the results of [ ] . actually, the logistic curve gof is in good agreement with the _cities gof too, and fits very satisfactorily the italy' real data after the fourth week (figure b ), but the cubic curve has an advantage in terms of benford gof and till the fourth week is also the best approximant. therefore, table reveals that the cubic curve determines the initial stage of the growth. again a note of caution: a small gof, say in the interval [ , ] as in table , is only a sufficient condition that guarantees a fast dynamic, but not a necessary one, meaning that a large gof > could represent a rapid growth too, as for the case of the function b * n . thus, most of the times, but not always, a large gof indicates a linear or sub-linear dynamics; instead, a small gof guarantees a rapid growth. a b figure . dotted blue: the italian cumulative curve, red: cubic approximant, black: logistic approximant. a) first four weeks, the cubic curve is very close to the real growth. b) first six weeks: the logistic curve now is a better approximant. the best approximating function function is the cubic i(ti)n . samples is the number of data-points used to calculate the first digit distribution and therefore also the gof; they are , except for the _cities gof, whose distribution is calculated using not more than data-points, the first data from each city. in orange * are indicated the non-congruent dynamics: for example, * n has a very large gof, nonetheless belongs to the class of fast dynamics, instead to that of the slow one. summarizing, the first digit distribution of the _cities sequence converges to the first digit distribution of the cumulative infected curve, that is benford during the ascending epidemic phase. if the _cities gof is small, on the basis of the berger-hill's theorem we have a fast epidemic growth; instead, if the gof is poor, the growth will be probably slow, although this cannot be assured formally. moreover, in order to determine the form of the approximating function that shadows the real epidemic growth, one might extrapolate some of them analytically or heuristically. at this point, it suffices to choose the approximating curve whose gof is close to that of the _cities. if the difference between the two gof is small (in our case study less than %), we assume that the approximating function shadows correctly the real epidemic growth. in this paper, we show how to estimate an approximating function of the epidemic curve within a time horizon of five or six weeks, using only the first seven epidemic data points of fifty italian cities, considered as an unique sequence. the level of compliance of the sequence to the benford test is used as a criterion to predict the approximating function accuracy with respect to the real epidemic national curve. this procedure is made possible by the convergence of the unique sequence first digit distribution to the benford distribution, since the national italian cumulative data first digit distribution is benford (almost always) when its sequence is a power, exponential or faster curve. therefore, when a fast epidemic spread is taking place on the territory, its fingerprint will be the benford distribution, otherwise the spreading will be, with high probability, slow (quasilinear) or made of sporadic outbursts. unfortunately, the benford compliance of the fast growth is only a sufficient condition, not a necessary one. our results make this point clear. on the other hand, early indications about the dynamic nature of the outbreak are made available through a simple statistical test applied to a handful of data. this method has been applied to the italian covid case study, where the most probable approximating function of the first five weeks has been identified clearly as a cubic curve. a generalized-growth model to characterize the early ascending phase of infectious disease outbreaks building test data from real outbreaks for evaluating detection algorithms bayesian prediction of an epidemic curve discrete signal processing on graphs: sampling theory the emerging field of signal processing on graphs: extending high-dimensional data analysis to networks and other irregular domains discrete signal processing on graphs adaptive least mean squares estimation of graph signals distributed recursive least squares strategies for adaptive reconstruction of graph signals online temperature estimation using graph signals predicting the sources of an outbreak with a spectral technique locating the source of diffusion in large scale network the role of the airline transportation network in the prediction and predictability of global epidemics locating multiple sources of contagion in complex networks under the sir model the network topology of connecting things: defense of iot graph in the smart city modelling dynamical processes in complex socio-technical systems epidemic spreading in scale-free networks epidemic thresholds in real networks epidemic spreading in real networks: an eigenvalue viewpoint utility and potential of rapid epidemic intelligence from internet-based sources a statistical derivation of the significant digit law a basic theory of benford's law the first digit phenomenon using benford's law to investigate natural hazard dataset homogeneity extinction and ergodic property of stochastic sis epidemic model with nonlinear incidence rate key: cord- -bjy ixcj authors: stella, massimo; restocchi, valerio; deyne, simon de title: #lockdown: network-enhanced emotional profiling at the times of covid- date: - - journal: nan doi: nan sha: doc_id: cord_uid: bjy ixcj the covid- pandemic forced countries all over the world to take unprecedented measures like nationwide lockdowns. to adequately understand the emotional and social repercussions, a large-scale reconstruction of how people perceived these unexpected events is necessary but currently missing. we address this gap through social media by introducing mercurial (multi-layer co-occurrence networks for emotional profiling), a framework which exploits linguistic networks of words and hashtags to reconstruct social discourse describing real-world events. we use mercurial to analyse , tweets from italy, the first country to react to the covid- threat with a nationwide lockdown. the data were collected between th and th march, immediately after the announcement of the italian lockdown and the who declaring covid- a pandemic. our analysis provides unique insights into the psychological burden of this crisis, focussing on: (i) the italian official campaign for self-quarantine (#iorestoacasa}), (ii) national lockdown (#italylockdown), and (iii) social denounce (#sciacalli). our exploration unveils evidence for the emergence of complex emotional profiles, where anger and fear (towards political debates and socio-economic repercussions) coexisted with trust, solidarity, and hope (related to the institutions and local communities). we discuss our findings in relation to mental well-being issues and coping mechanisms, like instigation to violence, grieving, and solidarity. we argue that our framework represents an innovative thermometer of emotional status, a powerful tool for policy makers to quickly gauge feelings in massive audiences and devise appropriate responses based on cognitive data. the stunningly quick spread of the covid- pandemic catalysed the attention of worldwide audiences, overwhelming individuals with a deluge of often contrasting content about the severity of the disease, the uncertainty of its transmission mechanisms, and the asperity of the measures taken by most countries to fight it [ , , , ] . although these policies have been seen as necessary, they had a tremendous impact on the mental well-being of large populations [ ] for a number of reasons. due to lockdowns, many are facing financial uncertainty, having lost or being on the verge of losing their source of income. moreover, there is much concern about the disease itself, and most people fear for their own health and that of their loved ones [ ] , further fueled by infodemics [ , , ] . finally, additional distress is caused by the inability of maintaining a normal life [ ] . the extent of the impact of these factors is such that, in countries greatly struck by covid- such as china, the population started to develop symptoms of post-traumatic stress disorder [ ] . during this time more than ever, people have shared their emotions on social media. these platforms provide an excellent emotional thermometer of the population, and have been widely explored in previous studies investigating how online social dynamics promote or hamper content diffusion [ , , , , ] and the adoption of specific positive/negative attitudes and behaviours [ , , ]. building on the above evidence, our goal is to draw a comprehensive quantitative picture of people's emotional profiles, emerging during the covid- crisis, through a cognitive analysis of online social discourse. we achieve this by introducing mercurial (multi-layer co-occurrence networks for emotional profiling), a framework that combines cognitive network science [ , , ] with computational social sciences [ , , , , ] . before outlining the methods and main contributions of our approach, we briefly review existing research on understanding emotions in social media. much of the research on emotions in social media has been consolidated into two themes. on the one hand, there is the data science approach, which mostly focused over large-scale positive/negative sentiment detection [ ] and recently identified the relevance of tracing more complex affect patterns for understanding social dynamics [ , , , ] . on the other hand, cognitive science research makes use of small-scale analysis tools, but explores the observed phenomena in much more detail in the light of its theoretical foundations [ , , ] . specifically, in cognitive science the massive spread of semantic and emotional information through verbal communication represent long-studied phenomena, known as cognitive contagion [ ] and emotional contagion [ , , ] , respectively. this research suggests that ideas are composed of a cognitive component and an emotional content, much alike viruses containing the genomic information necessary for their replication [ ] . both these types of contagion happen when an individual is affected in their behaviour by an idea. emotions elicited by ideas can influence users' behaviour without their awareness, resulting in the emergence of specific behavioural patterns such as implicit biases [ ] . unlike pathogen transmission, no direct contact is necessary for cognitive and emotional contagion to take place, since both are driven by information processing and diffusion, like it happens through social media [ , ] . in particular, during large-scale events, ripples of emotions can rapidly spread across information systems [ ] and have dramatic effects, as it has recently been demonstrated in elections and social movements [ , , ] . at the intersection of data-and cognitive science is emotional profiling, a set of techniques which enables the reconstruction of how concepts are emotionally perceived and assembled in usergenerated content [ , , , , , ] . emotional profiling conveys information about basic affective dimensions such how positive/negative or how arousing a message is, and also includes the analysis of more fine-grained emotions such as fear or trust that might be associated with the lockdown and people's hopes for the future [ , , ] . recently, an emerging important line of research has shown that reconstructing the knowledge embedded in messages through social and information network models [ , , ] successfully highlight important phenomena in a number of contexts, ranging from the diffusion of hate speech during massive voting events [ ] to reconstructing personality traits from social media [ ] . importantly, to reconstruct knowledge embedded in tweets, recent work has successfully merged data science and cognitive science, introducing linguistic networks of co-occurrence relationships between words in sentences [ , , ] and between hashtags in tweets [ ] . however, an important shortfall of these works is that these two types of networked knowledge representations were not merged together, thus missing on the important information revealed by studying their interdependence. we identify three important contributions that distinguish our paper from previous literature, and make a further step towards consolidating cognitive network science [ ] as a paradigm suitable to analyse people's emotions. first, we introduce a new framework exploiting the interdependence between hashtags and words, addressing the gap previously discussed. this framework, multi-layer co-occurrence networks for emotional profiling (mercurial), combines both the semantic structure encoded through the co-occurrence of hashtags and the textual message to construct a multi-layer lexical network [ ] . this multi-layer network structure allows us to contextualise hashtags and, therefore, improve the analysis of their meaning. importantly, these networks can be used to identify which concepts or words contribute to different emotions and how central they are. second, in contrast to previous work, which largely revolved around english tweets [ , ] , the current study focusses on italian twitter messages. there are several reasons why the emotional response of italians is particularly interesting. specifically, i) italy was the first western country to experience a vast number of covid- clusters; ii) the italian government was the first to declare a national lockdown ; iii), the italian lockdown was announced on th march, one day before the world health organization (who) declared the pandemic status of covid- . this enables us to address the urgent need of measuring the emotional perceptions and reactions to social distancing, lockdown, and, more generally, the covid- pandemic. third, thanks to mercurial, we obtain richer and more complex emotional profiles that we analyse through the lens of established psychological theories of emotion. this is a fundamental step in going beyond positive/neutral/negative sentiment and to provide accurate insights on the mental well-being of a population. to this end, we take into account three of the most trending hashtags, #iorestoacasa (english: "i stay at home"), #sciacalli (english: "jackals"), and #italylockdown, as representative of positive, negative, and neutral social discourse, respectively. we use these hashtags as a starting point to build multi-layer networks of word and hashtag co-occurrence, from which we derive our profiles. our results depict a complex map of emotions, suggesting that there is co-existence and polarisation of conflicting emotional states, importantly fear and trust towards the lockdown and social distancing. the combination of these emotions, further explored through semantic network analysis, indicates mournful submission and acceptance towards the lockdown, perceived as a measure for preventing contagion but with negative implications over economy. as further evidence of the complexity of the emotional response to the crisis, we also find strong signals of hope and social bonding, mainly in relation to social flash mobs, and interpreted here as psychological responses to deal with the distress caused by the threat of the pandemic. the paper is organised as follows. in the methods section we describe the data we used to perform our analysis, and describe mercurial in detail. in the results section we present the emotional profiles obtained from our data, which are then discussed in more detail in the section discussion. finally, the last section highlights the psychological implications of our exploratory investigation and its potential for follow-up monitoring of covid- perceptions in synergy with other datasets/approaches. we argue that our findings represent an important first step towards monitoring both mental well-being and emotional responses in real time, offering policy-makers a framework to make timely data-informed decisions. in this section we describe the methodology employed to collect our data and perform the emotional profiling analysis. first, we describe the dataset and how it was retrieved. then, we introduce cooccurrence networks, and specifically our novel method that combines hasthag co-occurrence with word co-occurrence on multi-layer networks. finally, we describe the cognitive science framework we used to perform the emotional profiling analysis on the so-obtained networks. we gathered , tweets in italian to monitor how online users perceived the covid- pandemic and its repercussions in italy. these tweets were gathered by crawling messages containing three trending hashtags of relevance for the covid- outbreak in italy and expressing three different sentiment polarities: • #iorestoacasa (english: "i stay at home"), a positive-sentiment hashtag introduced by the italian government in order to promote a responsible attitude during the lockdown; • #sciacalli, (english: "jackals"), a negative sentiment hashtag used by online users in order to address unfair behaviour rising during the health emergency; • #italylockdown, a neutral sentiment hashtag indicating the application of lockdown measures all over italy. we refer to #iorestoacasa, #sciacalli and #italylockdown as focal hashtags to distinguish them from other hashtags. we collected the tweets through complex science consulting (@complexconsult), which was authorised by twitter, and used the serviceconnect crawler implemented in mathematica . . the collection of tweets comprises , tweets for #iorestoacasa, , for #sciacalli and , for #italylockdown. retweets of the same text message were not considered. for each tweet, the language was detected. pictures, links, and non-italian content was discarded and stopwords (i.e. words without intrinsic meaning such as "di" (english: "of") and "ma" (english: "but") removed. other interesting datasets with tweets about covid- are available in [ , ] . word co-occurrence networks have been successfully used to characterise a wide variety of phenomena related to language acquisition and processing [ , , ] . recently, researchers have also used hashtags to investigate various aspects of social discourse. for instance, stella et al. [ ] showed that hashtag co-occurrence networks were able to characterise important differences in the social discourses promoted by opposing social groups during the catalan referendum. in this work we introduce mercurial (multi-layer co-occurrence networks for emotional profiling), a framework combining: • hashtag co-occurrence networks (or hashtag networks) [ ] . nodes represent hashtags and links indicate the co-occurrence of any two nodes in the same tweet. • word co-occurrence networks (or word networks) [ ] . nodes represent words and links represent the co-occurrence of any two words one after the other in a tweet without stop-words (i.e. words without an intrinsic meaning). we combine these two types of networks in a multi-layer network to exploit the interdependence between hashtags and words. this new, resulting network enables us to contextualise hashtags, and capture their real meaning through context, thereby enhancing the accuracy of the emerging emotional profile. to build the multi-layer network, we first build the single hashtag and word layers. for sake of simplicity, word networks are unweighted and undirected . note that the hashtag network was kept at a distinct level from word networks, e.g. common words were not explicitly linked with hashtags. as reported in figure , each co-occurrence link between any two hashtags a and b (#coronavirus and #restiamoacasa in the figure) is relative to a word network, including all words co-occurring in all tweets featuring hashtags a and b. the hashtag and word networks capture the co-occurrence of lexical entities within the structured online social discourse. words possess meaning in language [ ] and their network assembly is evidently a linguistic network. similar to words in natural language, hashtags possess linguistic features that express a specific meaning and convey rich affect patterns [ ] . the resulting networks capture the meaning of a collection of tweets by identifying which words/hashtags co-occurred together. this knowledge embedded in hashtag networks was used in order to identify the most relevant or central terms associated within a given collection of thematic tweets. rather than using frequency to indicate centrality, which makes it difficult to compare hashtags that do not co-occur in the same message, the current work relies on distance-based measures to detect how central a hashtag is in the network. the first measure that implements this notion is closeness centrality. closeness c(i) identifies how many links connect i to all its neighbours and is formalised as follows: where d ij is the network distance between i and j, i.e. the smallest amount of links connecting nodes i and j. in co-occurrence networks, nodes (i.e. hashtags and words) with a higher closeness tend to co-occur more often with each other or with other relevant nodes at short network distance. we expect that rankings of closeness centrality will reveal the most central hashtags in the networks for #iorestoacasa, #sciacalli and #italylockdown, in line with previous work in which closeness centrality was used to measure language acquisition and processing [ , , ] . importantly, closeness is a more comprehensive approach compared to the simpler frequency analysis. imagine a collection of hashtags a, b, c, d, .... computing the frequency of hashtag figure : top: example of co-occurrence networks for different hashtags: #distantimauniti (english: distant but united) in #iorestoacasa on the left, #incapaci (english: inept) in #sciacalli in the middle, and #futuro (english: future) in #italylockdown on the right. clusters of co-occurring hashtags were obtained through spectral clustering [ ] . these clusters highlight the co-occurrence of european-focused content, featuring hashtags like #bce (i.e. european central bank), #lagarde and #spread (i.e. spread between italian and german bonds) together with social distance practices related to #iorestoacasa. bottom: in mercurial, any link in a co-occurrence network of hashtags (left) corresponds to a collection of tweets whose words co-occur according to a word network (right). larger words have a higher closeness centrality. a co-occurring with hashtag b is informative about the frequency of the so-called -grams "ab" or "ba" but it does not consider how those hashtags co-occur with c, d, etc. in other words, a -gram captures the co-occurrence of two specific hashtags within tweets but does not provide the simultaneous structure of co-occurrences of all hashtags across tweets, for which a network of pairwise co-occurrences is required. on such a network, closeness can then highlight hashtags at short distance from all others, i.e. co-occurring in a number of contexts in the featured discourse. in addition to closeness, we also use graph distance entropy to measure centrality. this centrality measure captures which hashtags are uniformly closer to all other hashtags in a connected network. combining closeness with graph distance entropy led to successfully identifying words of relevance in conceptual networks with a few hundreds of nodes [ ] . the main idea behind graph distance entropy is that it provides info about the spread of the distribution of network distances between nodes (i.e. shortest path), a statistical quantity that cannot be extracted from closeness (which is, conversely, a mean inverse distance). considering the set d (i) ≡ (d i , ..., d ij , ..., d in ) of distances between i and any other node j connected to it ( ≤ j ≤ n ) and m i = m ax(d (i) ), then graph distance entropy is defined as: where p k is the probability of finding a distance equal to k. therefore, h(i) is a shannon entropy of distances and it ranges between and . in general, the lower the entropy, the more a node resembles a star centre [ ] and is at equal distances from all other nodes. thus, nodes with a lower h(i) and a higher closeness are more uniformly close to all other connected nodes in a network. words with simultaneously low graph distance entropy and high closeness were found to be prominent words for early word learning [ ] and mindset characterisation [ ] . in addition to hashtag networks, we also build word networks obtained from a collection of tweets containing any combination of the focal hashtags #iorestocasa or #sciacalli and #coronavirus. for all tweets containing a given set of hashtags, we performed the following: . subdivide the tweet in sentences and delete all stop-words from each sentence, preserving the original ordering of words; . stem all the remaining words, i.e. identify the root or stem composing a given word. in a language such as italian, in which there is a number of ways of adding suffixes to words, word stemming is essential in order to recognise the same word even when it is inflected for different gender, number or as a verb tense. for instance, abbandoneremo (we will abandon) and abbandono (abandon, abandonment) both represent the same stem abband; . draw links between a stemmed word and its subsequent one. store the resulting edge list of word co-occurrences. . sentences containing a negation (i.e. "not") underwent an additional step parsing their syntactic structure. this was done in order to identify the target of negation (e.g. in "this is not peace", the negation refers to "peace"). syntactic dependencies were not used for network construction but intervened in emotional profiling, instead (see below). the resulting word network also captures syntactic dependencies between words [ ] related by online users to a specific hashtag or combination of hashtag. we used closeness centrality to detect the relevance of words for a given hashtag. text pre-processing such as word stemming and syntactic dependencies was performed using mathematica . , which was also used to extract networks and compute network metrics. the presence of hashtags in word networks provided a way of linking words, which express common language, with hashtags, which express content but also summarise the topic of a tweet. consequently, by using this new approach, the meaning attributed by users to hashtags can be inferred not only from hashtag co-occurrence but also from word networks. an example of mercu-rial, featuring hashtag-hashtag and word-word co-occurrences, is reported in figure (bottom). in this example, hashtags #coronavirus and #restiamoacasa co-occurred together (left) in tweets featuring many co-occurring words (right). the resulting word network shows relevant concepts such as "incoraggiamenti" (english: encouragement) and "problemi" (english: problems), highlighting a positive attitude towards facing problems related to the pandemic. more in general, the attribution and reconstruction of such meaning was explored by considering conceptual relevance and emotional profiling in one or several word networks related to a given region of a hashtag co-occurrence network. as a first data source for emotional profiling, this work also used valence and arousal data from warriner and colleagues [ ] , whose combination can reconstruct emotional states according to the well-studied circumplex model of affect [ , ] . in psycholinguistics, word valence expresses how positively/negatively a concept is perceived (equivalently to sentiment in computer science). the second dimension, arousal, indicates the alertness or lethargy inspired by a concept. having a high arousal and valence indicates excitement and joy, whereas a negative valence combined with a high arousal can result in anxiety and alarm [ ] . finally, some studies also include dominance or potency as a measure of the degree of control experienced [ ] . however, for reasons of conciseness, we focus on the two primary dimensions of affect: valence and arousal. going beyond the standard positive/negative/neutral sentiment intensity is of utmost importance for characterising the overall online perception of massive events [ ] . beyond the primary affective dimension of sentiment, the affect associated with current events [ ] can also be described in terms of arousal [ ] and of basic emotions such as fear, disgust, anger, trust, joy, surprise, sadness, and anticipation. these emotions represent basic building blocks of many complex emotional states [ ] , and they are all self-explanatory except for anticipation, which indicates a projection into future events [ ] . whereas fear, disgust, and anger (trust and joy) elicit negative (positive) feedback, surprise, sadness and anticipation have been recently evaluated as neutral emotions, including both positive and negative feedback reactions to events in the external world [ ] . to attribute emotions to individual words, we use the nrc lexicon [ ] and the circumplex model [ ] . these two approaches allow us to quantify the emotional profile of a set of words related to hashtags or combinations of hashtags. the nrc lexicon enlists words eliciting a given emotion. the circumplex model attributes valence and arousal scores to words, which in turn determine their closest emotional states. because datasets of similar size were not available for italian, the data from the nrc lexicon and the warriner norms were translated from english to italian using a forward consensus translation of google translate, microsoft bing and deepl translator, which was successfully used in previous investigations with italian [ ] . although the valence of some concepts might change across languages [ ] , word stemming related several scores to the same stem, e.g. scores for "studio" (english: "study") and "studiare" (english: "to study") were averaged together and the average attributed to the stem root "stud". in this way, even if non-systematic cross-language valence shifting introduced inaccuracy in the score for one word (e.g. "studiare"), averaging over other words relative to the same stem reduced the influence of such inaccuracy. no statistically significant difference (α = . ) was found between the emotional profiles of italian tweets, including different stems, and their automatic translations in english, holding for each dimension separately (z-scores < . ). then, we build emotional profiles by considering the distribution of words eliciting a given emotion/valence/arousal and associated to specific hashtags in tweets. assertive tweets with no negation were evaluated directly through a bag of words model, i.e. by directly considering the words composing them. tweets including negations underwent an additional intermediate step where words syntactically linked to the negation were substituted with their antonyms [ ] and then evaluated. source-target syntactic dependencies were computed in mathematica . and all words targeted by a negation word (i.e. no, non and nessuno in italian) underwent the substitution with their antonyms. to determine whether the observed emotional intensity r(i) of a given emotion in a set s of words was compatible with random expectation, we perform a statistical test (z-test) using the nrc dataset. remember that emotional intensity here was measured in terms of richness or count of words eliciting a given emotion in a given network. as a null model, we use random samples as follows: let us denote by m the number of words stemmed from s that are also in the nrc dataset. then, m words from the nrc lexicon are sampled uniformly at random and their emotional profile is compared against that of the empirical sample. we repeated this random sampling times for each single empirical observed emotional profile {r(i)} i . to ensure the resulting profiles are indeed compatible with a gaussian distribution, we performed a kolmogorov-smirnov test (α = . ). all the tests we performed gave random distributions of emotional intensities compatible with a gaussian distribution, characterised by a mean random intensity for emotion i, r * (i) and a standard deviation σ * (i). for each emotion, a z-score was computed: in the remainder of the manuscript, every emotional profile incompatible with random expectation was highlighted in black or marked with a check. since we used a two-tailed z-test (with a significance level of . ), this means that an emotional richness can either be higher or lower than random expectation. the investigated corpus of tweets represents a complex multilevel system, where conceptual knowledge and emotional perceptions are entwined on a number of levels. tweets are made of text and include words, which convey meaning [ ] . from the analysis of word networks, we can obtain information on the organisation of knowledge proper of social media users, which is embedded in their generated content [ ] . however, tweets also convey meaning through the use of hashtags, which can either refer to specific words or point to the overall topic of the whole tweet. both words and hashtags can evoke emotions in different contexts, thus giving rise to complex patterns [ ] . similar to words in natural language, the same hashtags can be perceived and used in language differently by different users, according to the context. the simultaneous presence of word-and hashtag-occurrences in tweets is representative of the knowledge shared by social media users when conveying specific content and ideas. this interconnected representation of knowledge can be exploited by simultaneously considering both hashtag-level and word-level information, since words specify the meaning attributed to hashtags. in this section we use mercurial to analyse the data collected. we do so by characterising the hashtag networks, both in terms of meaning and emotional profiles. precedence is given to hashtags as they not only convey meaning as individual linguistic units but also represent more general-level topics characterising the online discourse. then, we inter-relate hashtag networks with word networks. finally, we perform the emotional profiling of hashtags in specific contexts. the combination of word-and hashtag-networks specifies the perceptions embedded by online users around the same entities, e.g. coronavirus, in social discourses coming from different contexts. the largest connected components of the three hashtag networks included: hashtags and links for #italylockdown; hashtags and links for #sciacalli; hashtags and links for #italylockdown. all three networks are found to be highly clustered (mean local clustering coefficient [ ] of . ) and with an average distance between any two hashtags of . . only hashtags were present in all the three networks. table reports the most central hashtags included in each corpus of tweets thematically revolving around #iorestoacasa, #sciacalli and #italylockdown. the ranking relies on closeness centrality, which in here quantifies the tendency for hashtags to co-occur with other hashtags expressing analogous concepts and, therefore, are at short network distance from each other (see methods). hence, hashtags with a higher closeness centrality represent the prominent concepts in the social discourse. this result is similar to those showing that closeness centrality captures concepts which are relevant for early word acquisition [ ] and production [ ] in language. additional evidence that closeness can capture semantically central concepts is represented by the closeness ranking, which assigns top-ranked positions to #coronavirus and #covid- in all three twitter corpora. this is a consequence of the corpora being about the covid- outbreak (and of the network metric being able to capture semantic relevance). in the hashtag network built around #italylockdown, the most central hashtags are relative to the coronavirus, including a mix of negative hashtags such as #pandemia (english: "pandemic") and positive ones such as #italystaystrong. similarly, the hashtag network built around #sciacalli highlighted both positive (#facciamorete (english: "let's network") and negative (#irresponsabili -english: "irresponsible") hashtags. however, the social discourse around #sciacalli also featured prominent hashtags from politics, including references to specific italian politicians, to the italian government, and hashtags expressing protest and shame towards the acts of a prominent italian politician. conversely, the social discourse around #iorestoacasa included many positive hashtags, eliciting hope for a better future and the need to act responsibly (e.g. #andratuttobene -english: "everything will be fine", or #restiamoacasa -english: "let's stay at home"). the most prominent hashtags in each network (cf. table ) indicate the prevalence of a positive social discourse around #iostoacasa and the percolation of strong political debate in relation to the negative topics conveyed by #sciacalli. however, we want to extend these punctual observations of negative/positive valences of single hashtags to the overall global networks. to achieve this, we use emotional profiling. hashtags can be composed of individual or multiple words. by extracting individual words from the hashtags of a given network, it is possible to reconstruct the emotional profile of the social discourse around the focal hashtags #sciacalli, #italylockdown and #iorestoacasa. we tackle this by using the emotion-based [ ] and the dimension based [ ] emotional profiles (see methods). the emotional profiles of hashtags featured in co-occurrence networks are reported in figure (top). the top section of the figure represents perceived valence and arousal represented as a circumplex model of affect [ ] . this d space or disk is called emotional circumplex and its coordinates represent emotional states that are well-supported by empirical behavioural data and brain research [ ] . as explained also in the figure caption, each word is endowed with an (x, y) table : top-ranked hashtags in co-occurrence networks based on closeness centrality. higher ranked hashtags co-occurred with more topic-related concepts in the same tweet. in all three rankings, the most central hashtag was the one defining the topic (e.g. #italylockdown) and was omitted from the ranking. (- . ,+ . ) represents a point of strong negative valence and positive arousal, i.e. alarm. figure reports the emotional profiles of all hashtags featured in co-occurrence networks for #italylockdown (left), #sciacalli (middle) and #iorestoacasa (right). to represent the interquartile range of all words for which valence/arousal rating are available, we use a neutrality range. histograms falling outside of the neutrality range indicate specific emotional states expressed by words included within hashtags (e.g. #pandemia contains the word "pandemia" with negative valence and high arousal). in figure (left, top), the peak of the emotional distribution for hashtags associated with #italylockdown falls within the neutrality range. this finding indicates that hashtags co-occurring with #italylockdown, a neutral hashtag by itself, were also mostly emotionally neutral conceptual entities. despite this main trend, the distribution also features deviations from the peak mostly in the areas of calmness and tranquillity (positive valence, lower arousal) and excitement (positive valence, higher arousal). weaker deviations (closer to the neutrality range) were present also in the area of anxiety. this reconstructed emotional profile indicates that the italian social discourse featuring #italylockdown was mostly calm and quiet, perceiving the lockdown as a positive measure for countering responsibly the covid- outbreak. not surprisingly, the social discourse around #sciacalli shows a less prominent positive emotional profile, with a higher probability of featuring hashtags eliciting anxiety, negative valence and increased states of arousal, as it can be seen in figure (center, top) . this polarised emotional profile represents quantitative evidence for the coexistence of mildly positive and strongly negative content within the online discourse labelled by #sciacalli. this is further evidence that the negative hashtag #sciacalli was indeed used by italian users to denounce or raise alarm over the negative implications of the lockdown, especially in relation to politics and politicians' actions. however, the polarisation of political content and debate over social media platforms has been encountered in many other studies [ , , ] and cannot be attributed to the covid- outbreak only. finally, figure (top right) shows that positive perception was more prominently reflected in the emotional profile of #iorestoacasa, which was the hashtag massively promoted by the italian government for supporting the introduction of the nationwide lockdown in italy. the emotional figure : emotional profiles of all hashtags featured in co-occurrence networks for #italylockdown (left), #sciacalli (middle) and #iorestoacasa (right). top: circumplex emotional profiling. all hashtags representing one or more words were considered. for each word, valence (x-coordinate) and arousal (y-coordinate) scores were attributed (see methods) resulting in a d density histogram (yellow overlay) relative to the probability of finding an hashtag in a given location in the circumplex, the higher the probability the stronger the colour. regions with the same probabilities are enclosed in grey lines. a neutrality range indicates where % of the words in the underlying valence/arousal dataset would fall and it thus serves as a reference value for detecting abnormal emotional profiles. distributions falling outside of this range indicate deviations from the median behaviour (i.e. interquartile range, see methods). bottom: nrc-based emotional profiling, detecting how many hashtags inspired a given emotion in a hashtag network. results are normalised over the total number of hashtags in a networks. emotions compatible with random expectation were highlighted in gray. profile of the hashtags co-occurring with #iorestoacasa indicate a considerably positive and calm perception of domestic confinement, seen as a positive tool to stay safe and healthy. the prominence of hopeful hashtags in association with #iorestoacasa, as reported in the previous subsection, indicate that many italian twitter users were serene and hopeful about staying at home at the start of lockdown. hashtag networks were emotionally profiled not only by using the circumplex model (see above) but also by using basic emotional associations taken from the nrc emotion lexicon (figure , bottom) . across all hashtag networks, we find a statistically significant peak in trust, analogous of the peaks close to emotions of calmness and serenity an observed in the circumplex models. however, all the hashtag networks included also negative emotions like anger and fear, which are natural human responses to unknown threats and were observed also with the circumplex representations. the intensity of fearful, alarming and angry emotions is stronger in the #sciacalli hashtag network, which was used by social users to denounce, complain and express alertness about the consequences of the lockdown. in addition to the politically-focused jargon highlighted by closeness centrality alone, by combining closeness with graph distance entropy (see methods and [ ] ) we identify other topics which are uniformly at short distance from others in the social discourse around #sciacalli, such as: #mascherine (english: "protective masks", which was also ranked high by using closeness only), #amuchina (the most popular brand, and synonym of, hand sanitiser), #supermercati (english: "supermarkets"). this result suggests an interesting interpretation of the negative emotions around #sciacalli. beside the inflaming political debate and the fear of the health emergency, in fact, a third element emerges: italian twitter users feared and were angry about the raiding and stockpiling of first aid items, symptoms of panic-buying in the wake of the lockdown. the above comparisons indicate consistency between dimension-based (i.e. the circumplex) and emotion-specific emotional profiling. since the latter offers also a more precise categorisation of words in emotions, we will focus on emotion-specific profiling. importantly, to fully understand the emotional profiles outlined above, it is necessary to identify the language expressed in tweets using a given combination of hashtags (see also figure , bottom). as the next step of the mercurial analysis, we gather all tweets featuring the focal hashtags #italylockdown, #sciacalli, or #iorestoacasa and any of their co-occurring hashtags and build the corresponding word networks, as explained in the methods. closeness centrality over these networks provided the relevance of each single word in the social discourse around the topic identified by a hashtag. only words with closeness higher than the median were reported. figure shows the cloud of words appearing in all tweets that include #sciacalli, displayed according to their nrc emotional profile. similar to the emotional profile extracted from hashtags co-occurring with #sciacalli, the words used in tweets with this hashtag also display a polarised emotional profile with high levels of fear and trust. thanks to the multi-layer analysis, this dichotomy can now be better understood in terms of the individual concepts eliciting it. by using closeness on word networks, we identified concepts such as "competente" (english: "competent"), "continua" (english: "continue", "keep going"), and "comitato" (english: "committee") to be relevant for the trust-sphere. these words convey trust in the expert committees appointed by the italian government to face the pandemic and protect the citizens. we find that other prominent words contributing to make the discourse around #sciacalli trustful are "aiutare" (english: "to help"), "serena" (english: "serene"), "rispetto" (english: "respect") and "veritÃă" (english: "truth"), which further validate a trustful, open-minded and fair perception of the political and emergency debate outlined above. this perception was mixed with negative elements, mainly eliciting fear but also sadness and anger. the jargon of a political debate emerges in the word cloud of fear: "dif-ficoltÃă" (english: "difficulty"), "criminale" (english: "criminal"), "dannati" (english: "scoundrel"), "crollare" (english: "to break down"), "banda" (english: "gang"), "panico" (english: "panic") and "caos" (english: "chaos"). these words indicate that twitter users felt fear directed to specific targets. a speculative explanation for exorcising fear can be finding a scapegoat and then target it with anger. the word cloud of such emotion supports the occurrence of such phenomenon by featuring words like "denuncia" (english: "denouncement"), "colpevoli" (english: "guilty"), "vergogna" (english: "shame"), "combattere" (english: "to fight") and "colpa" (english: "blame"). the above words are reflected also in other emotions like sadness, which features also words like "cadere" (english: "to fall") and "miseria" (english: "misery", "out of grace"). these prominent words in the polarised emotional profile of #sciacalli, suggest that twitter users feared criminal behaviour, possibly related to unwise political debates or improper stockpiling of supplies (as showed by the hashtag analysis). our findings also suggest that the reaction to such fearful state, which also projects sadness about negative economic repercussions, was split into a strong, angry denounce of criminal behaviour and messages of trust for the order promoted by competent organisations and committees. it is interesting to note that, according to ekman's theory of basic emotions [ ] , a combination of sadness and fear can be symptomatic of desperation, which is a critical emotional state for people in the midst of a pandemic-induced lockdown. the same analysis is reported in figure for the social discourse of #italylockdown (top) and #iorestoacasa (bottom). in agreement with the circumplex profiling, for both #italylockdown and #iorestoacasa the intensity of fear is considerably lower than trust. however, when investigated in conjunction with words, the overall emotional profile of #italylockdown appears to be more positive, displaying higher trust and joy and lower sadness, than the emotional profile of #iorestoacasa. although the difference is small, this suggests that hashtags alone are not enough to fully characterise the perception of a conceptual unit, and should always be analysed together with the natural language associated to them. figure : emotional profile and word cloud of the language used in tweets with #sciacalli. words are organised according to the emotion they evoke. font size is larger for words of higher closeness centrality in the word co-occurrence network relative to the hashtag (see methods). every emotion level incompatible with random expectation is highlighted with a check mark. the trust around #italylockdown comes from concepts like "consigli" (english: "tips", "advice"), "compagna" (english: "companion", "partner"), "chiara" (english: "clear"), "abbracci" (english:"hugs") and "canta" (english: "sing"). these words and the positive emotions they elicit suggest that italian users reacted to the early stages of the lockdown with a pervasive sense of commonality and companionship, reacting to the pandemic with externalisations of positive outlooks for the future, e.g. by playing music on the balconies . interestingly, this positive perception co-existed with a more complex and nuanced one. despite the overall positive reaction, in fact, the discourse on #italylockdown also shows fear for the difficult times facing the contagion ("contagi") and the lockdown restrictions ("restrizioni"), and also anger, identifying the current situation as a fierce battle ("battaglia") against the virus. the analysis of anticipation, the emotional state projecting desires and beliefs into the future, shows the emergence of concepts such as "speranza" (english: "hope"), "possibile" (english: "possible") and "domani" (english: "tomorrow"), suggesting a hopeful attitude towards a better future. the social discourse around #iorestoacasa brought to light a similar emotional profile, with a slightly higher fear towards being quarantined at home (quarantena (english: "quarantine"), comando (english: "command", "order", emergenza (english: "emergency"). both surprise and sadness were elicited by the the word "confinamento" (english: "confinement"), which was prominently featured in the network structure arising from the tweets we analysed. in summary, the above emotional profiles of hashtags and words from the , tweets suggest figure : emotional profile and word cloud of the language used in tweets with #italylockdown (top) and #iorestoacasa (bottom). words are organised according to the emotion they evoke. font size is larger for words of higher closeness centrality in the word co-occurrence network relative to the hashtag (see methods). every emotional richness incompatible with random expectation is highlighted with a check mark. that italians reacted to the lockdown measure with: . a fearful denounce of criminal acts with political nuances and sadness/desperation about negative economic repercussions (from #sciacalli); . positive and trustful externalisations of fraternity and affect, combined with hopeful attitudes towards a better future (from #italylockdown and #iorestoacasa); . a mournful concern about the psychological weight of being confined at home, inspiring sadness and disgust towards the health emergency (from #iorestoacasa). in the previous section we showed our findings on how italians perceived the early days of lockdown on social media. but what about their perception of the ultimate cause of such lockdown, covid- ? to better reconstruct the perception of #coronavirus, it is necessary to consider the different contexts where this hashtag occurs. figure displays the reconstruction of the emotional profile of words used in tweets with #coronavirus and either #italylockdown, #sciacalli, or #iorestoacasa. our results suggest that the emotional profiles of language used in these three categories of tweets are different. for example, when considering tweets including #sciacalli, which the previous analysis revealed being influenced by political and social denounces of criminal acts, #coronavirus is perceived with a more polarised fear/trust dichotomy. although #coronavirus was perceived as trustful as random expectations when co-occurring with #sciacalli (z-score: . < . ), it was perceived with significantly higher trust when appearing in tweets with #iorestoacasa (z-score: . > . ) and #italylockdown (z-score: . > . ). to reinforce this picture, the intensity of fear towards #coronavirus was statistically significantly lower than random expectation in the discourse of #iorestoacasa (z-score: - . < - . ) and #italylockdown (z-score: - . < - . ). this difference is prominently reflected in both the circumplex model ( figure , right) and the nrc emotional profile ( figure , left) , although in the latter both emotional intensities are compatible with random expectation. these quantitative comparisons provide data-driven evidence that twitter users perceived the same conceptual entity, i.e. covid- , with a higher trust when associating it to concrete means for hampering pathogen diffusion like lockdown and house confinement, and with a higher fear when denouncing the politics and economics behind the pandemic. however, social distancing, lockdown and house confinement clearly do not have only positive sides. rather, as suggested by our analysis, they bear complex emotional profiles, where sadness, anger and fear towards the current situation and future developments have been prominently expressed by italians on social media. this study delved into the massive information flow of italian social media users in reaction to the declaration of the pandemic status of covid- by who, and the announcement of the nationwide lockdown by the italian government in the first half of march . we explored the emotional profiles of italians during this period by analysing the social discourse around the official lockdown hashtag promoted by the italian government (#iorestoacasa), together with a most trending hashtag of social protest (#sciacalli), and a general hashtag about the lockdown (#italylockdown). the fundamental premise of this work is that social media opens a window on the minds of millions of people [ ] . monitoring social discourse on online platforms provides unprecedented opportunities for understanding how different categories of people react to real world events [ , , ] . here we introduced a new framework, multi-layer co-occurrence networks for emotional profiling (mercurial), which is based on cognitive network science and that allowed us to: (i) quantitatively structure social discourse as a multi-layer network of hashtag-hashtag and word-word co-occurrences in tweets; (ii) identify prominent discourse topics through network metrics backed up by cognitive interpretation [ ] ; (iii) reconstruct and cross-validate the emotional profile attributed to each hashtag or topic of conversation through the emotion lexicon and the circumplex model of affect from social psychology and cognitive neuroscience [ ] . our interdisciplinary framework provides a first step in combining network and cognitive science principles to quantify sentiment for specific topics. our analysis also included extensive robustness checks (e.g. selecting words based on different centrality measures, statistical testing for emotions), further highlighting the potential of the framework. the analysis of concept network centrality identified hashtags of political denounce and protest against irrational panic buying (e.g. face masks and hand sanitiser) around #sciacalli but not in the hashtag networks for #italylockdown and #iorestoacasa. our results also suggest that the social discourse around #sciacalli was further characterised by fear, anger, and trust, whose emotional intensity was significantly stronger than random expectation. we also found that the most prominent concepts eliciting these emotions revolve around social denounce (anger), concern for the collective well-being (fear), and the measures implemented by expert committees and authorities (hope). this interpretation is supported also by plutchik's wheel of emotions [ ] , according to which combinations of anger, disgust and anticipation can be symptoms of aggressiveness and contempt. however, within plutchik's wheel, trust and fear are not in direct opposition. the polarisation of positive/negative emotions observed around #sciacalli might be a direct consequence of a polarisation of different social users with heterogeneous beliefs, which is a phenomenon present in many social systems [ ] but is also strongly present in social media through the creation of echo chambers enforcing specific narratives and discouraging the discussion of opposing views [ , , , , ] . emotional polarisation might therefore be a symptom of a severe lack of social consensus across italian users in the early stages of the lockdown induced by covid- . in social psychology, social consensus is a self-built perception that the beliefs, feelings, and actions of others are analogous to one's own [ ] . destabilising this perception can have detrimental effects such as reducing social commitment towards public good or even lead to a distorted perception of society, favouring selfdistrust and even conditions such as social anxiety [ ] . instead, acts such as singing from the balconies together can reduce fear and enhance self-trust [ ] , as well as promote commitment and social bonding [ ] , which is also an evolutionary response to help coping with a threat, in this case a pandemic, through social consensus. when interpreted under the lens of social psychology, the flash mobs documented by traditional media and identified here as relevant by semantic network analysis for #italylockdown and #iorestoacasa become important means of facing the distress induced by confinement [ , , ]. anger and fear permeated not only #sciacalli but were found, to a lesser extent, also in association with other hashtags such as #iorestoacasa or #italylockdown. recent studies (cf. [ ] ) found that anger and fear can drastically reduce individuals' sense of agency, a subjective experience of being in control of our own actions, linking this behavioural/emotional pattern also to alteration in brain states. in turn, a reduced sense of agency can lead to losing control, potentially committing violent, irrational acts [ ] . consequently, the strong signals of anger and fear detected here represent red flags about a building tension manifested by social users which might contribute to the outbreak of violent acts or end up in serious psychological distress due to lowered self-control. one of the most direct implications of the detected strong signals of fear, anger and sadness is represented by increased violent behaviour. in cognitive psychology, the general aggression model (gam) [ ] is a well-studied model for predicting and understanding violent behaviour as the outcome of a variety of factors, including personality, situational context and the personal internal state of emotion and knowledge. according to gam, feeling emotions of anger in a situation of confinement can strongly promote violent behaviour. in italy, the emotions of anger and anxiety we detected through social media are well reflected in the dramatic rise in reported cases of domestic violence. for instance, the anti-violence centers of d.i.re (donne in rete contro la violenza) reported an anomalous increase of + . % in the number of women looking for help for domestic violence in march in italy . hence, monitoring social media can be insightful about potential tensions mediated and discussed by large populations, a topic in need for further research and with practical prominent repercussions for fighting covid- . as discussed, we found the hashtag #coronavirus to be central across all considered hashtag networks. however, our analysis outlined different emotional nuances of #coronavirus across different networks. in psycholinguistics, contextual valence shifting [ ] is a well-known phenomenon whereby the very same conceptual unit can be perceived wildly differently by people according to its context. this phenomenon suggests the importance of considering words in a contextual manner, by comparison to each other, as it was performed in this study, rather than alone. indeed, contexts can change the meaning and emotional perception of many words in language. we showed here that the same connotation shifting phenomenon [ ] can happen also for hashtags. online users perceived #coronavirus with stronger intensities of trust and lower fear (than random expectation) when using that hashtag in the context of #iorestoacasa and #italylockdown, but not when associated to #sciacalli. this shifting underlines the importance of considering contextual information surrounding a hashtag in order to better interpret its nuanced perception. to this aim, cognitive networks represent a powerful tool, providing quantitative metrics (such as graph distance entropy) that would be otherwise not applicable with mainstream frequency approaches in psycholinguistics. mercurial facilitates a quantitative characterisation of the emotions attributed to hashtags and discourses. nonetheless, it is important to bear in mind that the analysis we conducted relies on some assumptions and limitations. for instance, following previous work [ ] , we built unweighted and undirected networks, neglecting information on how many times hashtags co-occurred. including these weights would be important for detecting communities of hashtags, beyond network centrality. notice that including weights would come at the cost of not being able to use graph distance entropy, which is defined over unweighted networks and was successfully used here for exposing the denounce of panic buying in #sciacalli. another limitation is relative to the emotional profiling performed with the nrc lexicon, in which the same word can elicit multiple emotions. since we measured emotional intensity by counting words eliciting a given emotion (plus the negations, see methods), a consequence was the repetition of the same words across the sectors of the above word clouds. building or exploiting additional data about the predominance of a word in a given emotion would enable us to identify words which are peripheral to a given emotion, reduce repetitions and offer even more detailed emotional profiles. recently, forma mentis networks [ , ] have been introduced as a method to detect the organisation of positive/negative words in the mindsets of different individuals. a similar approach might be followed for emotions in future research. acting upon specific emotions rather than using the circumplex model would also solve another problem, in that the attribution of arousal to individual words is prone to more noise, even in mega-studies, compared to detecting word valence [ ] . another limitation is that emotional profiles might fluctuate over time. the insightful results outlined and discussed here were aggregated over a short time window, thus reducing the impact of aggregation itself. future analyses on longer time windows should adopt time-series for investigating emotional patterns, addressing key issues like non-stationary tweeting patterns over time and statistical scarcity due to tweet crawling (see also [ ] ). the current analysis has focused on aggregated tweets, but previous studies have shown both stable individual and intercultural differences in affect [ ] , especially for dimensions such as arousal. similarly, some emotions are harder to measure than others, which might affect reliability and thus underestimate their contribution. the current approach estimates emotional profiles on the basis of a large set of words, which will reduce some language-specific differences. the collection of currently missing large-scale italian normative datasets for lexical sentiment could further improve the accuracy of the findings. this study approaches the relation between emotions and mental distress mostly from the perspective that attitudes and emotions of the author are conveyed in the linguistic content. however, the emotion profile might also have implications for readers as well, as recent research suggests that even just reading words of strong valence/arousal can have deep somatic and visceral effects, e.g. raising heart beat or promoting involuntary muscle tension [ ] . furthermore, authors and readers participate in an information network, and quantifying which tweets are liked or retweeted depending on the structure of social network can provide further insight on their potential impact [ , , , , ] , which calls for future approaches merging social networks, cognitive networks and emotional profiling. finally, understanding the impact of nuanced emotional appraisals would also benefit from investigating how these are related to behavioural and societal outcomes including the numbers of the contagion (e.g. hospitalisations, death rate, etc.) and compliance with physical distancing [ ] . given the massive attention devoted to the covid- pandemic by social media, monitoring online discourse can offer an insightful thermometer of how individuals discussed and perceived the pandemic and the subsequent lockdown. our mercurial framework offered quantitative readings of the emotional profiles among italian twitter users during early covid- diffusion. the detected emotional signals of political and social denounce, the trust in local authorities, the fear and anger towards the health and economic repercussions, and the positive initiatives of fraternity, all outline a rich picture of emotional reactions from italians. importantly, the psychological interpretation of mercurial's results identified early signals of mental health distress and antisocial behaviour, both linked to violence and relevant for explaining increments in domestic abuse. future research will further explore and consolidate the behavioural implications of online cognitive and emotional profiles, relying on the promising significance of our current results. our cognitive network science approach offers decision-makers the prospect of being able to successfully detect global issues and design timely, data-informed policies. especially under a crisis, when time constraints and pressure prevent even the richest and most organised governments from fully understanding the implications of their choices, an ethical and accurate monitoring of online discourses and emotional profiles constitutes an incredibly powerful support for facing global threats. m.s. acknowledges daniele quercia, nicola perra and andrea baronchelli for stimulating discussion. how to fight an infodemic the covid- social media infodemic assessing the risks of "infodemics" in response to covid- epidemics covid- infodemic: more retweets for science-based information on coronavirus than for false information immediate psychological responses and associated factors during the initial stage of the coronavirus disease (covid- ) epidemic among the general population in china world health organization. mental health during covid- outbreak the immediate mental health impacts of the covid- pandemic among people with or without quarantine managements a longitudinal study on the mental health of general population during the covid- epidemic in china quantifying the effect of sentiment on information diffusion in social media phase transitions in information spreading on structured populations beating the news using social media: the case study of american idol bots increase exposure to negative and inflammatory content in online social systems exposure to opposing views on social media can increase political polarization cognitive network science: a review of research on cognition through the lens of network representations, processes, and dynamics text-mining forma mentis networks reconstruct public perception of the stem gender gap in social media probing the topological properties of complex networks modeling short written texts our twitter profiles, our selves: predicting personality with twitter emotions evoked by common words and phrases: using mechanical turk to create an emotion lexicon semeval- task : affect in tweets measuring emotions in the covid- real world worry dataset a complex network approach to political analysis: application to the brazilian chamber of deputies the emotions. university press of america the nature of emotion: fundamental questions emotional contagion the ripple effect: emotional contagion and its influence on group behavior experimental evidence of massivescale emotional contagion through social networks the rippling dynamics of valenced messages in naturalistic youth chat emotions and social movements: twenty years of theory and research forma mentis networks quantify crucial differences in stem perception between students and experts the circumplex model of affect: an integrative approach to affective neuroscience, cognitive development, and psychopathology largescale network representations of semantics in the mental lexicon forma mentis networks map how nursing and engineering students enhance their mindsets about innovation and health during professional growth from topic networks to distributed cognitive maps: zipfian topic universes in the area of volunteered geographic information distance entropy cartography characterises centrality in complex networks covid- labelled network subgraphs reveal stylistic subtleties in written texts predicting lexical norms: a comparison between a word association model and text-based word co-occurrence models modelling early word acquisition through multiplex lexical networks and machine learning norms of valence, arousal, and dominance for , english lemmas a circumplex model of affect the effects of group singing on mood affect regulation, mentalization and the development of the self forma mentis networks reconstruct how italian high schoolers and international stem experts perceive teachers, students, scientists, and school wordnet: an electronic lexical database the multiplex structure of the mental lexicon influences picture naming in people with aphasia recursive patterns in online echo chambers on the perception of social consensus the ice-breaker effect: singing mediates fast social bonding i just lost it! fear and anger reduce the sense of agency: a study using intentional binding the general aggression model: theoretical extensions to violence contextual valence shifters obtaining reliable human ratings of valence, arousal, and dominance for , english words the relation between valence and arousal in subjective experience varies with personality and culture somatic and visceral effects of word valence, arousal and concreteness in a continuum lexical space retweeting for covid- : consensus building, information sharing, dissent, and lockdown life the ids of the tweets analysed in this study are available on the open science foundation repository: https://osf.io/jy kz/. key: cord- -t l px authors: benetti, elisa; tita, rossella; spiga, ottavia; ciolfi, andrea; birolo, giovanni; bruselles, alessandro; doddato, gabriella; giliberti, annarita; marconi, caterina; musacchia, francesco; pippucci, tommaso; torella, annalaura; trezza, alfonso; valentino, floriana; baldassarri, margherita; brusco, alfredo; asselta, rosanna; bruttini, mirella; furini, simone; seri, marco; nigro, vincenzo; matullo, giuseppe; tartaglia, marco; mari, francesca; renieri, alessandra; pinto, anna maria title: ace gene variants may underlie interindividual variability and susceptibility to covid- in the italian population date: - - journal: eur j hum genet doi: . /s - - -z sha: doc_id: cord_uid: t l px in december , an initial cluster of interstitial bilateral pneumonia emerged in wuhan, china. a human-to-human transmission was assumed and a previously unrecognized entity, termed coronavirus disease- (covid- ) due to a novel coronavirus (sars-cov- ) was described. the infection has rapidly spread out all over the world and italy has been the first european country experiencing the endemic wave with unexpected clinical severity in comparison with asian countries. it has been shown that sars-cov- utilizes angiotensin converting enzyme (ace ) as host receptor and host proteases for cell surface binding and internalization. thus, a predisposing genetic background can give reason for interindividual disease susceptibility and/or severity. taking advantage of the network of italian genomes (nig), here we mined whole-exome sequencing data of italian control individuals from five different centers looking for ace variants. a number of variants with a potential impact on protein stability were identified. among these, three more common missense changes, p.(asn asp), p.(lys arg), and p.(gly arg) were predicted to interfere with protein structure and stabilization. rare variants likely interfering with the internalization process, namely p.(leu val) and p.(pro his), predicted to interfere with sars-cov- spike protein binding, were also observed. comparison of ace wes data between a cohort of patients and controls allowed identifying a statistically significant (p value < . ) higher allelic variability in controls compared with patients. these findings suggest that a predisposing genetic background may contribute to the observed interindividual clinical variability associated with covid- , allowing an evidence-based risk assessment leading to personalized preventive measures and therapeutic options. in december , a new infectious respiratory disease emerged in wuhan, hubei province, china [ - ] . an initial cluster of infections likely due to animal-to-human transmission was rapidly followed by a human-to-human transmission [ ] . the disease was recognized to be caused by a novel coronavirus (sars-cov- ) and termed coronavirus disease- . the infection spread within china and all over the world, and it has been declared as pandemic by the world health organization (who) on nd march . the symptoms of covid- range from fever, dry cough, fatigue, congestion, sore throat, and diarrhea to severe interstitial bilateral pneumonia with a ground-glass image at the ct scan. while recent studies provide evidence of a high number of asymptomatic or paucisymptomatic patients who represent the main reservoir for the infection progression, the severe cases can rapidly evolve towards a respiratory distress syndrome which can be lethal [ ] . although age and comorbidity have been described as the members main determinants of disease progression towards severe respiratory distress, the high variation in clinical severity among middle-age adults and children would likely suggest a strong role of the host genetic asset. a high sequence homology has been shown between sars-associated coronavirus (sars-cov) and sars-cov- [ ] . recent studies modeled the spike protein to identify the receptor for sars-cov- and indicated that angiotensin converting enzyme (ace ) is the receptor for this novel coronavirus [ , ] . zhou et al. conducted virus infectivity studies and showed that ace is essential for sars-cov- to enter hela cells [ ] . although the binding strength between sars-cov- and ace is weaker than that between sars-cov and ace , it is considered as much high as threshold necessary for virus infection. the spike glycoprotein (s-protein), a trimeric glycoprotein in the virion surface (giving the name of crown -corona in latin-), mediates receptor recognition throughout its receptor binding domain (rbd) and membrane fusion [ , ] . based on recent reports, sars-cov- protein binds to ace through leu , phe , gln , ala , and tyr . it has been postulated that residues , , , , , and of the ace receptor map to the surface of the protein interacting with sars-cov- spike protein [ ] , as previously documented for sars-cov. following interaction, cleavage of the c-terminal segment of ace by proteases, such as transmembrane protease serine (tmprss ), enhances the spike protein-driven viral entry [ , ] . thus, it is possible, in principle, that genetic variability of the ace receptor is one of the elements modulating virion intake and thus disease severity. ace is located on chromosome x. although it is one of the genes escaping x inactivation several lines of evidence suggest that a different degree of x-chromosome inactivation (xci) is present in distinct tissues [ ] . taking advantage of the network of italian genomes (nig), a consortium established to generate a public database (nig-db) containing aggregate variant frequencies data for the italian population (http://www.nig.cineca.it/), here we describe the genetic variation of ace in the italian population, one of the newly affected countries by the sars-cov- outbreak causing covid- . three common c. a>g p.(asn asp), c. a>g p.(lys arg), and c. g>a p.(gly arg) variants and rare missense variants were identified, of which had not previously been reported in public databases. we show that p.(asn asp), which lies in a residue located close to the cleavage sequence of tmprss , likely affects the cleavagedependent virion intake. along with the other two common variants, this substitution is represented in the italian and european populations but is extremely rare in the asian population. we also show that two rare variants, namely, c. c>g p.(leu val) and c. c>a p.(pro his) are predicted to cause conformational changes impacting rbd interaction. as the uncertainty regarding the transmissibility and severity of disease rise, we believe that a deeper characterization of the host genetics and functional characterization of variants may help not only in understanding the pathophysiology of the disease but also in envisaging risk assessment. the work has been realized in the context of the nig, with the contribution of centers: azienda ospedaliera universitaria senese, azienda ospedaliera universitaria policlinico sant'orsola-malpighi di bologna, città della salute e della scienza di torino, università della campania "luigi vanvitelli", ospedale pediatrico bambino gesù. the nig (http://www.nig.cineca.it/) aim is to create a shared database (nig-db) containing data from nucleic acids sequencing of italian subjects. this database allows defining an italian reference genome for the identification of genes responsible for genetic diseases or italian population susceptibility to complex disorders and for the detection of genetic variants responsible for interindividual differences in disease progression ad /or drug response among the italian population. individuals coming to our centers were offered to participate to the nig study and blood withdrawal was performed upon informed consent. individuals provided signed informed consents at each participating center for whole-exome sequencing analysis (wes), and clinical and molecular data storage and usage. all subjects were unrelated, healthy, and of italian ancestry. italian origin was ascertained asking for parents and grandparents origin. dna has been stored in the telethon network of genetic biobanks (project no. gtb ), funded by telethon italy. the study was consistent with institutional guidelines and approved by the university hospital (azienda ospedaliera universitaria senese) ethical committee, siena, italy (prot n. , dated march , ). written informed consent was obtained from all patients and controls. peripheral blood samples in edta-containing tubes and detailed clinical data were collected. all these data were inserted in a section of the established and certified biobank and registry of the medical genetics unit of the hospital dedicated to covid- . the cohort of covid- patients consists of individuals out of whom females and males belonging to the gen-covid multicenter study ([ ] , late breaking abstract eshg . virtual conference "wes profiling of covid- "). the cohort of controls consists of italian individuals ( males and females). all patients are of italian ethnicity. the median age is years (range - ): median age for women years and for males years. the population was clustered into four qualitative severity groups depending on the respiratory impairment and the need for ventilation: high care intensity group (those requiring invasive ventilation), intermediate care intensity group (those requiring noninvasive ventilation i.e., cpap and bipap, and high-flows oxygen therapy), low care intensity group (those requiring conventional oxygen therapy) and very low care intensity group (those not requiring oxygen therapy). targeted enrichment and massively parallel sequencing were performed on genomic dna extracted from circulating leukocytes of individuals. genomic dna was extracted from peripheral blood samples using standard procedures. exome capture was carried out using sur-eselect human all exon v /v /v /v ( broad institute, harvard, usa). alignment of raw reads against reference genome hg , variant calling and annotation were attained using in-house pipelines [ ] [ ] [ ] which take advantage of the gatk best practices workflow [ ] and of annovar, vep [ , ] . the genome aggregation database gnomad (https://gnomad.broadinstitute.org/) was used to assess allele frequency for each variant among different populations. the mean depth of coverage of each ace exon in all participants was ×. variants with a depth of coverage lower that × were filtered out according to ashg guidelines for germline variants [ ] . the the structure of native human angiotensin converting enzyme-related carboxypeptidase (ace ) was downloaded from protein data bank (https://www.rcsb.org/) (pdb id code r ) [ ] . the duet program [ ] was used to predict the possible effect of amino acids substitutions on the protein structure and function, based on the use of machine-learning algorithms exploiting the threedimensional structure to quantitatively predict the effects of residue substitutions on protein functionality. molecular dynamics (md) simulations of wild-type and variant ace proteins were carried out in gromacs . [ ] to calculate root mean square deviation (rmsd) to define structural stability. the graphs were plotted by the xmgrace software [ ] . md simulations were performed using a high parallel computing infrastructure (hpcs) with cpu within different nodes, t of ram, t hard disk partition size, and six nvidia tesla gpu with cuda support. pymol . was used as a molecular graphic interface. the protein structures were solvated in a triclinic box filled with tip p water molecules and na + /cl − ions were added to neutralize the system. the whole systems were then minimized with a maximal force tolerance of kj mol − nm − using the steepest descendent algorithm. the optimized systems were gradually heated to k in ns in the nvt ensemble, followed by ns equilibration in the npt ensemble at atm and k, using the v-rescale thermostat and berendsen barostat [ , ] . subsequently, a further ns md simulations were performed for data analysis. the extent of variability along the entire ace coding sequence and flanking intronic stretches was assessed using italian wes, out of which males and females which sum up to alleles. identified variants and predicted effects on protein stability are summarized in tables , , and table s , and represented in exomes. in addition to these variants, rare missense variants were identified, out of which ten had not previously been reported in gnomad database and nine truncating variants that had not been reported in gnomad database (table and supplementary table ). out of these variants, two fall in the neck domain, which is essential for dimerization and one in the intracellular domain. many of them truncate the protein in different positions of the protease domain embedded in the extracellular domain, which contains the receptor binding site for sars-cov- . only three truncating variants have been previously described for ace likely due to a low-tolerance for loss-of-function variants. in line with this evidence, all these variants were very rare and no homozygous females were detected for the identified variants. three missense changes c. t>c p. (fig. a) , the c. c>a p.(pro his) and c. c>g p.(leu val) variants show a difference in comparison with the native protein with a gradual increase in rmsd value, which stabilizes at an average of . nm (fig. a) . finally, the c. g>a p. (gly arg) shows a bigger difference with a higher increase in rmsd value, which stabilizes at an average of . nm (fig. a) . structural analysis between wt and mutant c. t>c p.(val ala) md simulations showed that the c. t>c p.(val ala) forms a hydrophobic center together with leu , leu , and phe with minimum differences in protein rearrangements when the residue is mutated in ala as reported in fig. and supplementary video s . the c. a>g p.(lys arg) is located at the n-terminus and the sidechain engages a hydrogen bond with asn thus determining a minimal destabilizing predicted effect as shown in table during md simulations, we have also investigated the surrounding region of ace wt and previously selected variants by calculating change in solvent accessibility surface area (sasa). differences in average sasa value would suggest for the native protein a wider surface exposed to solvent and subsequently a different ability to interact with spike sars-cov- in comparison with the studied variants (fig. b) . in order to shed light on the role of ace variants on interindividual variability and susceptibility to covid- in italian population we performed wes analysis on a cohort of patients and controls who agreed in participating to the study (see "materials and methods"). data analysis of ace variants identified a different distribution of variants in controls compared with patients ( fig. ) according to recent reports, ace is essential for sars-cov- to enter cells. recent single-cell rna studies have also shown that ace is expressed in human lung cells [ ] . the majority of ace -expressing cells are alveolar type cells. other ace -expressing cells include alveolar type cells, airway epithelial cells, fibroblasts, endothelial cells, and macrophages although their ace -expressing cell ratio is low and variable among individuals. the expression and distribution of the ace receptor can thus justify the route of infection and the main localization at the alveolar level. although the different density of ace receptors in the upper respiratory tract among individuals can partially give reason for the clinical variability, which ( Å) y , l , r , p , a , s , l , f , h , n , s y , e , v , n , g , v , y , d , y , p , t h , h , e , m , g , h , i , a , y , f , h , v , g , m l , p , a , h , l , l , w , w , v , v , e , p , w y , c , d , p , a , s , g , r , l , y , w , w , m a , y , v , r , a , k , l , duet program results that display predicted change in folding free energy upon ace missense variant (ΔΔg in kcal/mol). in the first three columns are reported single missense variants with specific position on ace protein. the residues in the first column highlighted in gray are involved in n-glycosylation pattern nxt/s, therefore those missense variants determine the loss glycosylation of asparagine and , respectively. in the fourth column is reported ΔΔg analysis predict effects of missense variants on protein stability using an integrated computational approach. the column "interaction network around ( Å)" shows for each single missense variant the residues around Å. in this column, we highlight in green residues involved in spike sars-cov protein interaction, in yellow residues involved in zinc coordination and finally in magenta residues of asn involved in n-glycosylation. the last column defines the outcome of protein stability for each single missense variant. an increasing negative value for the ΔΔg is correlated with a higher destabilizing effect, while a positive value is associated with a variant predicted as stabilizing. ranges from asymptomatic/paucisymptomatic patients to severely affected ones, it could not be the only reason for such variability. in addition, recent works did not observe significantly different viral loads in nasal swabs between symptomatic and asymptomatic patients [ ] . italy has been the first european country that experienced the covid- outbreak with a rapid increase in the positive cases in a very short-time period and a morbidity and lethality (~ %) definitely higher in comparison with asian countries, such as china ( %) and south korea ( . %) [ ] . these considerations raise the possibility of a predisposing genetic background accounting for or contributing to the wide interindividual clinical variability, as well for the differential morbidity and lethality observed among different countries, population awareness, and constrictive measures apart. we integrated genomic wes data produced by five italian centers (siena, naples, turin, bologna, and rome) interconnected by the nig in the attempt to identify variation encompassing the ace gene, which could account for a difference in sars-cov- spike binding affinity, processing, or internalization. previous studies showed that the residues near lysine , and tyrosine , - , and - in human ace are important for the binding of s-protein to coronavirus [ ] . in line with previous reports [ ] , we did not find polymorphism or rare variants in these residues in the italian population. however, we identified three variants namely c. a>g p.(asn asp), c. c>a p.(pro his), and c. c>g p.(leu val), one of which polymorphic c. a>g p.(asn asp), moderately expressed in the italian and european non-finnish populations and with a very low allele frequency or not occurring in the eastern asia population. these variants which surround residual essentials for the sars-cov- spike protein binding were predicted to likely affect the cleavage-dependent virion intake, such as the polymorphic c. a>g p.(asn asp) (allele frequency . ) which lies four amino acids from the cleavage sequence of tmprss or to have a substantial impact on protein structure and spike protein interaction by md simulation (fig. a) . the relatively frequent c. g>a p.(gly arg) (allele frequency . , / individuals) was predicted to confer a wide flexibility to the region because of the ability to engage different interactions with the nearby amino acid residues. along with these more common variants we also identified very rare variants such as the c. c>a p.(pro his) and the c. finnish european population, that could give reason for a different affinity for the sars-cov- spike protein (figs. , a and supplementary video s ). interestingly all the studied variants affect residues highly conserved among species (supplementary fig. s ). given their rarity in other populations, we cannot exclude that these variants can fig. structure superimposition snapshot between wild-type protein and variant proteins. a root mean square deviation (rmsd) trends for the backbone of ace wt (black line) and some selected variants (colored lines, see legend) during ns of simulation. the molecular dynamics simulation shows a good stability for all systems with exception of g r mutants. rmsd is a parameter used to define the stability of an element. wild type shows a steady course in the rmsd value, stabilizing at an average of . nm, while, the g r variant shows a gradual increase in rmsd value, stabilizing at an average of . nm. b sasa graphical representation of ace wt (black line) and ace variants (colored lines, see legend). partially account for the clinical outcome observed in the italian population. wes data generated from a wide cohort of covid- italian patients revealed a statistically significant (p < , ) higher allelic heterogeneity for ace in controls compared with patients with a higher chance to find at least one ace variant in the cohort of controls compared with the cohort of patients. therefore, it is plausible to think that the effect of allelic variability on ace conformation would at least partially account for the interindividual clinical differences and likely modulate clinical severity. this finding reinforces the hypothesis that at least some of the identified variants or the cumulative effect of few of them confer a different susceptibility to virus cell entry and consequently to disease onset and progression. we cannot exclude that also silent variants such as the c. g>a (p. val val) with no effect on the protein could play a role because of an unpredictable impact at a posttranscriptional level. notably, morbidity and lethality have been reported definitely higher in men compared with women (~ % vs. %, th march iss report). although several parameters have been brought to case to explain this difference, i.e., smoking, differences in ace localization and/or density in alveolar cells, hormonal asset, it is noteworthy that ace is located on chromosome x and that given the low allele frequency of the identified variants the rate of homozygous women is extremely low (see results section). the xci is incomplete in humans and some genes show a degree of xci escape which vary between individuals and tissues [ ] . ace is one of the genes escaping x inactivation, but it belongs to a subgroup of x-chromosome genes showing a higher expression in men in several tissues thus mostly suggesting that ace gene xci is present although different in distinct tissues [ ] . therefore, the impact of x inactivation on the alternate expression of the two alleles would guarantee, in the affected tissues, a heterogeneous population of ace molecules, some of which protective towards the infection until the point of a complete or almost complete protection in the case of a x inactivation skewed towards the less sars-cov- -binding prone allele. this hypothesis would justify the high rate of asymptomatic or paucisymptomatic patients. however, the presented data does not allow to confirm a clear cause-effect relationship and, since most of the identified variants have very low frequencies, further functional studies are needed to validate these results. ace is definitely one of the main molecules whose genetic heterogeneity can modulate infection and disease progression; however, a deeper characterization of the host genetics and functional variants in other pathwayrelated genes may help in understanding the pathophysiology of the disease opening up the way to a stratified risk assessment and to tailored preventive measures and treatments. questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. conflict of interest the authors declare that they have no conflict of interest. ethical approval the nig study was approved by the ethical committee of each center, prot name nig prot n _ , approved ceavse on . . n / . the covid- patients study was approved by the university hospital (azienda ospedaliera universitaria senese) ethical committee, siena, italy (prot n. , dated march , ). publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. open access this article is licensed under a creative commons attribution . international license, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the creative commons license, and indicate if changes were made. the images or other third party material in this article are included in the article's creative commons license, unless indicated otherwise in a credit line to the material. if material is not included in the article's creative commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. to view a copy of this license, visit http://creativecommons. org/licenses/by/ . /. department of specialized and internal medicine, pneumology unit and utip, san donato hospital department of health sciences, clinic of infectious diseases department of molecular medicine, university of padova, padova, italy; department of infectious and tropical diseases, university of brescia and asst spedali civili hospital clinical features of patients infected with novel coronavirus in wuhan a novel coronavirus outbreak of global health concern a familial cluster of pneumonia associated with the novel coronavirus indicating person-to-person transmission: a study of a family cluster asymptomatic carrier state, acute respiratory disease, and pneumonia due to severe acute respiratory syndrome coronavirus (sars-cov- ): facts and myths networkbased drug repurposing for novel coronavirus -ncov/sars-cov- pathological findings of covid- associated with acute respiratory distress syndrome viral dynamics in mild and severe cases of covid- a pneumonia outbreak associated with a new coronavirus of probable bat origin cryo-em structure of the -ncov spike in the prefusion conformation structural basis for the recognition of the sars-cov- by full-length human ace receptor and viral determinants of sars-coronavirus adaptation to human ace a transmembrane serine protease is linked to the severe acute respiratory syndrome coronavirus receptor and activates virus entry tmprss and adam cleave ace differentially and only proteolysis by tmprss augments entry driven by the severe acute respiratory syndrome coronavirus spike protein landscape of x chromosome inactivation across human tissues clinical and molecular characterization of covid- hospitalized patients loss of smpd causes a developmental disorder characterized by microcephaly and congenital arthrogryposis ssbp mutations cause mtdna depletion underlying a complex optic atrophy disorder aberrant function of the c-terminal tail of hist h e accelerates cellular senescence and causes premature aging the genome analysis toolkit: a mapreduce framework for analyzing next-generation dna sequencing data annovar: functional annotation of genetic variants from high-throughput sequencing data the ensembl variant effect predictor acmg clinical laboratory standards for nextgeneration sequencing ace x-ray structures reveal a large hinge-bending motion important for inhibitor binding and catalysis duet: a server for predicting effects of mutations on protein stability using an integrated computational approach gromacs: high performance molecular simulations through multi-level parallelism from laptops to supercomputers. soft-warex version . . canonical sampling through velocity rescaling molecular dynamics with coupling to an external bath single-cell rna expression profiling of ace , the putative receptor of wuhan the early phase of the covid- outbreak in lombardy how deadly is covid- ? a rigorous analysis of excess mortality and agedependent fatality rates in italy comparative genetic analysis of the novel coronavirus ( -ncov/sars-cov- ) receptor ace in different populations x-inactivation profile reveals extensive variability in x-linked gene expression in females • francesco musacchia • tommaso pippucci • annalaura torella • alfonso trezza • floriana valentino • margherita baldassarri • alfredo brusco , • rosanna asselta , • mirella bruttini , • simone furini • marco seri , • vincenzo nigro , • giuseppe matullo , • marco tartaglia • francesca mari , • gen-covid multicenter study • alessandra renieri author contributions eb, rt, os, amp, and ar have made substantial contributions to conception and design, acquisition of data, analysis and interpretation of data, and have been involved in drafting the paper. ra, gb, abruselles, abrusco, gd, ag, fm, tp, atorella, atrezza, and fv has made substantial contributions to acquisition and analysis of the data. mbaldassarri, mbruttini, ac, sf, fm, gm, vn, ms, and mt have made substantial contributions to interpretation of data and clinical evaluation. all authors have been involved in drafting the paper; have given final approval of the version to be published and agree to be accountable for all aspects of the work in ensuring that key: cord- -sonsrf b authors: franchini, linda; ragone, nicola; seghi, federico; barbini, barbara; colombo, cristina title: mental health services for mood disorder outpatients in milan during covid- outbreak: the experience of the health care providers at san raffaele hospital date: - - journal: psychiatry res doi: . /j.psychres. . sha: doc_id: cord_uid: sonsrf b the spread of covid- throughout italy, particularly lombardy, led to adopt quarantine measures, known to exacerbate pre-existing psychiatric conditions. we described a telephone-based surveillance on euthymic mood disorder outpatients in milan by a non-standardized survey to evaluate reactions to lockdown measures and the presence of quarantine stressors. frustration was the most represented quarantine stressor. being jobless was significantly related to the presence of frustration, somatization, increased alertness, psychic anxiety and low mood; younger age to the presence of psychic anxiety, alertness and financial concerns. no recurrences were observed at the time of writing. the rapid spread of the novel severe acute respiratory syndrome coronavirus- (sars-cov- ) throughout the italian country resulted in a tragic epidemic that led the italian government to impose emergency measures such as quarantine and social isolation. quarantine is an unpleasant experience for those who undergo it and boredom can occasionally create dramatic effects, sometimes leading to suicide (barbisch et al., ) . if domestic lockdown and social isolation have proved to be quite effective as physical containment for infections, it might prove risky from a psychological point of view due to the onset of distressing feelings such as uncertainty, fear, and despair (poletti et al., ) . if environmental stressors may have negative consequences on individuals without a psychiatric history, their effect on patients with pre-existing mental disorders may become even more detrimental, eventually resulting in relapses or worsening of their mental condition (hawryluck et al., ) . considering that this is the first experience of such a collective social trauma in italy, no analyses have focused on the potentially negative role of quarantine restrictions on mood disorder patients yet. lombardy is the italian region most severely affected by covid- and, at the time of writing, the total number of positive cases in milan reached , (data from civil protection); nationwide restrictions on travels and quarantine have resulted in difficult and impractical attending of regular psychiatric visits. to minimize the risk of infection arising from travels and social interactions, the italian ministry of health indicated that outpatients in clinically stable conditions should be monitored using telemedicine approaches (e.g., phone, smartphone, laptop, or other devices) for the duration of the italian lockdown (started from th march ). among these patients, it has been requested to evaluate the following parameters: ) their physical conditions; ) their knowledge of the pandemic situation; ) their adherence and tolerance to quarantine restrictions; ) occurrence of subclinical syndromes or new episodes of their psychiatric illness, to better understand their needs and concerns. the present report aims at describing the telephone-based mental illness surveillance on mood disorder patients attending the mood disorder outpatient center at san raffaele hospital in milan, with the aid of a non-standardized survey to evaluate reactions to lockdown measures. the subjects affected by major depressive disorder (mdd) or bipolar disorder (bp) who had a scheduled psychiatric visit during the italian lockdown period ( march- april ) at our center, were contacted by phone to assess their clinical conditions. the inclusion criteria for our study were being in euthymic conditions according to the clinical judgment of the referring psychiatrist, having previously signed a written informed consent to participate in the ongoing observational study at our hospital approved by the local ethical committee, and verbally agreeing to answer our questionnaire for clinical and research purposes. scheduled patients underwent a telephone-based consultation by their reference psychiatrist concerning their physical (temperature > . , cough, cold, sore throat, dyspnea) and psychiatric conditions. the consultation was integrated with a short, non-standardized instrument created in the emergency of the current pandemic situation. the tool consisted in a questionnaire assessing the presence of items regarding emotional stressors (fear of infection, frustration due to restrictions, adequacy of supplies, adequacy of information, financial concerns), unpleasant experiences during the lockdown (sleep disturbances, mood or anxiety symptoms, increased dosage of anxiolytics and/or hypnotics) (brooks et al., ) , and satisfaction with the questionnaire itself. each item was rated as present, absent, or not answered. if present, the severity was rated on a likert scale. answers were reported in the medical record and were collected together with demographic and clinical data of interest (age, sex, duration of euthymia, duration of maintenance treatment, household composition, employment status, close contact with people affected by covid- , and personal history of covid- ). using the stat-soft statistica . bivariate correlation analyses have been carried out to evaluate the association between clinical (diagnosis, euthymia and maintenance treatment duration), and demographic variables (age, sex , living alone, employment status) to quarantine stressor in order to identify predictive variables for multiple regression. at the time of writing quarantine has been going on for days. one-hundred and one patients were screened for enrolment, all of which met the inclusion criteria. no missing data were reported. our sample showed a mean age of . ± yrs, and an average period of euthymia of months, with a mean long-term treatment period of . ± years. fifteen out of patients ( . %) experienced covid- among their relatives whereas none of them reported to have personally suffered from covid- . furthermore, . % lived alone and . % were unemployed. table reports clinical and demographic characteristics of the sample, the percentage of each stressor reported and their significant correlations. among our sample, % of patients reported frustration due to restrictions, . % fears about infection, . % financial concerns, . % psychic anxiety, . % low mood, . % somatization, . % increased alertness, . % insomnia, . % self-medication with sedatives, . % inadequate supplies. current age and employment status were the variable significantly correlated to quarantine stressors. according to multiple regression analysis, employment status was negatively significantly associated with frustration (beta - . ), somatization (beta - . ), alertness (beta - . ), psychic anxiety (beta- . ) and low mood (beta - . ) ; age was negatively significantly associated with psychic anxiety (beta - . ), financial concerns (beta - . ) and alertness (beta- . ). no patient reported symptoms of early mood disorder recurrence. all patients showed appreciation for this type of tele-monitoring. the purpose of our brief report is to describe the telephone-based mental illness surveillance on the quarantine effects in a sample of euthymic mood disorder patients who could not attend their scheduled psychiatric visit due to lockdown measures. the potential negative role of quarantine restriction on mood disorder patients depends on the high susceptibility to lifestyle-mediated disturbances of biological and social rhythms (wang et al., ) . in this sense, social isolation, travel restrictions, and home confinement could predispose to an increased risk of affective recurrences. however, we have currently observed no recurrence of illness, supporting the crucial role of a sustained euthymia, reachable with a long-term maintenance treatment in mood disorders (cava et al., ) . frustration due to restrictions was the most represented stressor in our sample. this finding is not surprising in a long-lasting clinically stable sample and frustration could represent a normal and understandable response to an atypical and stressful situation such as the covid- . in fact, loss of usual routine and reduced social and physical contact have been linked to emotional responses, such as frustration, in a non-clinical population (bai y et al., ) . on the other hand, presence of frustration could be more harmful in mood disorder patients, considering that may lead to a worsening of the preexisting psychiatric disorders (yao h, et al., , lima ck et al., . in our sample, we found significant associations between unemployment and younger age as demographic variables and several quarantine stressors, including frustration. in particular being jobless was significantly related to the presence of frustration, somatization, psychic anxiety and increased alertness, while younger age to the presence of psychic anxiety and financial concerns. lower household income and financial loss are known to be post-quarantine stressors and having a history of psychiatric illness was associated with experiencing anxiety and anger - months after quarantine (jeong h et al., ) . in line with our results, our unemployed and younger patient might be at a higher risk for new recurrences and, therefore, they might benefit from a careful follow-up. we acknowledge that a non-standardized survey should be considered as a major limit, impeding the generalization of our observations. nevertheless, considering the lack of standardized tools and the responsibility to maintain operational psychiatric care during this exceptional gravity situation, our experience can provide a useful contribution. we found very low rates of concerns regarding having inadequate information or supplies, suggesting that the emergency measures taken by the italian authorities were appropriate and adequately released to the population. finally, our experience suggested that telemedicine, although infrequently practiced in the italian psychiatric setting until recently, might be considered as a valid ally against the clinical difficulties encountered during the covid- pandemic, supporting patients in their regular psychiatric monitoring. is there a case for quarantine? perspectives from sars to ebola the psychological impact of quarantine and how to reduce it: rapid review of the evidence the experience of quarantine for individuals affects by sars in toronto sars control and psychological effects of quarantine mental health status of people isolated due to middle east respiratory syndrome impact of early and recent stress on white matter microstructure in major depressive disorder circadian rhythm disruption and mental health mar ). immediate psychological responses and associated factors during the initial stage of the coronavirus disease (covid- ) epidemic among the general population in china * r= -. p< . ; ° r= -. p< . ; ** r= -. p< . ; *** r=-. p< . ; § r= . p< . ;**** r=-. p< . ; °° r= -. p< . ; °°° r= -. p< . ; ç r= - p< . key: cord- - km authors: iosa, marco; paolucci, stefano; morone, giovanni title: covid- : a dynamic analysis of fatality risk in italy date: - - journal: front med (lausanne) doi: . /fmed. . sha: doc_id: cord_uid: km italy was the second country in the world to face a wide epidemic of covid- after china. the ratio of the number of fatalities to the number of cases (case fatality ratio, cfr) recorded in italy was surprisingly high and increased in the month of march. the older mean age of population, the changes in testing policy, and the methodological computation of cfr were previously reported as possible explanations for the incremental trend of cfr, a parameter theoretically expected to be constant. in this brief report, the official data provided by the italian ministry of health were analyzed using fitting models and the linear fit method approach. this last methodology allowed us to reach two findings. the trend of the number of deaths followed a – -day delay of positive cases. this delay was not compatible with a biological course of covid- but was compatible with a health management explanation. the second finding is that the italian number of deaths did not increase linearly with the number of positive cases, but their relationship could be modeled by a second-order polynomial function. the high number of positive cases might have a direct and an indirect effect on the number of deaths, the latter being related to the overwhelmed bed capacity of intensive care units. the severe acute respiratory syndrome coronavirus (sars-cov- ) has developed worldwide into a pandemic ( , ) . there is a wide clinical debate on the different strategies required to minimize deaths and a political one on the economic impact of those strategies, but minimizing both fatalities and cost is proving to be quite difficult ( ) . in china, the epidemic seemed to be effectively contained by quarantine, social distancing, and the isolation of the infected population. conversely, on march , the spread of corona virus disease in italy largely increased despite the restrictions put in place by the government. in italy, the first case of sars-cov- was diagnosed in lombardy region on the th of february ( ). only month later, the number of deaths due to the covid- recorded in italy was the highest globally, even higher than that documented in china, and this was only recently exceeded by united states. many different mathematical models have been proposed to help governments to decide on what health policies they should follow. some models have been based on an exponential curve for fitting the number of infected cases and deaths. although mass media reported this initial exponential trend, it was conceivable to expect a deviation from that-rather than a plateau-followed by a progressive decrement, according to a bell-shaped curve ( ) . a recent study based on data recorded up to the th of march hypothesized for italy a trend similar to that observed in the hubei province in china, and it predicted a peak of cases at around the th of april ( ). anderson et al. ( ) have developed an illustrative simulation of the transmission model of covid- , showing that social distancing could flatten the curve of positive case frequency, retarding and reducing the peak of the curve estimation in case of no social restrictions ( ) . this theoretical modeling, reported by many mass media, suggested that a delay in contagions may reduce the number of deaths. this hypothesis was based on the idea that the number of beds in intensive care units (icus) could be sufficient only for a flattened curve of positive case frequency. otherwise, if the number of severely affected patients exceeded that of beds in icus, the number of deaths could dramatically increase. at the beginning, the italian case fatality rate (cfr) seemed to be similar to that of china, initially fixed at . % ( ) . the case fatality rate is the ratio of deaths caused by a given disease calculated on the total number of cases that the disease generated in a specific time period ( ) . updated with the new data from the th of march, the italian cfr exceeded the %. in a comparison report, a possible explanation was provided by the higher mean age of the italian population compared with the chinese one ( ) . but this may be only a partial explain of the difference in the case fatality rate of covid- in italy with respect to china. an older population, such as the italian one, may suffer from comorbidities, which increase the risk of death and hence the cfr ( ). however, the italian cfr has been higher than the chinese one even after being corrected for age: in patients older than years, cfr was . % in italy, and . % in china ( ) . furthermore, the mean older age of people did not explain the incremental trend for cfr within the italian population during the month of march. the authors of that research suggested also other possible explanations for the high cfr, mainly related to the methodological differences in case recording and case testing ( ) . in the early phase of the epidemic, italy carried out an extensive testing strategy by collecting swabs of both symptomatic and asymptomatic contacts of the infected patients, as was done in china. then, the italian ministry of health issued more stringent testing policies, prioritizing tests for patients with severe clinical symptoms who required hospitalization. this could have caused an increase in the computed value of cfr for the underestimation of the number of the asymptomatic or mildly affected patients for whom the tests were often not administered. it means that, in italy more than in other countries, the full denominator of cfr remains unknown because asymptomatic cases or patients with mild symptoms might not be tested and hence will not be identified. recent studies have faced the problem of a correct computation and interpretation of cfr related to covid- . one of them suggested, in this dynamic situation, to estimate the cfr as the number of deaths on the number of infected patients evaluated weeks before ( ) . this delay was suggested to be helpful for taking into account the incubation period and the median time from onset of symptoms to death ( , ) . a recent report, using a delay-adjusted cfr of . % (computed from a previous large study conducted in china), estimated that less than the % of the contagions in italy were actually diagnosed . however, it is noteworthy that the italian policy change on tests occurred on the th of february when the italian cfr was . % and had then continued to increase, hitting % only month later on the th of march. italy was the first western country with a wide spread of covid- , and it could be important, for other countries, to analyze in depth the italian case. the italian cfr increased day by day, despite, from a theoretical point of view, the cfr being expected to be constant ( ) . a constant cfr means that the number of deaths proportionally (linearly) increase with the number of cases. the above studies seemed to suggest that cfr was only miscomputed because the more severe cases the clinicians need to bring assistance to the less time they have to test non-severe cases, causing an apparent increase of cfr ( ) . in the present study, mathematical models were used to test if the high italian cfr was only apparent because it was related to an underestimation of positive cases or if it represents a real increment of covid- lethality, maybe related to the difficulties of the health national system to manage many cases in a short period and in a small region as occurred in the north of italy. these possibilities have led to the different theoretical scenarios depicted in figure . the cfr computed day by day could be high due to the need to take into account a biological delay of about days between deaths and the recorded number of positive cases ( ) or for the insufficient number of beds into icus. in the former case, there is a statistical problem, whereas, in the latter case, the health policy of other countries should take into account the italian lesson for covid- . the aim of this study was to provide a deeper insight into the italian cfr, testing the hypothesis that the number of deaths increased more than linearly with the number of positive cases. in this study, the data officially provided by the italian ministry of health and istituto superiore di sanità were used to monitor the increment of cases of contagion and death related to covid- in italy. data were collected from the th of february to the th of march (supplementary table ) . polynomial, logistic, and bell-shaped functions were applied to fit the data. the equation of a bell-shaped function was the following the adjusted coefficient of determination (r ) was preferred to the raw one to assess the goodness of the fitting models independently by the number of their coefficients. the approach of the linear fit method (lfm) was used to compare the number of cases and that of deaths. this method was previously validated for assessing the waveform similarity in clinical data. the lfm relies on the idea of plotting one dataset vs. another one to compare the similarity of their waveforms, such as the contemporaneity of their peaks ( ) . in the rapid evolution of the pandemic of covid- , the dayby-day cfr was computed. it means that, for each day, the cfr was the percentage of deaths on the number of actually positive patients plus dead patients plus discharged patients. the theoretical scenarios are depicted in figure , which reports the case of a constant cfr as theoretically expected ( ) and that of a cfr computed to take into account a biological delay ( ) . a third case is reported, related to a dynamic perspective of cfr taking into account a potential increase in the period in which the number of severe cases overwhelmed the capacity of intensive care units (icus), which was the worst-case scenario hypothesized by anderson et al. ( ) . the bell-shaped models of figure show that the number of positive cases in italy is still increasing day by day, as is that of deaths. although a prediction is very difficult, these models have exhibited very high values for the adjusted coefficient of determination r ( . for actually positive, total infected, and dead patients, whereas it was . for discharged patients). independently by the goodness of the predictions, the trend of deaths seems to follow that of infections, with a delay of about days. the linear fit method approach has allowed us to compare the trends of real data, as reported in figure . the number of dead patients increased with the increment of infected patients (left panel). as clearly shown by the data, this increment has a second-order polynomial trend more than the expected linear one. when the cfr was computed (right panel of figure ) , an initial quite constant low value of cfr was observed, and it was followed by a progressive increment. in fact, in the first days of data collection, the italian cfr was roughly constant and lower than . %. it then started to increase. the linear increment computed using the lfm showed that r = . . the model, based on a theoretical biological delay of days in the computation of deaths, showed a lower value r = . . furthermore, this model had a concavity opposite to that revealed by data. conversely, in this phase, a bell-shaped increment related to the overwhelmed icus showed that r = . in fitting the data. this last model coincided with a double bell-shape model with a delay of only day between positive tests and deaths. mathematical models and parameters are often used in epidemiology to generate insight into the transmission dynamics of infectious diseases and to assess the potential impact of the different intervention strategies. first of all, italian data and our models supported the theoretical prediction that the italian trend of infected patients could be similar to that one of china. this prediction was previously suggested by remuzzi and remuzzi on the basis of italian data recorded up to the th of march upon which a tend similar to that observed in the hubei province, china, was applied ( ). our results indirectly suggested that the italian interventions, mainly based on the social distancing, have been effective in reducing the speed of contagions, as occurred in china. these restrictions seemed to reduce the increment of infected patients (often incorrectly reported as an exponential growth), preventing the intensive care units in the rest of italy from being overwhelmed as occurred in lombardy ( ). however, the resulting italian cfr was very high and progressively increased throughout march. this could be due to a miscomputation of cfr ( ) . however, figure clearly shows the number of deaths increased following a second-order polynomial function with respect to the number of positive cases. in a theoretical stationary situation, cfr is expected to be constant, meaning that the number of deaths proportionally (linearly) increased with the number of positive cases. but the high number of positive cases that occurred in lombardy in a small period might have overwhelmed the icus, having a secondary effect on the number of deaths in that italian region. in the case of covid- , the case fatality rate might be relevant for optimizing a health policy. many recent studies investigating this cfr have tried to explain the high value recorded in italy and progressively in other western countries ( , ) . our study showed that the italian data had a different and unexpected second-order increment of the number of deaths related to covid- with respect to the relevant number of infected patients. some authors have suggested that it could be due to the change in testing policy ( ), but the increasing trend occurred even after this change. other authors have suggested a correction in cfr computation for taking into account the time of incubation and worsening ( ) , but it seemed to fail in modeling the italian data. in fact, our results, obtained with different data analysis, seemed to show a delay ranging from - days between the curve of positive cases and that of deaths. furthermore, the concavity of the -week delayed cfr seemed to be opposite to that of data. the small delay found in our analyses was not compatible with a biological explanation, but it could be compatible with a health management explanation. this hypothesis seemed to be confirmed by a bell-shaped increment of deaths related to the difficulties of icus in managing a high number of patients with severe symptoms. it is possible that, although all the possible miscomputation of cfr could be related to an underestimation of positive cases, the italian cfr was affected by what happened in lombardy region, the region most infected. it was a scenario of an unexpected high number of cases, most of them recorded in a small area and in a short period of time (about weeks). the italian health policy was conceivably effective in attenuating the lombardy trend in the other regions, reducing the velocity of contagions thanks to the imposed social distancing. furthermore, in lombardy and in other regions, the number of beds in icus was increased. this possible explanation did not exclude that the high cfr was also due to an underestimation of positive cases. the emergency might also have leaded clinicians to focus on severe cases, progressively applying the reduction of tests in mildly affected and asymptomatic people ( ) . both these explanations, related to health policy, could be concomitant with the progressively increased high value of italian cfr. many other countries are now facing the emergence of covid- , and the computation of cfr could be misleading, even taking into account the biological delay. in an emergency and rapidly changing scenario such as the italian one, the cfr should be interpreted from a dynamic perspective, as it is potentially affected by many changing variables with effects that are not necessarily linear. direct and indirect effects of a wide contagion should be taken into account. the analysis of the evolution of the italian cfr trend could be of help to further develop a suitable health policy in other countries. for example, in further studies, it could be important to assess the complementary value of cfr, which is related to recovered patients. there could be an important percentage of them needing rehabilitation of motor and respiratory functions. some of these patients may not be able to wait for the end of emergency, but the health policy should face the problem of rehabilitation with a respect for safety. even unaffected older people may have motor deficits related to the long period spent at home. another aspect could be the psychological effects of covid- in recovered patients, including the fear of being infected or the psychological effects of social distancing in uninfected people ( ) . the datasets for this study can be found in the repository of frontiers and also in the official site of italian ministery of health (www.salute.gov.it). ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. mi performed all the data analysis and wrote the first draft of the manuscript. gm and sp provided important clinical content to the manuscript and supervised the study. this study was financed by italian ministry of health in the framework of current research for the line d of irccs fondazione santa lucia. the supplementary material for this article can be found online at: https://www.frontiersin.org/articles/ . /fmed. . /full#supplementary-material time to use the p-word? coronavirus enter dangerous new phase early transmission dynamics in wuhan, china, of novel coronavirus-infected pneumonia how will country-based mitigation measures influence the course of the covid- epidemic? critical care utilization for the covid- outbreak in lombardy, italy. early experience and forecast during an emergency response covid- and italy: what next? encyclopedia of forensic and legal medicine case-fatality rate and characteristics of patients dying in relation to covid- in italy real estimates of mortality following covid- infection incubation period of novel coronavirus ( -ncov) infections among travellers from wuhan available online at assessment of waveform similarity in clinical gait data: the linear fit method the covid- outbreak and psychiatric hospitals in china: managing challenges through mental health service reform the authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.copyright © iosa, paolucci and morone. this is an open-access article distributed under the terms of the creative commons attribution license (cc by). the use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. no use, distribution or reproduction is permitted which does not comply with these terms. key: cord- -guzn peq authors: livadiotis, george title: statistical analysis of the impact of environmental temperature on the exponential growth rate of cases infected by covid- date: - - journal: plos one doi: . /journal.pone. sha: doc_id: cord_uid: guzn peq we perform a statistical analysis for understanding the effect of the environmental temperature on the exponential growth rate of the cases infected by covid- for us and italian regions. in particular, we analyze the datasets of regional infected cases, derive the growth rates for regions characterized by a readable exponential growth phase in their evolution spread curve and plot them against the environmental temperatures averaged within the same regions, derive the relationship between temperature and growth rate, and evaluate its statistical confidence. the results clearly support the first reported statistically significant relationship of negative correlation between the average environmental temperature and exponential growth rates of the infected cases. the critical temperature, which eliminates the exponential growth, and thus the covid- spread in us regions, is estimated to be t(c) = . ± . f( ). the daily number of new cases infected by covid- is currently exponentially growing for most countries affected by the virus. however, this exponential growth rate varies significantly for different regions over the globe. it is urgent and timely to understand the reasons behind this regional variation of the exponential growth rates. little information is known about this matter, while there are indications that the environmental temperature may be a factor; for instance, northern and colder us and italian regions experienced much more incidents than others. typically, the evolution curve of the spread of the coronavirus initiates with a pre-exponential phase, which is characterized by a mild logarithmic growth, followed by the outbreak, that is, the phase of the exponential growth. social-distancing measures against the spread may affect the evolution curve in a way that the exponential growth slows down (decelerated phase) and starts to decline (decline or decay phase [ ] ), depending though on the effectiveness and applicability of these measures. however, after the decline of the spread at some place, new infected cases may outbreak in other places, marked with insignificant number of cases until that moment. then, a newly growth phase may appear. for example, fig (left) shows the evolution curve of the spread for the infected cases in mainland china; clearly, we observe the whole growth−decay cycle, as well as, a new re-growth phase. super-strict measures, such as complete shut down and quarantines, can successfully lead to the deceleration of the exponential growth of infected cases [ ] . unfortunately, they cannot be successfully applied and followed within vast regions, and especially, for a long and indefinite period of time. inevitably, measures may be loosened during the decay phase − if not earlier, leading to the birth of an equally disastrous re-growth phase. the exponential growth is the most effective phase for the evolution curve of infected cases; and the most important question regarding this evolution is still open [ ] : what can influence the exponential growth rate, and thus, "flatten the curve"? measures, strict or not, may affect the evolution of new infected cases, by shifting the spread curve from the exponential to the decelerated growth. it should be noted though that measures do not affect the exponential growth rate itself, but only the period of time that this exponential phase applies. then, what factors do affect the exponential growth rate? the age distribution in the place where the outbreak occurs is unlikely to be a factor; indeed, the number of new cases is known to be positively correlated with age, however, the exponential growth rate (china: . ; us: . ; italy: . -decreasing rate) appears to be negatively correlated to the age median of these countries (china: . ; us: . ; italy: . increasing age); hence, the age is likely irrelevant to the rate variations. in addition, culture in social activities may be a factor; for example, this might be contributing in the observed differences among the exponential rates in the cases of china, italy, and us (fig ) . however, what is causing the major variation of exponential rates among different regions of the same culture? it is apparent that culture does not constitute the main factor influencing the exponential rate. fig shows the regional variation of infected cases (left) during the exponential growth phase and the average winter temperature (right) in italy. the possible negative correlation, observed between regional number of infected cases and winter temperature in italy, is an indication of the influence of temperature on the exponential growth, but it certainly does not constitute a necessary condition. the reason is that the map plots the total number of the infected cases n t , which is not dependent only on the exponential growth rate λ, but also on the initial number of cases n . it is generally accepted that the initial infected cases in italy were travelled directly from china; since some destinations are more favorable than others, then, the initial number of cases n , as well as the current number of cases n t (which is proportional to n ), should be subject of regional variation. therefore, there is a non-negligible possibility, the observed regional variation of the number of infected cases n t to be caused by the regional distribution of the initial cases n . in such a case, main airport cities would have incredibly high number of infected cases outplaying a possible negative correlation of daily infected cases with regional average temperature t; the latter may be one of the reasons of the high numbers of cases observed in new york city and rome. on the other hand, in their letter to the white house, members of a national academy of sciences committee said that "there is some evidence to suggest that [coronavirus] may transmit less efficiently in environments with higher ambient temperature and humidity; however, given the lack of host immunity globally, this reduction in transmission efficiency may not lead to a significant reduction in disease spread without the concomitant adoption of major public health interventions" [ ] . nevertheless, it has to be stressed out that there were no statistical analyses focused on the exponential growth rates of the infected cases in regions with different temperatures. for instance, several authors (e.g., [ , ] ) found insignificant correlations between temperatures and confirmed cases. however, their analysis was performed on the number of the infected cases n t , which is subject to the randomness of the initial cases n as explained above, and not pre-exponential (pre-exp), exponential (exp) growth, decelerated growth, decay (or decline), and possibly, a re-growth. day t = corresponds to / / for china, / / for italy, / / for us. evolution in china cases follows the whole growth-decay cycle, and a new re-growth phase. italian cases are characterized by a milder exponential rate, entered the phase of decelerated growth on march . us suffers with a larger exponential rate, and it is not clear whether has entered the decelerated growth phase. the exponential growth rate for china rose as high as λ = . ± . , while for italy and us the rates were λ = . ± . and . ± . , respectively (with correlation coefficient > . ). https://doi.org/ . /journal.pone. .g impact of environmental temperature on growth rate of covid- cases on the exponential growth rate λ, which is clearly dependent on physical characteristics of the coronavirus, binding protein, and environment. analysis of regional cases can show whether the speculated negative correlation between temperature and number of infected cases is true, meaning a negative correlation between temperature and exponential growth rate. if the environmental temperature plays indeed a substantial role on the virus spread, then, this can provide promising results, such as, the estimation of the critical temperature that may eliminate the number of daily new cases in heavily infected regions. the purpose of this paper is to improve our understanding of the effect of environmental temperature on the spread of covid- and its exponential growth rate. in particular, we calculate the exponential growth rates of infected cases for us and italian regions, derive the relationship of these rates with the environmental temperature, evaluate its statistical confidence, and determine the critical temperature that eliminates this rate. a standard model for describing the evolution of the infected cases by viruses can be constructed as follows where is the number of total infected cases evolved from the initial n �n( ) cases, n max is the maximum possible number of infected cass; λ is the exponential growth rate, and becomes clear for x(t) << where i is negligible, leading to: typically, data of infected cases are daily provided and updated. thus we set the readout of n(t) on a daily basis, such as: n t �n(t/[d]). hence, we may write with where x t �n t /n max with t indicating the time on a daily basis, (t = d, d, . . .). the function of negative feedback i models the factors that flattens the curve, such as, the measures taken against spreading. while these factors are not affecting the exponential growth rate λ, they become more effective as the number of cases increases, getting closer to n max ; exponent b controls the effectiveness of these factors; strict {loose} measures correspond to smaller {larger} values of b. as observed in fig (a) , stricter measures, nicely modeled by decreasing b, do not affect the exponential rate λ but they successfully flatten the curve. however, the same can be achieved by downgrading the exponential growth rate, as shown in fig (b) . it is apparent, then, how much useful would it be to know the factors that can flatten the curve by decreasing directly the exponential growth rate, instead of applying stricter measures. applied measures could be loosen and shorter! model ( ) originates from the logistic map family (e.g., [ ] , and references therein; [ ] ); other complicate versions, such as, the susceptible-infectious-recovered models (e.g., [ ] ) may be expressed by multi-dimensional differential or difference equations (e.g., [ ] , and references therein; [ ] ), but still, the curve flattening is governed by the same features. the two composites, the exponential growth e and the negative feedback i, are just the main and necessary conditions for reproducing the growth-decay phases of the spread curve. their interplay shows how the spread curve can be flattened as a result of stricter measures, independently of the existent exponential rate. what are the main factors that can affect the exponential growth rate λ of covid- spread? the rate λ is expected to have positive correlation with the reproduction number r (e.g., proportional to its logarithm), and negative correlation with the incubation period τ (e.g., inverse proportional) [ ] . the number r is a measure of how contagious a disease is; it provides the average number of people in a susceptible population that a single infected person will spread the disease over the course of their infection [ ] , and depends on the physical characteristics of coronavirus [ ] . the incubation period τ is the time elapsed between exposure to coronavirus and first symptoms; during this period, an infected individual cannot infect others; other characteristic periods and time intervals are the latent period between exposure and infection, and the generation time, mostly concerned with transmission process [ ] . characteristic values for covid- are τ~ - days and r ~ - [ ] . the rate expression can be written as λ/lnr /τ, and involves all the physical characteristics of the mechanisms of infection and the environmental interactions; this can be easily derived, considering difference equations (that is, iterated discrete maps) (e.g., see: [ ] [ ] [ ] ). setting the time to be given in discrete τ-steps (t = τ, τ,. . .), then, by definition of r (average number of people that a single infected person will spread the disease), we have n t = r n t−τ , that is, we note that the number of the infected cases does not vary for times t taken in-between the integer multiples of τ, but this is not expected in mixtures of populations with random characteristics. indeed, in a mixture of m evolving infected populations with different initial number of cases n (m) and starting times t (m) , m = , , . . ., m, the total number of infected cases n at a continuous time t is given by which coincides with ( b), but with time t varying on a daily basis, independently of the larger value of τ~ days [ ] . therefore, we set the readout of the total number of infected cases n on a daily basis, such as: ; then, eq ( d) matches eq ( b), where the exponential rate is given by: the main factors that can affect the exponential rate λ are: (a) culture in social activities, and (b) environmental temperature and/or other thermodynamic parameters. intense cultural and social activities have reasonably a positive correlation with r . as previously mentioned, measures against the virus spread do not effectively influence the exponential growth rate; e.g., they do not change the culture in social activities, which are characteristics of the particular population, but they just cease these social activities for some period of time. in terms of modeling, measures appear only in the negative feedback factor i and not in the e factor of model in eq ( ), while the culture, together with the environmental temperature, are the two main parameters affecting r directly. potentially, the environmental temperature t can affect all the parameters influencing exponential rate. we approach this dependence by (i) a linear approximation of the phenomenological relationship between exponential rate and temperature, and (ii) the connection of reproduction number with arrhenius behavior (with negative activation energy): (i) the temperature can affect the physical properties of coronavirus, such as, the incubation time τ, as well as, the reproduction number r that depends on these physical properties [ ] . a linear approximation absorbs the (weak) temperature dependence of any parameters involved in the exponential rate; then, eq ( ) is linearly expanded as: where we set the intercept to be given in normal conditions of atmospheric temperature and pressure (ntp) (that is, t = c , p = atm). then, we rewrite the exponential rate as: where λ and p are the intercept and slope of the linear relation ( a). (ii) coronavirus uses their major surface spike protein to bind on a receptor-another protein that acts like a doorway into a human cell [ ] . the whole process is a slow chemical reaction, where the mechanism behind can lead to rates negatively correlated with temperature, i.e., increasing rate with decreasing temperature. this is consistent to reaction rate expressed by the arrhenius exponential with negative activation energy exp[|e a |/(k b t)] [ ] . then, the effective reproduction number r (t) is expressed as a product combining the reproduction number in the absence of temperature effect, r , and the arrhenius exponential rate, namely, then, eq ( ) gives we rewrite this expression as: where −λ and p are the intercept and slope of the linear relation ( a), respectively. reactions of negative activation energy are barrier-less, relying on the capture of the molecules in a potential well. increasing {decreasing} the temperature leads to a reduced {gained} probability of the colliding molecules capturing one another. due to the negative activation energy, decreasing the environmental temperature reduces the probability of virus-protein reaction, thus the virus may stay inactive on air or surfaces and eventually die. exponential growth is related to community spread through outdoors activities, while the decelerated growth caused by effective measures is related to indoors activities: exponential growth exists once the disease is still effective and the measures are loosened, allowing people to outdoor social activities; however, exponential growth decelerates followed by the decay phase, once effective measures hold people in small groups indoors. therefore, exponential growth rate must be related to outdoors (rather than indoors) activities, and thus to the environmental temperature. as long as the exponential growth takes place, the environmental temperature has an effective role on the chemical reaction between virus and spike protein. it should be noted that both the models ( b) and ( b) consider that the exponential rate λ, or the reproduction number r , are subject to a component influenced by the culture in social activities (intercept λ ) and a component mostly influenced by the temperature (linear term with slope p ). in this way, the slope may indicate some universal quantity involved, such as, the (negative) activation energy. next, we employ the above two expressions of exponential rate λ and temperature t, eqs ( a and a), in order to set the two types of statistical models for fitting (Τ, λ) measurements for us and italian regions. we use publicly available datasets of: ( ) average environmental temperature of us and italian regions (e.g., see: www.ncdc.noaa.gov/data-access/land-based-station-data/land-baseddatasets/climate-normals; it.climate-data.org; www.weather-atlas.com); ( ) time series of the number of daily infected cases of us and italian regions (e.g., see: www.thelancet.com; www. protezionecivile.gov.it). we analyze the datasets of regional infected cases in us and italy, derive the relationship of the exponential growth rate of the number of cases with temperature, and evaluate its statistical confidence. first, we derive the exponential growth rates of the infected cases characterizing each examined region of us and italy; then, we plot these values against the environmental temperatures of each region, and perform the corresponding statistical analysis. we proceed according to the following steps: i. collect the time series of the current infected cases n t for all us and italian regions. ii. for each of the us and italian regions, we plot log (n t ) and log (Δn t ) with time t, detect the time intervals of linear relationship corresponding to the phase of exponential growth, fit the data-points within this region, and derive the slope (on linear-log scale), that is, the exponential growth rate λ. the total n t and new cases Δn t should be characterized by the same exponential rate, λ, thus the slopes resulted from the linear fits of log (n t ) and log (Δn t ) with time are (weighted) averaged (fig ) . iii. collect environmental temperature data, and calculate the temperature averaged over the whole examined region. the incubation period τ is longer than the time scale of a single day or night, thus the temperature is averaged over the daily and nightly measurements. iv. co-plot all the derived sample values (Τ±δΤ, λ±δλ), where each pair corresponds to each examined region; then, apply a linear fitting in order to derive the linear relationship between t and λ, as well as, evaluate the statistical confidence of this relationship; repeat the same for all us and italian regions. v. determine the critical temperature t c for which the rate becomes negligible; to eliminate the uncertainties of t c as a fitting parameter, we perform the linear fitting with the statistical model λ = λ ( −Τ/t c ) instead of λ = p +p �t. vi. repeat (iv) and (v) with pairs of (t - ±δt - , λ±δλ); we estimate again t c by performing the linear fitting with the statistical model λ = λ (- +Τ c /t) instead of λ = p +p �t - . the hypothesis to be tested is that the exponential growth rate λ varies linearly with temperature; (x is set to be the temperature or its inverse). this is tested by examining the chi-square corresponding to the fitting of the two-parameter linear statistical model λ(x;p ,p ) = p +p x to the given n data points, (the number of data point, n, should not to be confused with the number of cases, n t ). therefore, we minimize the chi-square w ðp ; where the total variance that characterize each data point is now given by σ i (p ) = σ λi +p σ xi [ ] . the global minimum of the chi-square function χ (p ,p ) gives the optimal parameter values, ðp � ; p � Þ, by solving the normal equations @χ (p ,p )/@p = and @χ (p ,p )/@p = ; the minimum chi-square value is w min ¼ w ðp � ; p � Þ. the statistical errors of these values are given by dp a; the statistical confidence of the dependence of the exponential growth rate on the environmental average temperature may be sufficiently high, leading to the acceptance of any of the two statistical models. the goodness of the fitting of each model is evaluated using two types of statistical tests, the "reduced chi-square", the "p-value of the extremes", and their combination (e.g., [ ] [ ] [ ] ), while student's t-test is also used for evaluating the statistical confidence of the derived slopes: • reduced chi-square: the goodness of fitting is estimated by the reduced chi-square value, the meaning of w red is the portion of w min that corresponds to each of the dof, and w red has to be~ for a good fit. therefore, fitting is characterized as "good" when w red~ , otherwise there is an overestimation, w red < , or underestimation, w red > , of the errors. one order of magnitude less, w red = . , or more, w red = , can be set as the accepted limits, i.e., . � w red � . • p-value of the extremes: the goodness of fitting is evaluated by comparing the estimated minimized chi-square value, w min , and the chi-square distribution, , that is, the distribution of all the possible χ values (parameterized by the dof = m). the likelihood of having a χ value, equal to or larger than the estimated value w min , is given by the complementary cumulative distribution. the probability of taking a result χ , larger than the estimated value w min , defines the p-value that equals the larger the p-value, the better the fitting. according to this method, the probability of taking a result with χ being extremer than the observed value w min , defines the p-value of the extremes; this equals the minimum between the two probabilities, pð � w � w min Þ and its complementary, pðw min � w < Þ. fits associated with p-values smaller than the significance level of . are typically rejected. • combined p-value and chi-square: the p-value of the extremes has very similar behavior with the reduced chi-square [ , ] , because, (i) the p-value attains the optimal value (p = . ) when chi-square does (w red = ), (ii) larger values of - p corresponds to larger values of jw red À j, (iii) both fractions ( − p)/( + p) and j À w red j=ð þ w red Þ range from to , reduced to / when the accepted limits, p = . or . �w red � , are reached. then, a combined measure can be defined by the sum of the squares of these fractions, i.e., ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ½ð À pÞ=ð þ pÞ� þ ½ð À w red Þ=ð þ w red Þ� q . • student's t-test: this is another test for evaluating the statistical confidence of the slope derived from the linear fitting of the temperature-rate sample points (t i ±δt i , λ i ±δλ i ) and , λ i ±δλ i ). we examine, whether the slope p ±δp has significant difference from the zero slope (null hypothesis: slope is zero), by performing the student's t-test with t m = p /δp , where the corresponding p-value is derived from the integration of t-distribution p t ðt; mÞ ¼ for t�t m < , i.e., p t ðt m ; mÞ ¼ r t m p t ðt; mÞdt. the student's t-test is not passed for the null hypothesis that the examined slope equals zero, when the corresponding p t -value is smaller than the acceptable confidence limit of . ; then, the null hypothesis is rejected, imposing that the slope has statistically significant difference for zero. in addition, we compare the slopes estimated for us with those estimated for italian regions, by deriving t m ¼ jb us À b it j= ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi ffi and then, finding again p t (t m ;m m ); the t-test is passed for the null hypothesis that the examined slopes are equal, when the corresponding p t -value is larger than . ; in this case, the null hypothesis is accepted, thus there is no statistically significant difference between the two slopes. the linear fitting of log (n t ) or log (Δn t ) with respect to time t within the region of exponential growth phase, resulted to the respective rates (which are given by the fitted slopes); their weighted averages are shown in table for us and in table for italian regions, while plotted against the average regional temperature in figs and , respectively. the method of weighted fitting for double uncertainties (x i ±δx i , λ i ±δλ i ), as described by [ ] , is used for estimating the fitting parameters λ , t c , together with their statistical, propagation, and total errors. the fits of the linear statistical model with temperature, x i = t i , (left panels in figs and ) , as well as of the alternative statistical model with inverse temperature, , (right panels in figs and ) , are both characterized with high statistical confidence, attaining high p-values (> . ) and reduced chi-squares w red values (close to ); also, both fits provide similar estimations of t c . the fitting results are shown in table . we also examine whether the sample points (t i ±δt i , λ i ±δλ i ) are subject to statistically significant concentrations or rarefactions, namely, whether possible heterogeneities within the distribution of sample points plays significant role in the fitted relationship. for this, we derive the temperature-rate relationship and its statistical confidence by fitting the homogenized set of sample points, instead of the raw sample points; then, we examine whether the fitting parameters differ from those derived from fitting the raw sample points. we homogenize the sample points by grouping them in temperature bins of Δt~ c (e.g., see: [ ] ). we estimate the weighted mean and error of the rates included in each bin. in the case of us regions we also performed a homogenization of rates, by grouping the temperature-binned means in rate bins of Δλ~ . d - . in the case of sample points with inverse temperatures, (t i , λ i ±δλ i ), the procedure is exactly the same. homogenized datasets result in a smooth relationship between the values of binned temperature and rate, as it can be observed in the plots of rate against temperature or inverse temperature (left or right lower panels, respectively), and for both us and italy regions (figs and , respectively). the results are highly supportive of the negative correlation between rate and temperature. the results are shown in table . we observe that the linear relationships of the growth rate with temperature or inverse temperature are characterized by high statistical confidence for the homogenized datasets (p-values much higher than the significant limit of . ; w red far from the significant limits of . and ). therefore, the arrangement of sample points do not affect significantly the fitting results. in addition, as shown in tables and , the linear fits of sample points (t i ±δt i , λ i ±δλ i ) and , λ i ±δλ i ) do not pass the student's t-test for the null hypothesis that their slopes equals zero, i.e., the corresponding p t -values are smaller than the acceptable confidence limit of . ; therefore, the negative correlation of environmental temperature with the exponential rate is statistically significant (accepted with confidence %). in order to improve the statistics of the estimated critical temperature, we combine the sample points (t i ±δt i , λ i ±δλ i ) of us and italian regions. first, we perform the student's t-test to compare the slopes from these regions; we find high p t -values (> . ) for both fits of x = t and x = t - , thus, the two populations are likely characterized by the same slope. the respective intercept λ does not pass the same test, i.e., the intercepts corresponding to us and italian regions are likely different; (that is expected, given of the different culture). a universality may characterize the slopes of two countries, either for the fits with x = t or x = t - , i.e., p = (@λ/ @t) ntp or p = |e a |/(τk b ), respectively. next, we perform the linear fits of the sample points (t i ±δt i , λ i ±δλ i ) and (t i , λ i ±δλ i ) for the mixed set of us and italian data, once the rates of the italian regions are shifted by Δλ; (this is allowed, since it has just be shown that a universality is likely characterized the slopes). the optimal fitting is obtained for that shift Δλ, for which the reduced chi-square is , the p-value of the extremes is~ . , and the combined measure~ (see previous section). fig shows how the combined datasets of temperature-rates from us and italian regions lead to the optimal fitting. (note that the optimization is not performed for the binned datasets, since they are characterized by smaller p-values-see, figs and ) . the results are shown in table ; we observe that the optimization is reached for two values of the shift Δλ; we estimate the weighted average of the results corresponding to the two shifts. the weighted mean is performed separately for the fitting cases of x = t and x = t - ; however, the weighted mean of the critical temperature is performed for all four results. table includes the weighted means of slopes for the fits x = t or x = t - , with slopes p = −|@λ/@t| ntp and p = |e a |/(τk b ), respectively. the latter can be used for deriving the activation table . fitting parameters of temperature-rate values for us and italian regions. b) we derive the reproduction number r , i.e., the two formulae in eq ( b) provide the value of t lnðr ntp Þ as . ± . and . ± . , respectively, with weighted mean . ± . ; then, we find r ntp ffi : � : , that is, the reproduction number for t = c . the corresponding number at t = c is r ( c )ffi . ± . , while by substituting the estimated parameters in eq ( ), we derive the up-to-date there is no systematic statistical analysis of the effect of the environmental temperature t (and possibly other weather parameters) on the exponential growth rate of the cases infected by covid- , while a statistically confident relationship between temperature and growth rate (either with positive or negative correlation) was unknown. the presented analysis led to the first statistically confident relationship of negative correlation between the exponential growth rate and the average environmental temperature, derived for us and italian regions. in particular, we analyzed datasets of regional infected cases in us and italy, derived the exponential growth rates for each of these regions and plotted them against environmental temperatures averaged within the same regions, derived the relationship of temperature-growth rate, and evaluated its statistical confidence. the performed statistical analysis involved fitting of linear statistical models with the datasets of environmental temperature (or its inverse) and exponential growth rate. the two linear models developed and used for the statistical analysis are (a) λ(t) . the statistical confidence of fitting was evaluated using the reduced chi-square values, the p-value of extremes, and a testing measure that combines both of these values; also, the student's t-test was used to compare the derived slopes. the sample points of temperature (or inverse temperature) and exponential growth rate were also tested for statistically significant concentrations or rarefactions, that is, for possible heterogeneities within the distribution of sample points that could have significant role in the results. the statistical analysis of the homogenized temperature-rate data points concluded that the negative correlation between temperature and exponential rate is stable, having no statistically significant variability due to concentrations or rarefactions, and it is characterized by a high statistical confidence. we also performed a student's t-test and ensured that the difference between the sample means of us and italian regions is not statistically significant. a universality is likely characterizing the slope of the temperature-rate relationship. this verifies the modeling developed and used by this analysis, where the exponential rate λ, or the reproduction number r , are subject to a component influenced by the culture in social activities (intercept λ ) and a component influenced by the temperature (slope p ). in this way, the slope may indicate to a universal quantity involved, such as, the (negative) activation energy. having shown that the derived slopes for us and italian regions are characterized by no statistically confident difference, we improved the statistics of the estimated fitting parameters by combined the sample points of us and italian regions. from the derived relationship, among others, we were able to estimate the values of the (negative) activation energy e a , as well as the reproduction number r at normal conditions and how this depends on temperature. therefore, the results clearly showed that there is indeed statistically significant negative correlation of temperature on the exponential growth rate of the cases infected by covid- . fig shows the anti-correlation between the mapped exponential rates and average environmental temperature of the us regions examined by this analysis, which they are characterized by a readable exponential growth phase in their evolution spread curve. the plot shows also the may-june daily, nightly, and h-averaged environmental temperatures in san antonio, texas, averaged over the last three years. the daily average temperatures will be clearly above the estimated t c threshold in the second half of may; thus, the plot suggests a possible date for loosening the strict measures in san antonio, that is, may . https://doi.org/ . /journal.pone. .g given the negative correlation of the environmental temperature with the exponential growth rate, it was reasonable to ask for the critical temperature that eliminates the exponential rate, and thus the number of daily new cases in infected regions. this was found to be t c~ . ± . f for us regions. it is straightforward to ask when the environmental temperature will climb above this critical value. as an example, fig plots the daily average temperatures in san antonio, texas, shown that it will be clearly above the estimated t c threshold by the end of may. the resulted high statistical confidence of the negative correlation of the environmental temperature on the exponential growth rate of the cases infected by covid- is certainly encouraging for loosening super-strict social-distancing measures, at least, during the summery high temperatures. however, we are, by no-means, recommending a return-to-work date based only on this study. but we do think that this should be part of the decision, as well as an inspiration for repeating the same analysis in other heavily infected regions. the steps of these analyses may be followed as: i. identify different outbreaks in regions with the same culture in social activities and different environmental temperature; ii. estimate the exponential growth rates for these regions from the time series of infected cases; iii. plot the derived rates against the environmental temperature averaged for these regions, and repeat the analysis of this study to determine the temperature-rate relationship and its statistical confidence. conceptualization: george livadiotis. formal analysis: george livadiotis. funding acquisition: george livadiotis. investigation: george livadiotis. methodology: george livadiotis. an accurate two-phase approximate solution to an acute viral infection model greece shows how to handle the crisis how to flatten the curve of coronavirus, a mathematician explains rapid expert consultation on sars-cov- survival in relation to temperature and humidity and potential for seasonality for the covid- pandemic. the national academies press effects of temperature on covid- transmission no association of covid- transmission with temperature or uv radiation in chinese cities numerical approximation of the percentage of order for one-dimensional maps generalized logistic growth modeling of the covid- outbreak in provinces in china and in the rest of the world how covid- and other infectious diseases spread: mathematical modeling an introduction to difference equations kappa function as a unifying framework for discrete population modeling vaccinology: an essential guide infection dynamics of coronavirus disease (covid- ) modeled with the integration of the eyring's rate process theory and free volume concept time variations in the generation time of an infectious disease: implications for sampling to appropriately quantify transmission potential a mathematical model for simulating the phase-based transmissibility of a novel coronavirus general allee effect in two-species population biology nearly exact discretization of single species population models mathematical model for the aggregation of β-amyloid cryo-em structure of the -ncov spike in the prefusion conformation negative activation energies and curved arrhenius plots. . theory of reactions over potential wells fitting a straight line with errors on both coordinates approach to general methods for fitting and their sensitivity geometric interpretation of errors in multi-parametrical fitting methods decades-long changes of the interstellar wind through our solar system solar radiation pressure and local interstellar medium flow parameters from ibex low energy hydrogen measurements low energy neutral atoms from the heliosheath chi-p distribution: characterization of the goodness of the fitting using l p norms general fitting methods based on l q norms and their optimization plasma-field coupling at small length scales in solar wind near au key: cord- -fsrdu tq authors: allieta, m.; allieta, a.; rossi sebastiano, d. title: covid- outbreak in italy: estimation of reproduction numbers over two months toward the phase date: - - journal: nan doi: . / . . . sha: doc_id: cord_uid: fsrdu tq after two months from the first case in covid- outbreak, italy counts more than , confirmed positive cases. from the beginning of april , the nationwide lockdown started to show early effects by reducing the total cumulative incidence reached by the epidemic wave. this allows the government to program the measures to loosen lockdown restrictions for the so called "phase ". here we provided the reproduction number estimation both in space and in time from february th to april th, across two months into the epidemic. our estimates suggest basic reproduction number averaged over all the regions of . , confirming that epidemiological figures of the sars-cov- epidemic in italy are higher than those observed at the early stage of wuhan (china) outbreak. based on the sars-cov- transmission dynamics reported here, we gave a quantitative evaluation of the efficiency of the government measures to low the reproduction number under the unity (control regime). we estimated that among the worst hit regions in italy, lombardy reached the control regime on march nd followed by emilia-romagna (march th), veneto (march th) and piemonte (march th). overall, we found that the mean value of time to reach the control regime in all the country is about days from the february th and about days from the first day of nationwide lockdown (march th). finally, we highlighted the interplay between the reproduction number and two demographic indices in order to probe the "state of activity" of the epidemic for each italian region in the control regime. we believe that this approach can provide a tool in the management of "phase ", potentially helping in challenging decision to continue, ease or tighten up restrictions. after the first covid- case was diagnosed in lombardy, italy, on february th, , [ ] the novel coronavirus rapidly spread across the country leading to a dramatic spike in the number of new positive cases and deaths. to minimize the likelihood that people who were not infected come into contact with people who had contracted the disease, the italian government imposed a series of progressively more strict social distancing measures which culminated in a national lock-down announced on march th, . [ ] around two months from the first case and more than , confirmed positive cases later, from the beginning of april, the effect of the nationwide lockdown started to achieve some level of success and the number of new infections began to smoothly decrease. these early signs of a slowdown of the covid- pandemic in italy provide a comforting picture of the outbreak's stabilization which is driving the government to periodically review its lockdown measures in view of the so called "phase ", i.e. the period during which citizens will have to live together with the virus as of all the industrial sector, including the nonessential economic activities, will start to reopen. however, since the regional differences in the number of new positive cases has been reported to be huge, with the northern regions of italy (namely lombardia) being most affected, the establishment of the proper precautions to plan the "phase " is a truly complicated task. the planned restrictions and permissions that will be applied could thus vary from region to region. in this context, the systematical estimation of key epidemiological parameters, for each region can provide insight into the speed at which the disease had spread and will give a useful tool to figure out if a differential approach at the regional level on the measures to apply for "phase " is feasible to keep down the transmission of sars-cov . at the beginning of epidemic and during the lockdown phase, italian government and the mainstream of the local and national mass media have been emphasized the relevance of the basic reproduction number (r ) , i.e. the average number of secondary cases generated by a single primary case in a theoretically fully susceptible ( %) population, as the most important and informative parameter to monitor the epidemic trends. obviously, r has an undoubted relevance since when r > the infection may spread in the population and more r is large and deeper would be the interventions needed to control the epidemic. on the other hand, if r < , on average the infectious individual infects less than one person and the epidemic falls in a so called "control regime" where it will not be sustained, and it will die out. nevertheless, r is not the only parameter that affect the impact and the spreading of all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint the disease over a population which may largely result even from several demographic and epidemiological factors. in this communication, we provided an estimation of the basic reproduction number r for all the italian regions by the cumulative confirmed covid- cases continuously updated and made public at the website of dipartimento della protezione civile. [ ] in addition, we estimated the time dependent reproduction number rt, which is the average number of secondary cases generated by an infectious individual at time t. we linked rt related to the last date of our period of observation (april th, ), with two demographic and epidemiologic indices in a simple three-dimensional array in order to highlight the "state of activity" of the epidemic for each italian region. we provide a useful tool in the management of "phase ", potentially helping in challenging decision to continue, ease or tighten up restrictions. the official demographic data of the resident population, the surface and the population density updated on january st, , for each italian region and italy were taken from the italian national institute of statistics (istituto nazionale di statistica, istat) and reported in the official data of covid- epidemic in italy was taken from the task force of the dipartimento della protezione civile. cumulative data are available at various aggregation levels, namely national, regional and provincial and are accessible on github. [ ] data for the analysis were considered from february th to april th, . in this period, we collected the daily cumulative number of confirmed positive cases (n), the number of "active" confirmed positive cases (na), i.e., the number of infected people living not recovered from covid- , and the "density of infected people" (da), calculated as na/surface and expressed like the population density as number of persons/km , for each italian region and italy. all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. to obtain the estimation of reproduction number we use the maximum likelihood estimation (ml) method which assumes that the number of secondary cases caused by an index case is poisson distributed with an expected value r. given then observation of (n , n ,..., nt) incident cases over consecutive time units, r is estimated by maximizing the following loglikelihood function [ ] : i is the distribution of the generation time corresponding to the distribution of the serial interval, i.e. the time between when a person gets infected and when they subsequently infect another other people, calculated at time i within the assumption that the incubation period does not change over the course of the epidemic [ ] . we consider that the distribution of the serial interval was expected to follow a gamma distribution with mean (±sd) of . ± . days as reported by the imperial college covid- response team [ ] . we note that this value agrees very well with gamma distribution with mean . days ( % ci, . to ) recently determined from the analysis of observations of individual serial intervals in clusters in lombardia (italy) [ ] . to estimate r = r , the ll(r) function must be calculated over a period where epidemic curves showed exponential growth. as a first guess, to select this time window we used the simple procedure described by obadia et al. [ ] in brief, we computed the function over a range of possibile time periods by determining the deviance r statistic for each iteration. largest r corresponds to the time window over which the ml model best described data. to evaluate the time dependent reproduction number rt we adopted the method developed by wallinga and teunis. [ ] the transmission probability (pij) of individual i being infected by individual j at ti, tj onsets, respectively, can be described mathematically as: [ ] the net reproduction number rj is then then sum of all pij involving j as the infector = ∑ and it can be averaged over all cases with same date of onset as all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint finally, since rt are computed by averaging over all transmission networks compatible with observed incidence data, no assumption is made about the time dependence of the epidemic unlike, for example the exponential growth in the well-known bayesian approach. [ ] , [ ] we believe, hence, that this model is particularly suitable to estimate the reproduction number in the post-peak period where the transmission is expected to decrease. all the above data analyses were performed using the r package [ ] as implemented in statistical software r. [ ] .results infection in italy shows that the exponential growth period may take place during the first - day from the national epidemic onset (february th , ). table s -s (supporting information) show demographic and epidemiological data, respectively. as it is widely known, table s shows that covid- epidemic affected (and is affecting) harder the northern italian regions, with n= and na= on april th, i.e. more than % of the cases of the country (with , % of the italian resident population), if we aggregate epidemiological and demographic data of the northern regions (lombardia, piemonte, veneto, emilia romagna, liguria, valle d'aosta, trentino-alto adige) plus marche and toscana regions. furthermore, in lombardia region epidemic had a huge spread, with n= and na= on april th, i.e. more than one third of the cases of the country (with . % of the italian resident population). in the top of the panels of figures s -s (supporting information) we reported the incidence data for all the regions plus italy. initial inspection of the datasets shows again that the exponential growth period may take place during the first - day from the relative all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint epidemic onset. it should be noted that for the evaluation of r in the initial outbreak stage, we considered data from february th , up to march th , since data in a wider range can be affected by the national lockdown on march th . in figure (a) we showed r values obtained for sars-cov- in all the regions and in italy. table reports the same data represented in figure (a), compared with those obtained by riccardo et al., [ ] d'arienzo et al., [ ] distante et al. [ ] . according to our ml estimation, (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. in this work, we analyzed the time evolution of incidence of the sars-cov- epidemic for two months from onset, february th to april th , in all the italian regions. we estimated the basic reproduction number (r ), by using the ml method in the early stage of the epidemic. in addition, we determined time evolution of this parameter across the two months of the observational period. finally, we linked rt, with two indices, the population density and da, the latter representing the density of infected people in a region as recorded on april th . firstly, we point out that these data can be considered only an approximation of the actual epidemic dynamics. indeed, the reported number of cases strictly depends on the number of swabs that are used for covid- testing and can be biased by several factors like underreporting, delays in recording as well as errors in classification of cases. [ ] therefore, large data noise is general observed, especially at the regional level, which requires a careful inspection of the epidemic curve as well as data smoothing in order to avoid unrealistic reproduction number estimation. as described in the results, for the evaluation of r in the initial outbreak stage, we considered data from february th , up to march th . data in a wider range can be affected by the national lockdown on march th . this period agrees well with previous investigation where the same time window has been assumed as the infection period to determine r for the whole italy. [ ] taking these preliminary considerations into account, our result of r = . for italy is highly consistent with values obtained by fitting the exponential growth rate of the infection across a -month period. [ ] similar conclusion has been drawn for northern regions transmission dynamics and the same results were found for the southern regions. [ ] in another work, riccardo f. et al. [ ] reported r ranging from . to . for six selected italian regions (lombardia, veneto, emilia-romagna, toscana, lazio, puglia). despite these values are lower than r obtained here, a variability of ~ . for most of the regions is thus confirmed independently of geographical location. again, gatto et al. [ ] , while including additional parameters like mobility and the spatial distribution of communities, determined a comparable initial generalized reproduction number r = . . overall, these data support the idea that epidemiological figures of the sars-cov- epidemic in italy are slightly higher than those observed at the early stage of outbreak in wuhan (china). [ ] the initial large values observed resulted from a sudden increase of independent first reported infections which in many cases can be related to the so called "super-spreading" all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint events. indeed, as observed for sars outbreak, [ ] in the early stage of the epidemic the time dependence of rt shows a fluctuating pattern characterized by wide confidence interval raised by the initial low number of cases used in the calculations. in this context, the superspreading events cannot be necessarily triggered by a single infector, but it can be related to few people which are perpetuating an epidemic in the susceptible population. [ ] here we observed that most of the regions have faced "super-spread events" in the early stage of epidemic. and significant is the observation of such event in southern regions after the early stages, the rt showed a decreasing trend which is likely to be affected by the temporal depletion of susceptible individuals (intrinsic factors) and by the implementation of control measures (extrinsic factors). [ ] both these factors slow down the growth rate of incidence and deeply affect the shape and time scaling of the epidemic peak driving rt to fall below . [ ] we found that the mean value of time to reach the control regime is about days from the (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . although the italian government's restrictive measures have proven to be of considerable utility in preventing even more devastating effects from the epidemic, the challenge in tackling "phase " appears even more demanding. in this line, obtaining simple and effective indices to evaluate the state of activity of the epidemic seems mandatory: if the rt index remains essential for understanding the trend in a given area, however it is not the only parameter to account for. briefly, if we consider two areas with the same rt, that of the two that has a population density and a higher percentage of infected people must be considered more at risk, monitored more carefully and potentially the target of more timely restrictive measures. the population density of a given area is clearly an rt-independent risk factor for the development of an epidemic that spreads through human infection, although the population density of the different italian regions may not be truly representative of the distribution of the population. urban areas and in particular metropolitan areas (rome, milan, naples) have a population density higher than the regional one. furthermore, due to the peculiar italian orography, some regions (for example liguria, valle d'aosta, trentino-alto adige) concentrate the population in a "habitable" area much less large than the total surface. while admitting its arbitrariness, da (ie the number of infected people per km ) is in some way a representative parameter of how much the epidemic was active in the previous period and, above all, what is the generic risk of "meeting" a subject affection in a given area. therefore, we suggest to associate a combined use of rt with da and population density to evaluate the epidemic risk of a specific area, in our case of the italian regions. all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint caption: ( + ) date of epidemic onset february th ; (*) the original incidence data related to trento and bolzano were merged into a single region called trentino-alto adige resulting in a geographical disaggregation of italy into regions. all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. caption: official data from the dipartimento della protezione civile available at https://github.com/pcm-dpc/covid - ; (*) the original data related to trento and bolzano were merged into a single region called trentino-alto adige resulting in a geographical disaggregation of italy into regions; n = aggregate number of infected people, na = number of active infected people, da = density of active infected people. all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint figure s . daily incidence as numbers of new cases from february to april for covid- outbreaks (upper panel) and the corresponding time dependent reproduction number (rt) (lower panel) for the following italian regions: (a) abruzzo, (b) basilicata, (c) calabria, (d) campania, (e) emilia-romagna, (f) friuli v. g., (g) lazio. in upper panels vertical bars are the incidence data whereas in lower panels black dots are the rt mean values accompanying by grey vertical lines standing for % confidence intervals. in the same panel the horizontal solid line indicates the threshold value r = , above which an epidemic will spread and below which the epidemic is controlled. days are listed from the onset february th , . all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint figure s . daily incidence as numbers of new cases from february to april for covid- outbreaks (upper panel) and the corresponding time dependent reproduction number (rt) (lower panel) for the following italian regions: (a) liguria, (b) lombardia, (c) marche, (d) molise, (e) piemonte, (f) puglia, (g) sardegna, in upper panels vertical bars are the incidence data whereas in lower panels black dots are the rt mean values accompanying by grey vertical lines standing for % confidence intervals. in the same panel the horizontal solid line indicates the threshold value r = , above which an epidemic will spread and below which the epidemic is controlled. days are listed from the onset february th , . all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . in upper panels vertical bars are the incidence data whereas in lower panels black dots are the rt mean values accompanying by grey vertical lines standing for % confidence intervals. in the same panel the horizontal solid line indicates the threshold value r = , above which an epidemic will spread and below which the epidemic is controlled. days are listed from the onset february th , . all rights reserved. no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint covid- : preparedness, decentralization, and the hunt for patient zero covid- working group, epidemiological characteristics of covid- cases in italy and estimates of the reproductive numbers one month into the epidemic the r package: a toolbox to estimate reproduction numbers for epidemic outbreaks estimation in emerging epidemics: biases and remedies estimating the number of infections and the impact of non-pharmaceutical interventions on covid- in european countries the early phase of the covid- outbreak in lombardy, arxiv different epidemic curves for severe acute respiratory syndrome reveal similar impacts of control measures r: a language and environment for statistical computing. r foundation for statistical computing assessment of the sars-cov- basic reproduction number, r , based on the early phase of covid- outbreak in italy covid- outbreak progression in italian regions: approaching the peak by the end of march in northern italy and first week of april in southern italy int no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity guida alla lettura e all'interpretazione dei dati covid- covid- and italy: what next? lancet ( ) spread and dynamics of the covid- epidemic in italy: effects of emergency containment measures preliminary estimation of the basic reproduction number of novel coronavirus ( -ncov) in china, from to : a data-driven analysis in the early phase of the outbreak the effective reproduction number as a prelude to statistical estimation of time-dependent epidemic trends no reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted key: cord- -gtvi gh authors: flesia, luca; monaro, merylin; mazza, cristina; fietta, valentina; colicino, elena; segatto, barbara; roma, paolo title: predicting perceived stress related to the covid- outbreak through stable psychological traits and machine learning models date: - - journal: j clin med doi: . /jcm sha: doc_id: cord_uid: gtvi gh the global sars-cov- outbreak and subsequent lockdown had a significant impact on people’s daily lives, with strong implications for stress levels due to the threat of contagion and restrictions to freedom. given the link between high stress levels and adverse physical and mental consequences, the covid- pandemic is certainly a global public health issue. in the present study, we assessed the effect of the pandemic on stress levels in n = italian adults, and characterized more vulnerable individuals on the basis of sociodemographic features and stable psychological traits. a set of psycho-social variables, generalized regressions, and predictive machine learning approaches were leveraged. we identified higher levels of perceived stress in the study sample relative to italian normative values. higher levels of distress were found in women, participants with lower income, and participants living with others. higher rates of emotional stability and self-control, as well as a positive coping style and internal locus of control, emerged as protective factors. predictive learning models identified participants with high perceived stress, with a sensitivity greater than %. the results suggest a characterization of people who are more vulnerable to experiencing high levels of stress during the covid- pandemic. this characterization may contribute to early and targeted intervention strategies. sars-cov- (severe acute respiratory syndrome coronavirus ; henceforth referred to as covid- ) is a strain of coronavirus that can infect humans, attacking the lungs and causing symptoms ranging from those of the common cold to those of severe acute respiratory syndrome (sars) [ ] . while approximately % of those who are infected recover with no special treatment (i.e., they are either asymptomatic or suffer from mild pneumonia) [ ] , recent data have confirmed that older persons ( + years old) [ ] and persons with certain pre-existing medical conditions are more likely to develop serious respiratory distress that can lead to death ( - % of the population) [ ] . covid- spreads very easily between persons and, at the time of writing, no drugs or biologics have proven effective for preventing or treating the virus [ , ] . covid- was first identified in the chinese region of wuhan in december [ ] . between december and april , the virus spread throughout the world, causing more than , , infections and over , deaths [ ] . on march , the world health organization (who) declared covid- a pandemic [ ] . to contain the number of victims and prevent the collapse of the healthcare system, most national governments imposed strict restrictions on residents' freedom, forcing those infected with the virus to self-isolate and requiring all residents (infected or not) to stay at home and move through public areas only for reasons of absolute necessity. among western governments, the italian government was the first to apply such restrictions [ ] . a few weeks after the spread of covid- in china, the first scientific studies investigating the psychological impact of the outbreak highlighted a mild to severe negative psychological impact of the event within a significant proportion of the chinese population ( . %); specifically, this impact on mental health was reported to include symptoms of anxiety, depression, and stress [ , ] . more broadly, the literature on the impact of infectious outbreaks on mental health shows that pandemics are extremely stressful events that force people to cope with totally unexpected, ambiguous, and uncertain situations [ ] . specifically, two main aspects of pandemics have been found to affect people's mental state. the first relates to danger (i.e., the fear of contagion), which can increase perceived threat and sometimes lead to panic, behavioral contagion, and an emotional epidemic [ , ] . the second regards the multiple and rapid changes to social, working, and familiar habits, due to self-isolation and social distancing measures [ ] [ ] [ ] [ ] [ ] . the longer the duration of self-isolation, the more people experience frustration and boredom, along with concerns about infection [ ] . well-documented psychological reactions to epidemics include emotional distress, anxiety behaviors, sleeping disorders, fear, anger, depression, health concerns, a sense of powerlessness, and uncertainty [ , , [ ] [ ] [ ] [ ] [ ] . furthermore, studies examining the long-term consequences of infectious epidemics have shown that some individuals may even develop symptoms of post-traumatic stress disorder (ptsd) [ , , ] . one review indicated that those who develop ptsd may experience the symptoms for years following the end of the epidemic [ , ] . stress is defined as an adaptive psycho-physical reaction to a physical, social or psychological stimulus, called a stressor [ ] . stress-related responses may be cognitive, emotional, behavioral, or physiological. depending on the type, timing, and severity of exposure to a stressor, the resulting stress may become a risk factor for a number of illnesses, including those of a psychiatric or cardiovascular nature [ ] [ ] [ ] [ ] [ ] . an emergency such as the covid- outbreak might rightly be considered a severe stressor, as it is a new and unexpected situation with a potentially serious impact on health (experienced both personally and through loved ones) that also involves social restrictions [ ] . nevertheless, no event, in and of itself, is the precipitating cause of pathology and illness. rather, it is the perception of stress (i.e., the degree to which one considers the event stressful) that accounts for the varying physical and mental responses to the situation [ ] . in this sense, it is important to detect vulnerable persons early, and to promote effective preventive programs in order to treat such persons rapidly and limit negative psychological outcomes. the identification of psycho-social risk and resilience factors for psychological distress during the covid- emergency comprises a significant step in this direction [ ] . to date, studies on the psychological impact of covid- have mainly focused on the role of sociodemographic variables (e.g., gender, age, education level, and social connections) in moderating reactions to the outbreak [ , , , , ] . the identified sociodemographic risk factors for psychological distress include gender (female), age ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) or + years old), student status, education level, perception of the public health system, specific physical symptoms (e.g., coryza, cough, sore throat, headache), and a low reported level of health [ , , ] . for chinese students, living in an urban area, having a stable family income, living with parents, and having good social support were also found to protect against anxiety [ ] . in addition to investigating sociodemographic factors, many studies have also outlined the role of certain dispositional traits in modulating responses to stressful events. however, these constructs have been poorly investigated in relation to the psychological impact of epidemics and, specifically, covid- . more generally, research has shown how individual differences, including dispositional traits, can explain life outcomes [ ] . according to the theory of traits (or dispositional theory), individual differences may be explained by certain predispositions (traits), which are expressed in a relatively stable way across situations and time. traits comprise a person's manner of thinking, feeling, perceiving, and relating to others [ ] . based on these considerations, dispositional traits might play a relevant role in predicting perceived stress in relation to coping is one of the most widely studied dispositional traits, and it has been found to be significant in modulating responses to stressful events. coping is defined as the effort to solve personal and interpersonal problems in an attempt to master, minimize, or tolerate stress and conflict [ ] . distinct coping strategies have been found to be differentially associated with specific emotional responses [ ] , physiological stress responses [ ] , and self-efficacy [ , ] . one investigation into the role of coping strategies during a virus outbreak ( h n flu pandemic) found coping style to influence the perceived risk of contagion and vaccination intentions among canadian adults [ ] . another study, based in singapore, found coping strategies to be associated with post-traumatic outcomes within visitors to community health care services during the national outbreak of sars [ ] . as regards emotional self-regulation and adaptation to the world, self-control may represent a significant protective factor. there is empirical evidence that people with high dispositional self-control have better psychological adjustment and impulse control [ ] ; this suggests that good self-control may mitigate the influence of a negative environment. similarly, perceived control over life outcomes has been shown to be positively associated with well-being and health-related quality of life, and negatively associated with emotional distress, in the context of stressful events [ , ] . in this regard, locus of control [ ] is a relevant concept, describing the degree to which an individual believes that they have control over the outcome of life events, as opposed to feeling that their life is subject to external forces beyond their control. finally, among the many individual difference variables that might influence reactions to covid- , personality traits merit significant attention. several studies have highlighted an association between the big five personality traits [ ] and various health behaviors, such as sedentary behavior [ ] , sexual health behavior [ ] , physical activity [ ] , and alcohol consumption [ ] . the first aim of the present study was to investigate the impact of the covid- pandemic and the related government-imposed restrictions on perceived stress in a western country (i.e., italy). as reported above, most studies on the psychological impact of covid- have related to the chinese population. however, countries differ from one another in many important aspects (i.e., social, cultural, political, and economic aspects, to name only a few); consequently, psychological responses may also vary between contexts and communities, revealing unique qualitative and quantitative psychological reactions and psychological needs. our sample of italian adults, collected during the covid- outbreak, would show higher levels of perceived stress compared to italian normative values. the second aim of the study was to confirm the role of certain sociodemographic factors in modulating stress responses to the covid- pandemic, as reported in the recent literature. (a) participants who were female, younger, and students, with a lower level of education and lower income, would report higher levels of stress, and (b) participants who were living with others would report lower levels of psychological distress. the third aim of the work was to investigate the association between certain stable psychological traits and psychological distress relating to the current situation. to this end, participants were tested for coping strategies, self-control, locus of control, and select personality traits. (a) participants with positive coping strategies, higher levels of self-control, an internal locus of control, and higher levels of emotional stability would report lower levels of stress, and (b) participants with negative coping strategies, lower levels of self-control, an external locus of control, and lower levels of emotional stability would report higher levels of stress. finally, with the goal of anticipating persons in need of treatment and improving the targeting and overall effectiveness of preventive programs, we aimed at developing machine learning models to predict individual psychological responses to the covid- pandemic, based on sociodemographic and psychological variables with maximal sensitivity in classifying subjects with high versus low levels of perceived stress. to summarize, the study was novel in the following two respects: first, it considered the role of not only sociodemographic variables, but also stable psychological traits, as predictors of a stressful reaction to covid- ; and second, it leveraged machine learning techniques to identify people at the greatest risk of developing severe and negative psychological outcomes due to the pandemic. to test the abovementioned hypotheses, we implemented a cross-sectional study. using google forms, we designed an ad hoc online questionnaire to collect data on participants' stress reactions to covid- , demographical variables, and psychological traits. the questionnaire also assessed the following sociodemographic factors: gender, age, education, number of family members and/or others living in the household, monthly household income, and student status. subsequently, we administered five standardized questionnaires, as follows: • the italian version of the -item perceived stress scale (pss- ; cronbach's alpha = . ) [ ] . the pss- is a frequently used psychological instrument to measure perceived stress [ ] . respondents are asked to answer questions pertaining to the frequency of experiences of stressful situations during the last month on a five-point scale ranging from (never) to (very often) [ , ] . example items include "in the last month, how often have you been upset because of something that happened unexpectedly?" higher scores indicate higher levels of perceived stress. moreover, in the present study, the score corresponding to . sd above the italian normative score [ ] was used as a cut-off to divide participants into two classes: low perceived stress (males: pss- score < . ; females: pss- score < . ) and high perceived stress (males: pss- score ≥ . ; females: pss score- ≥ . ); • the italian shortened version of the coping orientations to the problems experienced (cope-nvi- ; cronbach's alpha of factors range . - . ) [ ] . the cope-nvi- is a multi-dimensional inventory that assesses individual differences in coping styles. it is comprised of items, which are rated on a -point scale ranging from (i usually don't do this at all) to (i usually do this a lot) [ ] . the instrument includes five subscales corresponding to five different coping styles: social support, avoidance strategies, positive attitude, problem solving, and turning to religion [ ] . an example item is "i admit to myself that i can't deal with it, and quit trying" (avoidance strategies). a higher score on a particular subscale indicates a greater use of that specific coping strategy. the italian translation of the brief self-control scale (bscs; cronbach's alpha = . ) [ ] . the bscs measures individual differences in dispositional capacity for self-control. the scale is comprised of items that are rated on a five-point scale ranging from (not at all) to (very much). an example item is "i do certain things that are bad for me, if they are fun." higher scores on the bscs indicate a greater capacity for self-control, and they are also correlated with better psychological adjustment, interpersonal skills, and emotional responses [ ] ; • the italian short version of the locus of control (loc) scale [ ] . this -item questionnaire is used to measure generalized expectancies relating to an internal versus external locus of control, rated via dichotomous options ("yes" vs. "no"), similar to rotter's original internal-external locus of control scale [ ] . an example item is "to do well in life, luck is more important than commitment." respondents with an internal locus of control (i.e., a high score on the internal loc scale) tend to attribute life outcomes-and general life events-to their own behavior, whereas those with a prevalent external locus of control (i.e., a high score on the external loc scale) tend to attribute life events to fate, others, or external causes beyond their control [ , ] ; • the italian version of the -item big five inventory (bfi- ; spearman-brown coefficients ≥ . ) [ ] . the bfi- assesses personality traits according to the five-factor approach [ ] . it is comprised of items rated on a five-point scale ranging from (strongly disagree) to (strongly agree), measuring five dimensions of personality, which are extraversion, agreeableness, conscientiousness, neuroticism (or, if reversed, emotional stability), and openness [ ] . an example item is "i see myself as someone who is outgoing, sociable" (extraversion). the higher the score on a particular subscale, the more that specific dimension represents a characteristic trait of the respondent's personality. the complete list of variables that were extracted from the responses to the questionnaire is provided in the supplementary materials. the present research was designed in accordance with the declaration of helsinki and approved by the ethics committee for psychological research at the university of padova (protocol number , unique code b fe f a d b d bb). data were collected during the period of - march, . participants were recruited online through an invitation posted on social media (facebook and whatsapp). this approach of online recruitment was selected primarily due to the lockdown situation, which prevented us from collecting data in the laboratory. according to the aim of the study, it was necessary for us to capture the psychological state of participants at the time of the pandemic; thus an a posteriori study would not have provided useful and reliable information. participants were invited to complete an anonymous online questionnaire to report their personal experiences with the covid- emergency and their mental state. the inclusion criteria were the following: (a) living in italy at the time of data collection and (b) being aged + years ( years is the legal age in italy, defined by the capacity to act and be emancipated). participation was voluntary. all participants were required to read and provide informed consent before beginning the online questionnaire. they received no compensation for their participation. in total, volunteers took part in the study. of these, were excluded on the basis that they responded to the questionnaire twice (we kept only their first response). thus, the final sample was comprised of participants, of whom were female, were male, and were reported as "other." the participants' average age was . (sd = . ; range: and their average education level was . years (sd = . ; range: - ). a more detailed description of the sample's demographic characteristics is provided in the supplementary materials. it has been calculated that a sample size of is sufficiently large to achieve at least a statistical power ( -β) = . in a linear multivariable regression analysis involving predictors, given a significance level α = . and an effect size of . [ ] . data are provided in the supplementary materials. data analysis was conducted using the jasp software [ ] . a single sample t-test (t, two-sided) was performed in relation to the pss- score, in order to determine whether the sample's true mean (µ) was statistically different from that of the known population (m ). a multivariable regression analysis was run to investigate the relationship between the pss- score and the independent variables that were hypothesized to impact the level of perceived stress. the collinearity assumption was checked prior to running the model, using the tolerance and variance inflation factor (vif). as a rule of thumb, if vif > and tolerance < . , the assumption is greatly violated, whereas if vif > and tolerance < . , the model may be biased [ ] . the results indicated that the collinearity assumption was not violated by any of the independent variables entered in the regression model. the analysis was performed using the stepwise variable selection method, which identified predictors with a significant (p < . ) individual association with the outcome (pss- score). the results were reported using unstandardized coefficients, as recommended by friedrich [ ] . recently, researchers in different scientific fields, including the clinical and social sciences, have emphasized the utility of focusing on prediction, rather than explanation, during data analysis [ ] [ ] [ ] [ ] . this increased attention to predictive models may be largely attributed to the significant spread of machine learning (ml)-a branch of artificial intelligence that trains algorithms on data samples (i.e., training sets) in order to make predictions on completely new data (i.e., test sets) without being explicitly programmed to do so [ ] . as regards psychology, ml techniques have been shown to be particularly useful for predicting human behavior, including high-risk behavior; thus, they may be applied to improve the effectiveness and targeting of preventive programs and interventions [ ] . in brief, ml models are capable of predicting the behavior of individual subjects, allowing greater attention to be paid to those considered most critical [ ] . in the present study, ml algorithms were trained on psycho-social data to identify subjects who were more likely to present high levels of perceived stress during the covid- emergency, and who were consequently at the greatest risk of developing psychological symptoms, including those of ptsd. for this purpose, participants were split into two classes: high perceived stress and low perceived stress. the high perceived stress class included participants with a pss- score of more than . sd above the italian population mean (n = ) for men and women, respectively. conversely, the low perceived stress class included participants whose pss- did not exceed . sd above the italian normative value (n = ). it should be noted that participants who reported their gender as "other" (n = ) were excluded from this analysis, as the italian normative values were available for males and females only [ ] . as ml models are built to fit particular data, it is important to test how each model fits new (i.e., unseen) data. for this reason, part of the data (the training set) is generally used to train and validate the model, while another part (the test set) is used to test the model's accuracy on new examples [ , ] . this procedure guarantees the model generalization and increases the replicability of the results [ , ] . in the present study, % [ , ] of the participants were randomly chosen and retained as the test set. accordingly, the training set consisted of participants ( with high perceived stress and with low perceived stress), and the test set consisted of participants ( with high perceived stress and with low perceived stress). in the first step, feature selection was performed to remove redundant and irrelevant features and to increase model generalization by reducing overfitting and noise in the data [ ] . a good strategy for feature selection is to identify the subset of features that are highly correlated with the class to predict, but not correlated with each other [ ] . this procedure was performed in the present study using the correlation-based feature selector (cfs) in the weka . software [ ] . the problem of class imbalance was addressed while running the classification algorithms. the ratio between participants with high perceived stress and those with low perceived stress was approximately : . as ml methods work best with balanced datasets, it is necessary to account for any class imbalance, especially when training examples are limited-a condition that is frequently met by datasets in health and clinical psychology [ ] . at the same time, it is equally important for ml models to be built on samples that are representative of the population, reflecting real distribution [ ] . one strategy to overcome these two limitations consists of altering the relative costs associated with misclassifying the minority and majority classes, in order to compensate for the class imbalance [ ] . in the present study, ml algorithms were set in such a way that any algorithmic error made in classifying the minority class (high perceived stress) was weighted four times more than any error in classifying the majority class (low perceived stress). this cost-modifying strategy has been shown to provide better results than other methods in addressing the class imbalance problem [ ] . moreover, it should be noted that, for the goal of the present task, it was more beneficial to minimize false negatives than to minimize false positives (i.e., to have a model with high sensitivity rather than high specificity). in other words, it was more important to identify people who were truly at risk than to avoid misclassifying people who were not truly at risk. ml models were trained and validated on the training sample (n= ) through a -fold cross-validation procedure using the weka . software [ ] . the different algorithms (i.e., logistic regression [ ] , support vector machine (svm) [ ] , naïve bayes [ ] , random forest [ ] ) were chosen as representatives of different classification strategies, to ensure that the results would be stable across classifiers and not dependent on specific model assumptions (details on the parameters of the ml classifiers are reported in the supplementary materials). k-fold cross-validation is a resampling procedure that seeks to reduce the variance in model performance relative to the performance that may be obtained from a single training set and a single test set. the procedure consists of portioning the sample into k subsets (i.e., folds; in the present study, k = ), and using k- (i.e., ) subsets to train the model and the remaining subset to validate the model's accuracy. this is repeated k (i.e., ) times [ ] . the final model metrics are obtained by averaging the metrics obtained in all validation subsets. in the present study, the models developed from the -fold cross-validation procedure were tested on the test sample (n = ). the main results of the data analysis are reported in this section. a more complete descriptive analysis of each variable, including the composition of high perceived stress versus low perceived stress samples, is reported in the supplementary materials. the average pss- score of the entire sample was . (sd = . ). analyzing the responses of males and females separately (note that participants who reported a gender of "other" were excluded from this analysis due to a lack of normative data), males obtained an average score of . (sd = . ) and females obtained an average score of . (sd = . ). to determine whether the sample mean statistically differed from that of the italian normative population (males: average = . , sd = . ; females: average = . , sd = . ) [ ] , a one-sample t-test was run separately for each gender. a first multiple regression analysis was run, including sociodemographic variables that have been shown to potentially impact the level of perceived stress during a pandemic [ , , , , ] . the pss- score was set as the dependent variable, while gender (male), age, education, monthly income, number of family members, and student status (student) were entered as covariates. the final model accounted for a significant proportion of the variance in the level of perceived stress (r = . , adjusted r = . , f-change ( , ) = . , p < . ). all of the aforementioned variables, with the exception of student status, were found to contribute to the level of perceived stress. results are reported in table . to better understand the role of stable psychological traits in predicting the level of perceived stress (pss- score), a second multiple linear regression was run, adding to the previous model the scores of the five coping styles measured by the cope-nvi- (cope positive, cope problem, cope avoidance, cope religion and cope support), the bscs total score, the internal loc score, and the scores for the five personality traits measured by the bfi- (bfi- agreeableness, bfi- conscientiousness, bfi- emotional stability, bfi- extraversion and bfi- openness). this second model accounted for a larger proportion of the variance in the level of perceived stress (r = . , adjusted r = . , f-change ( , ) = . , p < . ) compared to the previous model. bfi- emotional stability, cope positive, age, bcsc total score, gender (male), cope avoidance, internal loc, number of family members, cope support, monthly income, and bfi- conscientiousness were identified as significant predictors of the level of perceived stress during the covid- epidemic (see table ). education, cope religion, cope problem solving, bfi- agreeableness, bfi- extraversion and bfi- openness were excluded. ultimately, the questionnaire variables were considered predictors of perceived stress. the entire list of predictors, along with their descriptions, is provided in the supplementary materials. of these variables, the following were identified as the best set of predictors using correlation-based feature selection: age, monthly income, cope avoidance, cope positive, bscs total score, bfi- emotional stability, and bfi- agreeableness. using these predictors, ml algorithms were trained and tested according to the procedure described in the "data analysis" section. classification results for the test set are reported in table , which quantifies predictive performance according to the following metrics: receiver operating characteristic curve (roc) area, precision, recall and f-measure (f score). it is worth noting that the classifiers showed an roc area ranging from . to . in the test set. however, the random forest algorithm highlighted the lower sensitivity (recall) of the high perceived stress class compared to the other classifiers, making it a weaker model for the purposes of prediction. the present study measured the impact of the covid- emergency on perceived levels of stress, taking into account sociodemographic variables and stable psychological traits. the results confirmed that participants perceived the covid- crisis as a stressful experience; in the present sample, the level of perceived stress was higher than that of the general population in a non-emergency condition. indeed, almost % of the sample scored above the results from the normative data on measures of perceived stress. these results are in line with the findings of recent studies on the psychological impact of covid- [ , , ] and the international literature on epidemic outbreaks [ ] . the mean values of the single items of the pss- suggest that, in addition to nervousness and stress, feelings of being unable to control one's personal life accounted for the majority of participants' perceived stress. this suggests that the unpredictability and uncontrollability of the pandemic may play a significant role in determining levels of perceived stress during the crisis. moreover, it may reflect participants' attitudes toward the significant lifestyle changes demanded of them due to the lockdown and other restrictive measures. as regards sociodemographic variables, the results suggest that the female gender is associated with higher levels of stress. this is consistent with the literature indicating gender differences in the psychological response to covid- [ , ] and other epidemics [ ] ; it is also in line with the normative data for the general population. consistent with other studies ( , ) , the present study found an association between higher incomes and lower levels of perceived stress. one explanation for this is that higher incomes might be related to less concern about the economic effects of self-isolation and/or with more comfortable housing solutions (e.g., larger living spaces, access to outdoor spaces (such as gardens), and access to leisure activities). moreover, people with higher incomes may be more likely to perform work that can easily and fully transition to the online environment, thereby reducing some sources of stress. in the present sample, older age was found to be associated with lower levels of stress. this finding might appear surprising, since it contradicts both the results of studies on the chinese population [ , ] and the association between older age and higher covid- mortality. however, the result is in line with recent italian data [ ] . several studies have indicated age-related differences in coping and locus of control, with older adults presenting greater self-control and emotional self-regulation relative to younger adults [ ] [ ] [ ] . considering the current pandemic, older people may be more used to staying at home, so their daily routines might be less impacted by mandatory self-isolation measures. data from previous investigations on age differences in stress responses to the sars epidemic reflect inconsistencies [ , ] , but sociopolitical and cultural aspects, such as differences in elder care services and policies, might account for these discrepancies. the present study did not find education to be a significant predictor of the level of perceived stress. the large percentage of highly educated participants in the sample might partially explain this finding. however, prior research on this subject has generated mixed results-recent studies on the chinese population [ , ] have found that education does not seem to affect mental health, while data from a spanish sample [ ] and from previous studies on psychological adjustment to sars [ ] have confirmed an association between a higher level of education and better mental health. the present study also found that living alone or with few family members was a protective factor against perceived stress. we might argue that this condition both conveys a sense of protection from contagion and offers continuity with pre-epidemic economic and social conditions. moreover, when cohabiting with family members, concern for loved ones might contribute to increasing perceived stress. as regards psychological variables, emotional stability was found to be an important protective factor. according to the five-factors model [ ] , people with high emotional stability remain calm in response to stressful situations, and view problems in proportion to their importance. as a result, they tend to worry less about problems than do people with low emotional stability [ ] . many studies have found that emotional stability is able to buffer stress responses to adverse events [ , ] . in the present study, conscientiousness and agreeableness were found to predict psychological distress. generally speaking, individuals who score high on agreeableness tend to dislike conflict and be less suspicious of others; generally, they seek to pacify and mediate. in this sense, agreeable people might be more flexible and accepting when faced with unexpected and undesired situations, such as restrictions and changes to daily routines. conscientious people are more likely to perceive lower levels of stress (see correlation analysis). according to the literature, they are more aware of their actions and tend to exhibit more goal-oriented behavior. in this sense, in times of self-isolation, conscientious people might have a greater tolerance for frustration and imposition relative to less conscientious people, who might engage in more impulsive behavior [ ] . moreover, previous research has indicated that conscientiousness may influence adaptive behavior, especially in health-related programs [ , ] . the present study found higher levels of dispositional self-control to predict lower levels of psychological distress in response to the covid- emergency. personal self-control skills may play a role in determining tolerance to restrictions to personal freedom during self-isolation. this result further suggests that self-regulatory processes may have a strong influence on responses to the outbreak. the results regarding dispositional self-control are consistent with those relating to the emotional stability trait of the big five model. in fact, these dimensions are often correlated [ ] , and this specific pattern may indicate the importance of personal skills, such as the ability to remain calm, and maintain emotional balance and a sense of acceptance. in this sense, practices that enhance emotional stability and acceptance, such as mindfulness, could be useful in reducing the stressful impact of the emergency [ , ] . the results concerning coping styles are also in line with this. besides confirming the protective effect of functional coping styles and the adverse impact of dysfunctional coping styles, the results of the present study suggest that people who use a positive attitude as a coping strategy may be much less likely to experience psychological distress during the present emergency. such persons may appraise the emergency as a unique opportunity, and feel less need for psychological support. in contrast, the present findings suggest that people who use avoidance strategies may be more likely to experience higher levels of stress during the emergency. these results are consistent with the findings of previous investigations into the relation between coping style and response to an epidemic [ , , , ] . the results relating to dispositional locus of control indicate that people with an internal locus of control may be less likely to feel stressed. again, these results suggest that the more people are inclined to confidently rely on themselves, the better they will cope with uncertainty and change. several studies have indicated an association between an internal locus of control and self-efficacy and emotional stability [ , ] . furthermore, previous studies have found a relation between an internal locus of control and the positive appraisal of an emerging infectious disease outbreak [ ] . it could be hypothesized that people with an internal locus of control interpret self-isolating as something that they determine and enact for themselves as a protective behavior, rather than something that is imposed on them; this might account for their lower levels of perceived stress. overall, the results of the present study identify some population subgroups that may be more vulnerable to experiencing stress during the covid- emergency. specifically, a set of seven psycho-social variables may identify a high percentage of people experiencing high stress during the covid- pandemic, with sensitivity approaching . (roc area of predictive models ranging between . and . ). according to this model, we may develop targeted preventive interventions. furthermore, self-regulatory skills (including emotional stability, an internal locus of control, and self-control) were shown to be a protective factor, indicating the importance of raising awareness of these skills during the emergency and offering training and education to increase personal abilities in these areas (e.g., mindfulness programs). the present research aimed at improving our understanding of the possible risk and protective factors for high perceived stress during the covid- outbreak. it is worth noting that all data were collected from an italian population. therefore, the findings were inevitably influenced by specific contextual and socio-cultural aspects. further investigations involving people of different ethnicities and residents of other countries would deepen our understanding of the generalizability of these results, and the effective influence of psychological traits. in this regard, the open access data reported in the supplementary materials may contribute to facilitating comparisons between ethnicities, countries, and specific traumatic events. some further limitations should be considered when interpreting the findings of this research. first, participants were recruited via an online link posted on social networks. while online recruitment guarantees large samples, it does not guarantee sample representativeness. for this reason, very vulnerable groups, such as homeless or low-income persons, may not be well represented in this study. similarly, the average age of the sample was young, predominantly female, and largely well educated, as indicative of a sample that is more likely to participate in an online survey. second, the use of self-report measures did not enable us to verify the reliability of the responses, or to ensure that participants correctly understood the questions. future research should aim at overcoming these shortcomings. finally, future research should also investigate the interplay and mutual interrelationship between protective and risk factors, to improve the targeting and overall effectiveness of preventive programs and interventions. indeed, the literature suggests that, during a pandemic, it is extremely important for people to sustain their use of psychological services, either online or in the context of social distancing [ , ] . this is particularly essential for those who are more vulnerable to experiencing high levels of stress, and it is important that we ensure that such persons can access timely and high-quality psychological services in order to prevent the development of chronic outcomes, including ptsd. supplementary materials: the following are available online at http://www.mdpi.com/ - / / / /s , table s : descriptive statistics, table s : item by item analysis of the pss- , table s : list of predictors, table s : details on ml classifiers parameters. funding: this research received no external funding. the authors declare no conflict of interest. the epidemiology and pathogenesis of coronavirus disease (covid- ) outbreak characteristics of and important lessons from the coronavirus disease (covid- ) outbreak in china supporting older people during the covid- pandemic is everyone's business unique epidemiological and clinical features of the emerging novel coronavirus pneumonia (covid- ) implicate special control measures treatment of covid- : old tricks for new challenges the sars-cov- outbreak: what we know who coronavirus disease (covid- ) dashboard. available online who director-general's opening remarks at the mission briefing on covid- pandemic and lockdown: a territorial approach to covid- in china, italy and the united states immediate psychological responses and associated factors during the initial stage of the coronavirus disease (covid- ) epidemic among the general population in china the psychological impact of the covid- epidemic on college students in china long-term psychiatric morbidities among sars survivors psychiatry of pandemics: a mental health response to infection outbreak vicarious traumatization in the general public, members, and non-members of medical teams aiding in covid- control the psychological impact of quarantine and how to reduce it: rapid review of the evidence timely mental health care for the novel coronavirus outbreak is urgently needed how personality relates to distress in parents during the covid- lockdown: the mediating role of child's emotional and behavioral difficulties and the moderating effect of living with other people the effect of the covid- lockdown on parents: a call to adopt urgent measures how to improve compliance with protective health measures during the covid- outbreak: testing a moderated mediation model and machine learning algorithms psychological responses to outbreak of severe acute respiratory syndrome: a prospective, multiple time-point study how health anxiety influences responses to viral outbreaks like covid- : what all decision-makers, health authorities, and health care professionals need to know the psychological burden experienced by hong kong midlife women during the sars epidemic psychosocial and coping responses within the community health care setting towards a national outbreak of an infectious disease prevalence and predictors of ptss during covid- outbreak in china hardest-hit areas: gender differences matter predictors of symptoms of posttraumatic stress in chinese university students during the h n influenza pandemic individual differences, economic stability, and fear of contagion as risk factors for ptsd symptoms in the covid- emergency neurobiological and systemic effects of chronic stress effects of lifetime stress exposure on mental and physical health in young adulthood: how stress degrades and forgiveness protects health the effects of psychological stress on depression chronic stress, hair cortisol and depression: a prospective and longitudinal study of medical internship heightened biological stress response during exposure to a trauma film predicts an increase in intrusive memories a direct test of the diathesis-stress model for depression psychological stress and coping in adaptation and illness recommended psychological crisis intervention response to the novel coronavirus pneumonia outbreak in china: a model of west china hospital. precis a nationwide survey of psychological distress among chinese people in the covid- epidemic: implications and policy recommendations the impact of covid- epidemic declaration on psychological consequences: a study on active weibo users the power of personality: the comparative validity of personality traits, socioeconomic status, and cognitive ability for predicting important life outcomes an analysis of coping in a middle-aged community sample coping as a mediator of emotion coping as a moderator of relations between reactivity to interpersonal stress, health status, and internalizing problems human agency in social cognitive theory self-efficacy: the exercise of control h n influenza pandemic: the role of threat, coping, and media trust on vaccination intentions in canada high self-control predicts good adjustment, less pathology, better grades, and interpersonal success harpaz-rotem, i. locus of control in us combat veterans: unique associations with posttraumatic stress disorder -factor model symptom clusters locus of control', health-related quality of life, emotional distress and disability in parkinson's disease generalized expectancies for internal versus external control of reinforcement personality in adulthood personality and sedentary behavior: a systematic review and meta-analysis. health psychol linking big five personality traits to sexuality and sexual health: a meta-analytic review personality and physical activity: a systematic review and meta-analysis alcohol involvement and the five-factor model of personality: a meta-analysis psychometric evaluation of three versions of the italian perceived stress scale the perceived stress scale (pss- ) in women experiencing infertility: a reliability and validity study a global measure of perceived stress positive events and social supports as buffers of life change stress cope-nvi- : validazione italiana della versione ridotta della coping orientation to the problems experienced (cope-nvi) assessing coping strategies: a theoretically based approach coping orientation to problems experienced-nuova versione italiana (cope-nvi): uno strumento per la misura degli stili di coping vivere con il sole in tasca: cos' è la psicologia positiva e cosa può fare per noi contributo alla taratura italiana della scala di controllo interno-esterno (i-e) di rotter an italian version of the -item big five inventory: an application to hedonic and utilitarian shopping values the stability of personality: observations and evaluations measuring personality in one minute or less: a -item short version of the big five inventory in english and german power : a flexible statistical power analysis program for the social, behavioral, and biomedical sciences statistical analysis in jasp: a guide for students in defense of multiplicative terms in multiple regression equations choosing prediction over explanation in psychology: lessons from machine learning indicators to distinguish symptom accentuators from symptom producers in individuals with a diagnosed adjustment disorder: a pilot study on inconsistency subtypes using sims and mmpi- -rf introducing machine learning to detect personality faking-good in a male sample: a new model based on minnesota multiphasic personality inventory- restructured form scales and reaction times the development of a short version of the sims using machine learning to detect feigning in forensic assessment machine learning in psychometrics and psychological research machine learning approaches for clinical psychology and psychiatry nonlinear system identification replication and p intervals: p values predict the future only vaguely, but confidence intervals do much better the reusable holdout: preserving validity in adaptive data analysis correlation-based feature selection for machine learning the weka data mining software the class imbalance problem: significance and strategies the class imbalance problem: a systematic study ridge estimators in logistic regression improvements to platt's smo algorithm for svm classifier design estimating continuous distributions in bayesian classifiers random forest a study of cross-validation and bootstrap for accuracy estimation and model selection a nationwide survey of psychological distress among italian people during the covid- pandemic: immediate psychological responses and associated factors association of self-perceptions of aging, personal and family resources, and loneliness with psychological distress during the lock-down period of covid- narrative synthesis of psychological and coping responses towards emerging infectious disease outbreaks in the general population: practical considerations for the covid- pandemic dynamic integration: affect, cognition, and the self in adulthood social and emotional patterns in adulthood: support for socioemotional selectivity theory population-based post-crisis psychological distress: an example from the sars outbreak in taiwan relations of sars-related stressors and coping to chinese college students' psychological adjustment during the beijing sars epidemic social support and psychological adjustment to sars: the mediating role of self-care self-efficacy the "big five questionnaire": a new questionnaire to assess the five factor model emotional stability is associated with the maoa promoter uvntr polymorphism in women self-efficacy and emotional stability buffer negative effects of adverse childhood experiences on young adult health-related quality of life the influence of personality on trajectories of distress, health and functioning in mild-to-moderately depressed adults with type diabetes impact of baseline patient characteristics on interventions to reduce diabetes distress: the role of personal conscientiousness and diabetes self-efficacy higher-order factors of the big five let's conquer covid- and sustain our abilities mindfulness mediates the physiological markers of stress: systematic review and meta-analysis coping responses of emergency physicians and nurses to the severe acute respiratory syndrome outbreak the importance of coping appraisal in behavioural responses to pandemic flu relationship of core self-evaluations traits-self-esteem, generalized self-efficacy, locus of control, and emotional stability-with job satisfaction and job performance: a meta-analysis perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior psychological interventions for people affected by the covid- epidemic the role of telehealth in reducing the mental health burden from covid- key: cord- -qz g v u authors: livadiotis, george title: statistical analysis of the impact of environmental temperature on the exponential growth rate of cases infected by covid- date: - - journal: nan doi: . / . . . sha: doc_id: cord_uid: qz g v u we perform a statistical analysis for understanding the effect of the environmental temperature on the exponential growth rate of the cases infected by covid- for us and italian regions. in particular, we analyze the datasets of regional infected cases, derive the growth rates for regions characterized by readable exponential growth phase in their evolution spread curve and plot them against the environmental temperatures averaged within the same regions, derive the relationship between temperature and growth rate, and evaluate its statistical confidence. the results clearly support the first reported statistically significant relationship of negative correlation between the average environmental temperature and exponential growth rates of the infected cases. the critical temperature, which eliminates the exponential growth, and thus the covid- spread in us regions, is estimated to be tc = . ± . f. the daily number of new cases infected by covid- is currently exponentially growing for most countries affected by the virus. however, this exponential growth rate varies significantly for different regions over the globe. it is urgent and timely to understand the reasons behind this regional variation of the exponential growth rates. little information is known about this matter, while there are indications that the environmental temperature may be a factor; for instance, northern and colder us and italian regions experienced much more incidents than others. typically, the evolution curve of the spread of the coronavirus initiates with a pre-exponential phase characterized by a mild logarithmic growth, followed by the outburst phase of the exponential growth. social-distancing measures against the spread may affect the evolution curve in a way that the exponential growth slows down and starts to decay, depending though on the effectiveness and applicability of these measures. however, after the decay of the spread at some place, new infected cases may outburst in other places, marked with insignificant number of cases until that moment. then, a newly growth phase may appear. for example, fig. (left) shows the evolution curve of spread for the infected cases in mainland china; clearly, we observe the whole growth−decay cycle, as well as, a new re-growth phase. super-strict measures, such as complete shut down and quarantines, can successfully lead to the deceleration of the exponential growth of infected cases (giugliano, ) . unfortunately, they cannot be successfully applied and followed within vast regions, and especially, for a long and indefinite period of time. inevitably, measures may be loosened during the decay phase, leading to the birth of an equally disastrous re-growth phase. us & italy (right); phases (color-coded): pre-exponential (pre-exp), exponential (exp) growth, decelerated growth, decay, and possibly, a re-growth. day t= corresponds to / / for china, / / for italy, / / for us. evolution in china cases follows the whole growth-decay cycle, and a new re-growth phase. italian cases are characterized by a milder exponential rate, entered the phase of decelerated growth on march . us suffers with a larger exponential rate, and it is not clear whether has entered the decelerated growth phase. the exponential growth rate for china rose as high as λ= . ± . , while for italy and us the rates were λ= . ± . and . ± . , respectively (with correlation coefficient > . ). the exponential growth is the most effective phase for the evolution curve of infected cases; and the most important question regarding this evolution is still open (black et al. ) : what can influence the exponential growth rate, and thus, "flatten the curve"? measures, strict or not, may affect the evolution of new infected cases, by shifting the spread curve from the exponential to the decelerated growth. it should be noted though that measures do not affect the exponential growth rate itself, but only the period of time that this exponential phase applies. then, what factors do affect the exponential growth rate? the age distribution in the place where the outburst occurs is unlikely to be a factor; indeed, the number of new cases is known to be positively correlated with age, however, the exponential growth rate (china: . ; us: . ; italy: . -decreasing rate) appears to be negatively correlated to the age median of these countries (china: . ; us: . ; italy: . -increasing age); hence, the age is likely irrelevant to the rate variations. in addition, culture in social activities may be a factor; for example, this might be contributing in the observed differences among the exponential rates in the cases of china, italy, and us ( fig. ) . however, what is causing the major variation of exponential rates among different regions of the same culture? it is apparent that culture does not constitute the main factor influencing the exponential rate. figure shows the regional variation of infected cases (left) and average winter temperature (right) in italy. the possible negative correlation, observed between regional number of infected cases and winter temperature in italy, is an indication of the influence of temperature on the exponential growth, but it certainly does not constitute a necessary condition. the reason is that the map plots the total number of the infected cases n t , which does not depend only on the exponential rate λ, but also on the initial number of cases n . it is generally accepted that the initial infected cases in italy were travelled directly from china; since some destinations are more favorable than others, then, the initial number of cases n , as well as the current number of cases n t (which is proportional to n ), should be subject of regional variation. therefore, there is a non-negligible possibility, the observed regional variation of the number of infected cases n t to be caused by the regional distribution of the initial cases n . in such a case, main airport cities would have incredibly high number of infected cases outplaying a possible negative correlation of daily infected cases with regional average temperature t; the latter may be one of the reasons of the high numbers of cases observed in new york city and rome. on the other hand, in their letter to the white house, members of a national academy of sciences committee said that "there is some evidence to suggest that [coronavirus] may transmit less efficiently in environments with higher ambient temperature and humidity; however, given the lack of host immunity globally, this reduction in transmission efficiency may not lead to a significant reduction in disease spread without the concomitant adoption of major public health interventions" (relman, ) . nevertheless, it has to be stressed out that there were no statistical analyses focused on the exponential growth rates of the infected cases in regions with different temperatures. for instance, several authors (e.g., pawar et al., ; yao et al., ) found insignificant correlations between temperatures and confirmed cases. however, their analysis was performed on the number of the infected cases n t , which is subject to the randomness of the initial cases n as explained above, and not to the exponential growth rate λ, which is dependent on physical characteristics of the coronavirus, binding protein, and environment. analysis of regional cases can show whether the speculated negative correlation between temperature and number of infected cases is true, meaning a negative correlation between temperature and exponential growth rate. if the environmental temperature plays indeed a substantial role on the virus spread, then, this can provide promising results, such as, the estimation of the critical temperature that may eliminate the number of daily new cases in heavily infected regions. the purpose of this paper is to improve understanding of the effect of environmental temperature on the spread of covid- and its exponential growth rate. then, we calculate the exponential growth rates of infected cases for us and italian regions, derive the relationship of these rates with the environmental temperature, evaluate its statistical confidence, and determine the critical temperature that eliminates this rate. a standard model for describing the evolution of the infected cases by viruses is given by ( ) ( ) ; n t is the number of total infected cases evolved from the initial n cases, n max is the maximum possible number of infected cases; λ is the exponential growth rate, and becomes clear for x t << , where i is negligible, leading to: the function of negative feedback i models factors that flattens the curve, such as, the measures taken against spreading. while these factors are not affecting the exponential growth rate λ, they become more effective as the number of cases increases, getting closer to n max ; exponent b controls the effectiveness of these factors; strict {loose} measures correspond to smaller {larger} values of b. figure shows the evolution curve of the number of new (Δn t = n t+ −n t ) and total infected cases (n t ) and how this curve flattens for stricter measures (smaller values of b). as observed in fig. (b) , stricter measures, nicely modeled by decreasing b, do not affect the exponential rate λ but they successfully flatten the curve. however, the same can be achieved by downgrading the exponential growth rate. it is apparent, then, how much useful would it be to know the factors that can flatten the curve by decreasing directly the exponential rate. applied measures could be loosen and shorter! model ( ) originates from the logistic map family (e.g., livadiotis, ; and references therein; wu et al. ) ; other complicate versions, such as, the susceptible-infectious-recovered models (e.g., ciarochi, ) may be expressed by multi-dimensional differential or difference equations (e.g., elaydi, , and references therein; livadiotis et al. ), but still, the curve flattening is governed by the same features. the two composites, the exponential growth e and the negative feedback i, are just the main and necessary conditions for reproducing the growth-decay phases of the spread curve. their interplay shows how the spread curve can be flattened as a result of stricter measures, independently of the existent exponential rate. . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . what are the main factors that can affect the exponential growth rate λ of covid- spread? the rate λ is expected to have positive correlation with the reproduction number r (e.g., proportional to its logarithm), and negative correlation with the incubation period τ (e.g., inverse proportional) (milligan and barrett, ) . the number r is a measure of how contagious a disease is; it provides the average number of people in a susceptible population that a single infected person will spread the disease to over the course of their infection (ciarochi, ) , and depends on the physical characteristics of coronavirus (hao ) . the incubation period τ is the time elapsed between exposure to coronavirus and first symptoms; during this period, an infected individual cannot infect others; other characteristic periods and time intervals are the latent period between exposure and infection, and the generation time, mostly concerned with transmission process (nishiura, ) . characteristic values for covid- are τ~ - days and r ~ - (chen, ) . the rate expression can be written as , and involves all the physical characteristics of the mechanisms of infection and the environmental interactions; this can be easily derived, considering difference equations (that is, iterated discrete maps) (e.g., see: livadiotis and elaydi, ; kwessi et al., ; dayeh et al, ) . setting the time to be given in discrete τ-steps, then, which can be written in terms of eq. ( ), where the exponential rate is given by: the main factors that can affect the exponential rate λ are: (a) culture in social activities, and (b) environmental temperature and/or other thermodynamic parameters. intense cultural and social activities have reasonably a positive correlation with r . as previously mentioned, measures against the virus spread do not effectively influence the exponential growth rate; e.g., they do not change the culture in social activities, which are characteristics of the particular population, but they can just cease these activities for some period of time. on the other hand, the environmental temperature t can affect all the parameters influencing exponential rate. we approach this dependence by (i) a linear approximation of the phenomenological relationship between exponential rate and temperature, and (ii) the connection of reproduction number with arrhenius behavior (with negative activation energy): (i) the temperature can affect the physical properties of coronavirus, such as, the incubation time τ, as well as, the reproduction number r that depends on these physical properties (hao, ) . a linear approximation absorbs the (weak) temperature dependence of any parameters involved in the exponential rate; then, eq.( ) gives: where we set the intercept to be given in normal conditions of atmospheric temperature and pressure (ntp) (that is, t= c , p= atm). then, we rewrite the exponential rate as: (ii) coronavirus uses their major surface spike protein to bind on a receptor -another protein that acts like a doorway into a human cell (wrapp et al., ) . the whole process is a slow chemical reaction, where . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . / . . . doi: medrxiv preprint the mechanism behind can lead to rates negatively correlated with temperature, i.e., increasing rate with decreasing temperature. this is consistent to reaction rate expressed by the arrhenius exponential with and benson, ) . then, the effective reproduction number ( ) r t is expressed as a product combining the reproduction number in the absence of temperature effect, r  , and the arrhenius exponential rate, namely, then, eq. ( ) gives we rewrite this expression as: reactions of negative activation energy are barrier-less, relying on the capture of the molecules in a potential well. increasing {decreasing} the temperature leads to a reduced {gained} probability of the colliding molecules capturing one another. exponential spread is mainly related to outdoors activities while the decelerated growth caused by effective measures is related to indoors activities. therefore, as long as the exponential growth takes place, the environmental temperature has an effective role on the chemical reaction between virus and spike protein. due to the negative activation energy, decreasing the environmental temperature reduces the probability of virus-protein reaction, thus the virus may stay inactive on air or surfaces and eventually die. it should be noted that both the models ( b) and ( b) consider that the exponential rate λ, or the reproduction number r , are subjects to a component influenced by the culture in social activities (intercept λ ) and a component mostly influenced by the temperature (slope p ). in this way, the slope may indicate to some universal quantity involved, such as, the (negative) activation energy. next, we employ the above two expressions of exponential rate λ and temperature t, eqs.( a, a), in order to set the two types of statistical models for fitting (Τ, λ) measurements for us and italian regions. we use publicly available datasets of: ( ) average enviromental temperature of us and italian regions (e.g., see: www.ncdc.noaa.gov/data-access/land-based-station-data/land-based-datasets/climate-normals; it.climate-data.org; www.weather-atlas.com); ( ) time series of the number of daily infected cases of us and italian regions (e.g., see: www.thelancet.com; www.protezionecivile.gov.it). we analyze the datasets of regional infected cases in us and italy, derive the relationship of the exponential growth rate of the number of cases with temperature, and evaluate its statistical confidence. first, we derive the exponential growth rates of the infected cases characterizing each examined region of italy and us; then, we plot these values against the environmental temperatures of each region, and perform the corresponding statistical analysis. we proceed according to the following steps: (i) collect the time series of the current infected cases n t for all us and italian regions. for each of the us and italian regions, we plot log(n t ) and log(Δn t ) with time t, detect the time intervals of linear relationship corresponding to the phase of exponential growth, fit the datapoints within this region, and derive the slope (on linear-log scale), that is, the exponential growth rate λ. the total n t and new cases Δn t should be characterized by the same exponential . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . rate, λ, thus the slopes resulted from the linear fits of log(n t ) and log(Δn t ) with time are (weighted) averaged (fig. ) . (iii) collect environmental temperature data and calculate the temperature averaged over the whole examined region. the incubation period τ is longer than the time scale of a single day or night, thus the temperature is averaged over the daily and nightly measurements. (iv) co-plot all the derived sample values (Τ±δΤ, λ±δλ), where each pair corresponds to each examined region; then, apply a linear fitting in order to derive the linear relationship between t and λ, as well as evaluate the statistical confidence of this relationship; repeat the same for all us and italian regions. determine the critical temperature t c for which the rate becomes negligible; to eliminate the uncertainties of t c as a fitting parameter, we perform the linear fitting with the statistical model repeat (iv) and (v) with pairs of (t - ±δt - , λ±δλ); we estimate again t c by performing the linear fitting with the statistical model λ=λ (- +Τ c /t) instead of λ=p +p •t - . linear fitting of the number of the total n t and new Δn t infected cases with time (on linear-log scale) for the states of california and illinois, where the slope reads the exponential rate λ. the resulted rates from the linear fitting of log(n t ) (black) and log(Δn t ) (red) are (weighted) averaged (blue). the phases are color-coded as in figure . the hypothesis to be tested is that the exponential growth rate λ varies linearly with temperature; (x is set to be the temperature or its inverse). this is tested by examining the chi-square corresponding to the fitting of the two-parameter linear statistical model h is the α-th diagonal element of its . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . , α= , , where the derivatives are numerically derived. we will use two linear statistical models, (a) , as given by eqs.( a, a) ; both can be written with the linear expression: the statistical confidence of the dependence of the exponential growth rate on the environmental average temperature may be sufficiently high for the acceptance of any of the two statistical models. the goodness of the fitting of each model is evaluated using two types of statistical tests, the "reduced chi-square", the "p-value of the extremes", and their combination (e.g., frisch et al., ; schwadron et al., , fuselier et al., , while student's t-test is also used for evaluating the statistical confidence of the derived slopes: -reduced chi-square: the goodness of a fit is estimated by the reduced chi-square value, -student's t-test: this is another test for evaluating the statistical confidence of the slope derived from the linear fitting of the temperature-rate sample points (t i ±δt i , λ i ±δλ i ) and (t i - ±δt i - , λ i ±δλ i ). we examine, whether the slope p ±δp has significant difference from the zero slope (null hypothesis: slope is zero), by performing the student's t-test with t m =p /δp , where the corresponding p-value is derived from the integration of t-distribution . the student's t-test is not passed for the null hypothesis that they examined slope equals zero, when the corresponding p t -value is less than the acceptable confidence limit of . ; then, the null hypothesis is rejected, meaning the slope has statistically significant difference for zero. in addition, we examine the slopes estimated for us with those estimated for italian regions, by deriving the linear fitting of log(n t ) or log(Δn t ) with time t within the region of exponential growth phase, resulted to the respective rates (given by the fitted slope); their weighted averages are shown in table for us and in table for italian regions, while plotted against the average regional temperature in figs. and , respectively. the method of weighted fitting for double uncertainties (x i ±δx i , λ i ±δλ i ), as described by fasano and vio, ) , is used for estimating the fitting parameters λ , t c , together with their statistical, propagation, and total errors. the fits of the linear statistical model with temperature, x i =t i , (left panels in figs. , ), as well as of the alternative statistical model with inverse temperature, (right panels in figs. , ) , are both characterized with high statistical confidence, attaining high p-values (> . ) and reduced chi-squares red  values (close to ); also, both fits provide similar estimations of t c . the fitting results are shown in table . we also examine whether the sample points (t i ±δt i , λ i ±δλ i ) are subject to statistically significant concentrations or rarefactions, namely, whether possible heterogeneities within the distribution of sample points plays significant role in the fitted relationship. for this, we derive the temperature-rate relationship and its statistical confidence by fitting the homogenized set of sample points, instead of the raw sample points; then, we examine whether the fitting parameters differ from those derived from fitting the raw sample points. we homogenize the sample points by grouping them in temperature binning of Δt ~ c (e.g., see: livadiotis & desai ). we estimate the weighted mean and error of the rates included in each bin. in the case of us regions we also performed a homogenization of rates, by grouping the temperaturebinned means in rate binning of Δλ ~ . d - . in the case of sample points with inverse temperatures, (t i - ±δt i - , λ i ±δλ i ), the procedure is exactly the same. homogenized datasets result in a smooth relationship between the values of binned temperature and rate, as it can be observed in the plots of rate against temperature or inverse temperature (left or right lower panels, respectively), and for both us and italy regions (figs. and , respectively) . the results are highly supportive of the negative correlation between rate and temperature. the results are shown in table . we observe that the linear relationships of the growth rate with temperature or inverse temperature are characterized by high statistical confidence for the homogenized datasets (p-values much higher than significant limit of . ; red  far from the significant limits of . and ). therefore, the arrangement of sample points do not affect significantly the fitting results. in addition, as shown in tables and , the linear fits of sample points (t i ±δt i , λ i ±δλ i ) and (t i - ±δt i - , λ i ±δλ i ) do not pass the student's t-test for the null hypothesis that their slopes equals zero, i.e., the corresponding p t -values are less than the acceptable confidence limit of . ; therefore, the negative correlation of environmental temperature with the exponential rate is statistically significant (accepted with confidence %). . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . ( ) the exponential growth rate and its uncertainty is the weighted averaging of the rates derived from total and new infected cases; ( ) the environmental temperature is averaged over the time period, from τ~ days before the appearance of the st case, to st april; ( ) the standard deviation of temperature is given by the half difference between highest and lowest values within the examined time period, divided by √ (similar to the standard deviation for a sinusoidal function); ( ) ny: the temperature refers to the new york city, instead of the whole state, which suffers from the vast majority of the state infected cases. . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . / . . . doi: medrxiv preprint figure . linear fitting of rates with (left) temperatures and (right) inverse temperatures for us regions. the fitting is weighted with double uncertainties (on both the temperature and rate values). the analysis is first completed for the raw measurements (upper) and then repeated for the binned averages (lower). . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . / . . . doi: medrxiv preprint figure . as in figure , but for italian regions. table . . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . in order to improve the statistics of the estimated critical temperature, we combine the sample points (t i ±δt i , λ i ±δλ i ) of us and italian regions. first, we perform the student's t-test to compare the slopes corresponding to us and italian regions; we find high p t -values (> . ) for both fits of x=t and x=t - , thus, the two populations are likely characterized by the same slope. tthe respective intercept λ does not pass the same test, i.e., the intercepts corresponding to us and italian regions are likely different. a universality may characterize the slopes of fits x=t or x=t - , i.e., next, we perform the linear fits of the sample points (t i ±δt i , λ i ±δλ i ) and (t i - ±δt i - , λ i ±δλ i ) for the mixed set of us and italian data, once the rates of the italian regions are shifted by Δλ; the optimal fitting is obtained for that shift Δλ, for which the reduced chi-square is ~ , the p-value of the extremes is ~ . , and the combined measure ~ (see previous section). figure shows how the combined datasets of temperaturerates from us and italian regions lead to the optimal fitting. (note that the optimization is not performed for the binned datasets, since they are characterized by smaller p-values -see, figs. and ). the results are shown in table ; we observe that the optimization is reached for two values of the shift Δλ; we estimate the weighted average of the results corresponding toe the two shifts. the weighted mean is performed separately for the fitting cases of x=t and x=t - ; however, the weighted mean of the critical temperature is performed for all four results. fitting of datasets combined for us and italian regions, with the latter's rates shifted by Δλ. the optimal fitting corresponds to shifts Δλ~ . and ~ . , for which the reduced chi-square is ~ , the pvalue of the extremes is ~ . , and the combined measure ~ . . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . figure . relationship of the reproduction number r and its uncertainty with environmental temperature t. according to this, new affected cases cease (r = ) when temperature climbs to t c~ c or (~ f ). up-to-date there is no systematic statistical analysis of the effect of the environmental temperature t (and possibly other weather parameters) on the exponential growth rate of the cases infected by covid- , while a statistically confident relationship between temperature and growth rate (either with positive or negative correlation) is still unknown. the presented analysis led to a statistically confident relationship of negative correlation between the exponential growth rate and the average environmental temperature, derived for us and italian regions. in particular, we analyzed datasets of regional infected cases in us and italy, derived the exponential growth rates for each of these regions and plotted them against environmental temperatures averaged within the same regions, derived the relationship of temperature -growth rate, and evaluated its statistical confidence. . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted april , . . https://doi.org/ . the performed statistical analysis involved fitting of linear statistical models with the datasets of environmental temperature (or its inverse) and exponential growth rate, finding their relationship, and evaluating its statistical confidence. the two linear models developed and used for the statistical analysis are (a) the statistical confidence of fitting was evaluated using the reduced chi-square values, the p-value of extremes, and a testing measure that combines both of these values. also, the statistical confidence was also evaluated using the student's t-test, where the derived slopes compared to a hypothetical zero slope. the sample points of temperature (or inverse temperature) and exponential growth rate were also tested for statistically significant concentrations or rarefactions; namely, for possible heterogeneities within the distribution of sample points that could have significant role in the results. the statistical analysis of the homogenized temperature-rate data points concluded that the negative correlation between temperature and exponential rate is stable, having no statistically significant variability due to concentrations or rarefactions, and it is characterized by a high statistical confidence. we also performed a student's t-test and ensured that the difference between two sample means of us and italian regions is not statistically significant. a possible universality may characterize the slope of the temperature-rate relationship. this verifies the modeling developed and used by this analysis, where the exponential rate λ, or the reproduction number r , are subjects to a component influenced by the culture in social activities (intercept λ ) and a component influenced by the temperature (slope p ). in this way, the slope may indicate to a universal quantity involved, such as, the (negative) activation energy. since the derived slopes for us and italian regions are characterized by no statistically confident difference, we improved the statistics of the estimated fitting parameters by combined the sample points of us and italian regions. from the derived relationship, among others, we were able to derive the values of the (negative) activation energy e a , and the reproduction number r at normal conditions and how this depends on temperature. therefore, the results clearly showed that there is indeed statistically significant negative correlation of temperature on the exponential growth rate of the cases infected by covid- . figure shows the anticorrelation between the mapped exponential rates and average environmental temperature of the usa regions examined by this analysis, which they are characterized by readable exponential growth phase in their evolution spread curve. figure . anti-correlation between the spatial distributions of the exponential growth rates of the infected cases (left) and the average environmental temperature (right). . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. (which was not certified by peer review) the copyright holder for this preprint this version posted april , . . https://doi.org/ . given the negative correlation of the environmental temperature with the exponential growth rate, it was reasonable to ask for the critical temperature that eliminates the exponential rate, and thus the number of daily new cases in infected regions. this was found to be t c~ . ± . f for us regions. it is straightforward to ask when the environmental temperature will climb above this critical value. as an example, figure plots the daily average temperatures in san antonio, texas, shown that will be clearly above the estimated t c threshold in the second half of may. figure . according to the statistically confident relationship between exponential growth rate of infected cases shown in fig. , the critical temperature, which eliminates the exponential growth, and thus the covid- spread, is t c = . ± . f . the plot shows also the may-june daily, nightly, and h-averaged environmental temperatures in san antonio, texas, averaged over the last three years. the daily average temperatures will be clearly above the estimated t c threshold in the second half of may; thus, the plot suggests a possible date for loosening the strict measures in san antonio, that is, may . the resulted high statistical confidence of the negative correlation of the environmental temperature on the exponential growth rate of the cases infected by covid- is certainly encouraging for loosening super-strict social-distancing measures, at least, during the summery high temperatures. however, we are, by no-means, recommending a return-to-work date based only on this study. but we do think that this should be part of the decision, as well as an inspiration for repeating the same analysis in other heavily infected regions. the steps of these analyses may be followed as: (i) identify different outbreaks in regions with the same culture in social activities and different environmental temperature; (ii) estimate the exponential growth rates for these regions from the time series of infected cases; (iii) plot the derived rates against the environmental temperature averaged for these regions, and repeat the analysis of this study to determine the temperature-rate relationship and its statistical confidence. . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. (which was not certified by peer review) the copyright holder for this preprint this version posted april , . . https://doi.org/ . how to flatten the curve of coronavirus, a mathematician explains a mathematical model for simulating the phase-based transmissibility of a novel coronavirus how covid- and other infectious diseases spread: mathematical modeling a discrete mathematical model for the aggregation of β-amyloid an introduction to difference equations fitting a straight line with errors on both coordinates decades-long changes of the interstellar wind through our solar system low energy neutral atoms from the heliosheath negative activation energies and curved arrhenius plots. . theory of reactions over potential wells time variations in the generation time of an infectious disease: implications for sampling to appropriately quantify transmission potential effects of temperature on covid- transmission rapid expert consultation on sars-cov- survival in relation to temperature and humidity and potential for seasonality for the covid- pandemic solar radiation pressure and local interstellar medium flow parameters from ibex low energy hydrogen measurements generalized logistic growth modeling of the covid- outbreak in provinces in china and in the rest of the world no association of covid- transmission with temperature or uv radiation in chinese cities key: cord- -l zxic y authors: volpato, stefano; landi, francesco; incalzi, raffaele antonelli title: a frail health care system for an old population: lesson form the covid- outbreak in italy date: - - journal: j gerontol a biol sci med sci doi: . /gerona/glaa sha: doc_id: cord_uid: l zxic y nan italy is one of the oldest countries in europe and worldwide with an average life expectancy at birth of . years, an old age index of , and % of the population being and older ( ). on the second half of february , lombardy, one of the richest regions of the country, and northern emilia-romagna experienced the beginning of one of the largest and most serious clusters of covid- in the world. despite aggressive containment efforts, the disease continued to spread, and the number of affected patients has been rising steeply. according to the bi-weekly bulletin of the italian national institute of health (istituto superiore di sanità) as to march , , italian people ( % men) have been recognized as affected by reverse transcriptase-polymerase chain reaction (rt-pcr) testing over the country ( ) . of them, % were older than years and about % were older than years. since most of the italian regions decided to test only symptomatic patients, these figures clearly show that most cases of covid- affect the geriatric population. older people also have a worse clinical course, confirmed by case fatality rates by age group, showing a steeper increase in mortality from . % in the - age group to . % in the - group. furthermore, comparison with other european countries suggested an excess mortality in italy as compared, for example, with france ( . %) and germany ( . %) ( ) confirmed also in age-stratified analysis comparing italy with china demonstrating an excess mortality rate in italy after the age of ( ). according to the data released by the italian ministry of health, mortality rates varied across different regions of italy. for example, considering the four regions with more than deaths, mortality rate was . % for the veneto region, . % for piemonte, . % for emilia-romagna, and . % for lombardia. although age-stratified mortality rates across different regions are not available now, it is likely that regions with higher total mortality had even higher mortality rates for older people. higher mortality rates of older patients are expected as complicated covid- is characterized by severe interstitial pneumonia followed by acute respiratory distress syndrome, thromboembolic events, and eventually multiorgan failure, a cascade of negative events that is obviously more likely in older frail patients, those with elevated multimorbidity and reduced functional reserve. although it remains unclear how the different mortality rates vary across different regions of the country, the first spontaneous question is whether the italian health care system is well suited to respond to an unexpected and so huge health emergency. first, the italian health system, regarded as one of top ten in the world ( ), is equipped with . hospital beds for , persons, with extremely high bed occupancy rate, that is far below the european union average . / , and less than half of what is available in germany ( . / , ) ( ). the sudden explosion of the covid- outbreak, with almost % of infected patients having respiratory failure and requiring mechanical ventilation has immediately saturated the acute care beds availability of lombardy and northern emilia-romagna, including intensive care units' (icu) beds. the lack of availability of icu beds and mechanical ventilators has forced italian physicians to make quick and dramatic decisions on who had to be ventilated, intubated, and admitted to icu wards and who not. the guidelines of italian college of anesthesia, analgesia, resuscitation, and intensive care (siaarti) ( ) did not suggest that age should be the only factor determining resource allocation, but the document acknowledged that an age limit for icu admission may ultimately needs to be set. although chronological age was not the only criterium for selection, decision was certainly not based on a comprehensive geriatric assessment and therefore, it is likely that older patients and particularly geriatric patients, even with limited multimorbidity and good functional reserve, have had very limited access to icu and mechanical ventilation. as a matter of fact, it has been reported that several older patients have been dying at home, often alone, because of the lack of hospital beds' availability ( ) . a second important issue is that, because of the italian regionalization of the health care system, the availability of home care service and long-term care for older adults is substantially different across regions. home care and long-term care are cornerstones for continuity of care and prevention in older patients, particularly for those socially disadvantaged. from this point of view, lombardy has a very robust hospital network, but a limited long-term and home care organization that might have hampered early recognition of covid- cases among the oldest patients and might have postponed the access to care. third, according to the world health organization recommendation, laboratory testing for sars-cov- virus detection from nasopharyngeal swab or sputum sample has been performed only on symptomatic individuals or persons who had a close unprotected contact with confirmed patients. this policy may have limited the benefits of social isolation because asymptomatic people, including caregivers and health care personnel, might have carried the virus to older persons, particularly the most disabled and dependent and those living in nursing homes. from this point of view, it is striking that the veneto region that is experiencing the lowest mortality rates in italy (about %) performed the highest level of laboratory testing ( . % of the population), whereas, lombardy that tested only % of the total population is having the highest mortality rate (almost %). finally, it is possible that because of the higher burden of multimorbidity and consequent atypical presentation of covid- in older people, a diagnostic delay may have occurred in some cases leading to more severe pneumonia and worst prognosis. overall, these data suggest that italian national health care system was not suited and prepared to respond to this sudden and dramatic outbreak. first, our system, like our older frail patients, is lacking functional reserve able to increment the overall efficiency and maintain the system homeostasis under stressful conditions. the immediate shortages of icu beds and ventilators are one of the major causes of the incredibly high mortality rates observed in italy. second, our health system in probably well oriented to the cure and prevention of multifactorial conditions very common in western countries like cancer, cardiovascular diseases, and metabolic conditions but it is no longer used to counteract infection diseases outbreaks. from this point of view, the italian tragedy suggests that a prompt population mass testing for detecting asymptomatic infected people along with immediate, widespread, draconian measures of social isolation, along with contract tracing and quarantine, might have more strongly reduced the rate of covid- transmission, with the greatest benefit for frail patients, more prone to a complicated course of the disease. third, retirement homes for the older adults and nursing home should develop and implement specific strategies and protocols to prevent external contamination and widespread of the infection between their residents. fourth, social isolation might be particularly negative for older people and might be associated with worsening of many chronic conditions. it is of paramount importance to guarantee indirect contact with older patients and their caregivers, like a teleconsultation service for older persons. geriatric care teams must be integrated along with infectious disease specialists, pneumologists, and anesthesiologists into the management of this sars-cov- infection crisis and must develop tailored management strategies for older patients and should promptly investigate and describe the peculiar clinical presentation of covid- in older patients in order to facilitate immediate recognitions of the disease and start available therapy to prevent the most severe form of pneumonia complicated by respiratory failure ( ) . in this perspective, the gerocovid initiative of the italian society of gerontology and geriatrics aims at expanding the knowledge of the disease and its impact on health status and needs of care as well as at promoting health care strategies tailored to multiple different settings and different available resources. any geriatric society or working group can voluntarily join the gerocovid group at www.sigg.it. this would allow explore the effects of covid pandemic on the older population through the same protocol and, thus, have comparable data. each geriatric society or geriatric working group will use its own data besides contributing to the whole study. we hope that such a working method will make geriatricians able to recognize the disease, help their patients, and provide valuable information to the health authorities. case-fatality rate and characteristics of patients dying in relation to covid- in italy italy's health performance, - : findings from the global burden of disease study clinical ethics recommendations for the allocation of intensive care treatments, in exceptional, resource-limited circumstances troppi malati lasciati a casa. non c'è più la sanità pubblica the new challenge of geriatrics: saving frail older people from the sars-cov- pandemic infection key: cord- -zie uv authors: godio, alberto; pace, francesca; vergnano, andrea title: seir modeling of the italian epidemic of sars-cov- using computational swarm intelligence date: - - journal: int j environ res public health doi: . /ijerph sha: doc_id: cord_uid: zie uv we applied a generalized seir epidemiological model to the recent sars-cov- outbreak in the world, with a focus on italy and its lombardy, piedmont, and veneto regions. we focused on the application of a stochastic approach in fitting the model parameters using a particle swarm optimization (pso) solver, to improve the reliability of predictions in the medium term ( days). we analyzed the official data and the predicted evolution of the epidemic in the italian regions, and we compared the results with the data and predictions of spain and south korea. we linked the model equations to the changes in people’s mobility, with reference to google’s covid- community mobility reports. we discussed the effectiveness of policies taken by different regions and countries and how they have an impact on past and future infection scenarios. we present an updated version of the predictive model of epidemic phenomena based on the approach called seir (susceptible-exposed-infective-recovered), widely used to analyze infection data during the different stages of an epidemic outbreak. the seir model represents one of the most adopted mathematical models to characterize the epidemic dynamics and to predict possible contagion scenarios. the seir model can be useful to assess the effectiveness of various measures, such as lock-down, since the infectious disease outbreak. it is based on a series of dynamic ordinary differential equations that consider the amount of the population subject to contagion, the trend over time of individuals who recover after infection, and the individuals who unfortunately die [ ] . this work was carried out during the crucial development phase of the epidemic in italy (mid-april ), with the operational difficulties linked to the impossibility of verifying and validating the databases, and with the difficulty of comparing and calibrating the results with other studies. the purpose, however, is to provide an easy-to-read and useful tool that can help the policymakers, responsible for strategic choices, in assessing the social and economic scenarios related to the development of the epidemic. we are conscious that it is a predictive model which, although based on a scientific approach, is conditioned by a series of intrinsic and endogenous factors that can affect its medium-term reliability. nevertheless, we are also aware that any political decision potentially lacking any rational and critical evaluations of all the available data risks being based on mere sensations, often dictated by sentimental suggestions [ ] . the generalized seir model is based on a system of differential equations, as discussed by peng et al. ( ) [ ] in the analysis of the sars-cov- outbreak in china. the model, that adds complexity to the classical sir or seir models, represents the various conditions of susceptible the seir model simulates the time-histories of an epidemic phenomenon. in its classical form, it models the mutual and dynamic interaction of people between four different conditions, the susceptible (s), exposed (e), infective (i), and recovered (r). the classical seir model can be described by a series of ordinary differential equations: dr(t) dt = (λ + κ)i(t) the susceptible (s) is the part of the population that could be potentially subjected to the infection: at the initial time, without further information, it is represented by the whole population. the exposed (e) is the fraction of the population that has been infected but does not show symptoms yet: it can be called a latent phase, and at this stage, a disease can be infectious, partially infectious or not infectious [ ] . the infective (i) represents the infective population after the latent period. the recovered (r) are the people after healing, and they are generally not reintroduced into the susceptible category if it is supposed that they became immune to the disease. this aspect is strongly debated, as in some countries a second infection of recovered people has been recorded. at this stage of our study, we do not have enough data to include this effect on the model; this would require the introduction of another term in the previous system of equations, including another coefficient that takes into account the re-population of the susceptible compartment. in the classical seir model, the r category also comprehends the individuals who died of the disease. a characteristic of this model is that the sum of the four categories is equal to the total population (n) at any time: as can be seen, it does not consider the natural births and deaths of the population during the time span of the disease. the equations of the classical seir model are governed by the parameters β, γ, λ and κ. we adopt the symbols used in peng et al. ( ) [ ] . as usual in this field, the following parameters have day − as a unit of measurement. • β is called infection rate. it is the number of people that an infective person infects each day. it is equal to p, where b, or the contact rate, is the number of people an average person enters into contact with each day, and p is the probability that a contact provokes the transmission of the disease. in the seir model, β is the vector which transports people from the s category to the e category. it is multiplied by the ratio s/n to avoid counting contacts between two people who cannot infect each other (e.g., because one of them has already recovered, or because both are infective). • γ is the inverse of the average latent time and governs the lag between having undergone an infectious contact and showing symptoms: in the equations, it brings people from the e category to the i category. • λ and κ are the recovery rate and the death rate, respectively, and they are united together in a single parameter in the classical seir model. they give information about how fast the people may recover from the disease ( /λ is the average recovery time), and how many of them, unfortunately, die. given the complexity of the disease, many authors have implemented different variations of the classical seir model, regarding both the equations and the parameters, or managing different fitting techniques to make the model representing the reality as close as possible. we adopted a generalized seir model following the recent publication by peng et al. ( ) [ ] , who studied the covid- infection in several chinese provinces. we applied the model of peng et al. ( ) to the italian situation, following this system of equations: this seir model adds some features to those of the classical seir model (equations ( )-( )): it supposes that the susceptible population decreases thanks to lockdown policies and improvements in public health behaviors, such as wearing face masks. each day, several individuals (s·α) passes from the susceptible category to the protected category (p), being α the protection rate. it adds the category of quarantined people (q). the passage from the infective to the quarantined category is done through the parameter δ, which is the inverse of the average time required to quarantine a person with symptoms: this happens usually after the person has been tested positive. the quarantined people are hence excluded from the infective category (i) because they are supposed not to have had any contact with others. the quarantined category matches the "active confirmed cases" in italian official datasheets, and, according to the common habit of quarantining positive people, it is true also for data from most developed countries. this is a critical point for the system of equations, that according to us should be better defined. in fact, an infected individual is not automatically quarantined, because the authorities were often unable to test enough people while keeping pace with the spread. this is especially difficult because many people do not develop symptoms at all, but can transmit the infection to others. so, we think that δ also contains some information about the percentage of the detected infective people. a study from calafiore et al. ( ) [ ] proposed the introduction of an additional parameter to better understand this issue. it separates the categories of recovered (r) and dead (d) people, linked to the quarantined category through the λ and κ parameters, the cure rate, and the mortality rate respectively. λ and κ are time-dependent because the health system can improve its capability to treat people over time, e.g., with the introduction of a new therapy. based on the data collected from chinese reports in peng et al. ( ) [ ] , which suggested an exponential evolution of the two parameters, we constrained λ and κ to fit an exponential trend. similarly to cheynet ( ) [ ] , the assumption is that the death rate should become closer to zero as time increases, while the recovery rate converges toward a constant value: the λ represents the final asymptotic value of the cure rate. it is related to the health system's ability to tackle the infection after adapting to the new outbreak and depends also on other factors like the good health of the citizens. λ is related to how fast the adaptation to the emergency was. at the beginning of april, south korea was already in a post-peak phase of the disease spread. from our initial tests, equation ( ) did not ideally match the data of south korea, probably because of the more complicated trend, compared to other countries. therefore, only for the south korean model, the λ parameter was not constrained by an exponential law. we increased the degrees of freedom of its trend by imposing a sinusoidal law: the fitting of the six parameters was performed by rearranging equation ( ): where r and q are the data series of recovered and quarantined people. this approach was successful in providing a better fitting of the model prediction to the data. the parameter κ represents the initial value of the mortality rate. the κ value is related to the initial health system's ability to tackle the infection and depends on the good health of the citizens. the κ value measures how the rate has changed with time. the mortality rate is supposed to decrease over time, and the higher κ is, the faster this decrease. we introduced an improvement to the β parameter, compared to peng's model, that is, its time dependency. since the infection rate β is proportional to the contact rate b, as stated before, we estimated the variation in the contact rate according to the recent publication of google's covid- community mobility reports [ ] , a database built on gps data collected from mobile devices with the "location history" option turned on. it provides data about the reduction in the mobility of people over the recent few months. for each investigated region, we calculated the average mobility decrease over time, and we fitted the curve with a second-order polynomial trend line. then, we constrained β to be proportional to that specific trend line. in a preliminary test on a simple data set, we noticed how the introduction of this constraint allowed the model to obtain a better data fitting. therefore, we applied this approach to the seir model to ensure good data fitting for all the regions and countries investigated. we noticed that the mobility data of south korea did not show a significant decrease in people's mobility, because the government adopted a different strong approach in lockdown policies compared to most european countries. strict lockdowns were not imposed, but efforts were addressed to track the infection spread at the early stages, with tight controls and strict quarantine protocols for infected individuals. for the days of model prediction, the time-dependent β parameter had the same value of the last observed day (i.e., mid-april) since we did not have reliable predictions about future mobility. this means that the lockdown policies will continue, or the reopening of business will be made paying close attention to health protection procedures. the generalized seir-model scheme is described in figure . the expected evolution of the equation terms is: the susceptible category decreases over time, feeding the exposed (e) category through the beta parameter β and the protected (p) category through the α parameter; the latter represents the part of the population who for various reasons becomes insusceptible to the disease. the exposed (e) category is only a temporary category: its individuals pass into the infective (i) category after a latent time ( /γ), on average. the infective category generates newly infected people over time, removing them from the susceptible category. the detected infected individuals are quarantined (q) to avoid spread. then, they evolve into recovered (r) or death (d) cases, according to various causes, like health care system effectiveness, age, co-morbidity of other diseases. it is important to note that the most reliable data series provided by national agencies are q(t), r(t), and d(t). the fitting of these data in the structure of the generalized seir model allows the trend of other categories to be estimated with some degrees of uncertainty, as well as their prediction for the subsequent days. the main outputs of the model are the following data series: • s, target time-histories of the susceptible cases, • e, the target time-histories of the exposed cases, • i the target time-histories of the infective cases, • q, the target time-histories of the quarantined cases, • r, the target time-histories of the recovered cases, • d, the target time-histories of the death cases, • p, the target time-histories of the insusceptible cases. the α, β, γ, δ, λ and κ parameters can be considered a major output of the model. in particular, the evolution over time of λ and κ could provide information about the changes in the health system response to the contagion. β is also time-dependent, and it is constrained to be proportional to the people's mobility trend extrapolated from google's big data. the α-value is also related to policies, although it is not closely related to a precise aspect of the government strategies, as the β parameter. the α-value was not forced to follow any particular law during the modeling since we tested no significant improvement by modifying it from a constant to a time-dependent parameter. the α, β, γ, δ, λ and κ parameters can be considered a major output of the model. in particular, the evolution over time of λ and κ could provide information about the changes in the health system response to the contagion.  is also time-dependent, and it is constrained to be proportional to the people's mobility trend extrapolated from google's big data. the -value is also related to policies, although it is not closely related to a precise aspect of the government strategies, as the  parameter. the -value was not forced to follow any particular law during the modeling since we tested no significant improvement by modifying it from a constant to a time-dependent parameter. the various researches about seir-like models applied to the current sars-cov- epidemic have introduced minor or major changes to the classical seir model. it can be useful to overview the values of the main parameters, to define a realistic range of values, and to understand their meaning. we used the coefficient values found in literature to set the lower and upper boundaries of the parameters in our modeling. the values of these parameters are related to different methods. therefore, the bibliographic research summarized in table should be only considered as a qualitative benchmark. we highlight some peculiarities of the various studies. the various researches about seir-like models applied to the current sars-cov- epidemic have introduced minor or major changes to the classical seir model. it can be useful to overview the values of the main parameters, to define a realistic range of values, and to understand their meaning. we used the coefficient values found in literature to set the lower and upper boundaries of the parameters in our modeling. the values of these parameters are related to different methods. therefore, the bibliographic research summarized in table should be only considered as a qualitative benchmark. we highlight some peculiarities of the various studies. table . values of α, β, γ, δ, λ, and κ describing sars-cov- outbreak in the recent literature. the study by calafiore et al. [ ] presents the values of the parameters also for each italian region. the peculiar seir model used here introduces two new parameters: α (with a meaning different from ours) and ω. α represents how many times the real number of infected people is higher than the number of detected infected people, and it is estimated to be about . the ω parameter defines a fixed percentage of total people susceptible to the disease, and according to this model, it is about . ( %). one of the who reports [ ] shows a list of estimates of serial interval, which is the average time between infection and subsequent transmission. we reported the latter since it is closely related to the inverse of β. it ranges from to days, according to the various studies considered. dandekar et al. [ ] calculates the change over time of a term, called quarantine strength (q(t)), fitted thanks to a neural network-based approach. shaikh et al. [ ] considers as separated the asymptomatic and symptomatic categories. the most interesting feature of the model proposed by lin et al. [ ] is the time-dependent β, implemented with a different design compared to our model: where α is a stepwise function that represents the governmental action, estimated to range from to . according to the strength of lockdown policies. while k is the citizen response, estimated to be about . iwata et al. [ ] proposes a model which does not fit real data but investigates possible scenarios deriving from a different combination of parameters. their range is reported as a credible range reference. the model equations were implemented on the basis of the matlab code provided by [ ] , available in matlab file exchange. the data are extracted from the official repository and are composed of: confirmed, recovered, and death cases (q, r, and d, respectively). these values represent the initial assumptions, while the parameters α, β, γ, δ, λ, and κ are the problem unknowns. the differential equations are numerically solved by means of the runge-kutta method. the standard approach of the source code uses as default a least-square fitting solver to match the observed data and the calculated response (of q, r, and d). at the beginning of modeling, the initial values of the six parameters are given as first esteem. then, their values are calculated following a least-square solver that considers the observed data (q, r, d) with time. we modified the standard release of the code by introducing a new solver, the pso algorithm, belonging to the family of computational swarm intelligence (population-based nature-inspired metaheuristics) [ , ] . this optimization solver minimizes an objective function, which is set to decrease the misfit between observed data and calculated responses of q, r, and d by varying the six parameters, i.e., the problem unknowns. the main advantages of the stochastic approach over the deterministic method to solve the seir model are briefly discussed. the adaptive exploration and exploitation of the search space of the model solutions avoids the risk of being trapped in some local-minima solutions and also enhances the independence from the initial assumption of the six parameters which could bias the final solution. the solution search-space is sampled by a set of particles, representing the possible solutions, which are randomly initialized. the adaptive behavior and the convergence and stability of the final solution are ensured by using a pso variant, the hierarchical pso with time-varying acceleration coefficients (hpso-tvac) [ ] . convergence was achieved in iterations. each run of iterations was repeated for trials to test the variability of the solutions due to the random initialization of the parameters. finally, the trial showing the minimum normalized root mean square error (nrmse) was selected as the best solution. the solutions from the remaining trials were a-posteriori evaluated with their probability density distribution. the solutions within % of the minimum nrmse were chosen as representative of other probable scenarios. deploying a stochastic approach increased the computational cost of the modeling. therefore, the code was parallelized to be run on multiple cores. the simulations ran on the academic high performance computing (hpc) cluster of politecnico di torino. the sustained performance of the cluster is globally . tflops and the cpu model of one node is x intel xeon e - v . ghz cores. we adopted cores of a single node. here we present the time series obtained by the standard deterministic approach and the data series obtained by the stochastic approach, based on the particle swarm optimization (pso) algorithm. first, we analyze the italian framework at a national and regional scale. then, we provide the results of seir modeling for two other countries: spain and south korea. spain was chosen because the epidemic spread is similar to the italian one, while south korea represented a testing data set as the epidemic peak had already been overcome. the final nmrse of the modeling and the values of the seir coefficients are supplied in table . the prediction of the italian situation according to the deterministic solver shows the trends given in figure . the result of the pso approach is shown in figure . observed data of quarantined, recovered, and death cases are marked in red, green, and black circles, respectively. the individuals tested positive and placed in quarantine (at home, or hospitalized, or in intensive care) are plotted in red color. the sum of quarantined, recovered, and deaths, at a certain date, represents the total confirmed cases at that moment. the italian data set starts from st march, because we start to model from the day when the confirmed cases were % of the maximum counted cases. the predicted curves are plotted with solid lines in figures and . the set of most probable pso solutions (within % of the minimum nmrse) is plotted with dashed lines in figure . the predicted peak in the red curve represents the status in which the rate of recoveries becomes greater than the rate of infection. it reflects the most relevant impact on the health system, because the numbers of quarantined people, both at home and in the hospitals, are at their maximum. for the italian situation, the maximum number of the predicted quarantined cases is expected after th april according to the deterministic approach and some days before the day according to the best solution of the pso approach. the curves of recovered and deceased cases in figures and are similar. the final nrmse was . and . for pso and deterministic modeling, respectively (table ). confirmed cases were % of the maximum counted cases. the predicted curves are plotted with solid lines in figures and . the set of most probable pso solutions (within % of the minimum nmrse) is plotted with dashed lines in figure . the predicted peak in the red curve represents the status in which the rate of recoveries becomes greater than the rate of infection. it reflects the most relevant impact on the health system, because the numbers of quarantined people, both at home and in the hospitals, are at their maximum. for the italian situation, the maximum number of the predicted quarantined cases is expected after th april according to the deterministic approach and some days before the day according to the best solution of the pso approach. the curves of recovered and deceased cases in figures and are similar. the final nrmse was . and . for pso and deterministic modeling, respectively (table ). the italian data set starts from st march, because we start to model from the day when the confirmed cases were % of the maximum counted cases. the predicted curves are plotted with solid lines in figures and . the set of most probable pso solutions (within % of the minimum nmrse) is plotted with dashed lines in figure . the predicted peak in the red curve represents the status in which the rate of recoveries becomes greater than the rate of infection. it reflects the most relevant impact on the health system, because the numbers of quarantined people, both at home and in the hospitals, are at their maximum. for the italian situation, the maximum number of the predicted quarantined cases is expected after th april according to the deterministic approach and some days before the day according to the best solution of the pso approach. the curves of recovered and deceased cases in figures and are similar. the final nrmse was . and . for pso and deterministic modeling, respectively ( table ) . the analysis of the situation of lombardy, veneto, and piedmont regions is depicted in figures - , respectively. lombardy was strongly impacted by sars-cov- , as at the end of march, nearly , novel infected cases and more than deaths were recorded in a population of million. on the contrary, the veneto region evidenced cases and about deaths in a population of million people. the analysis of the situation of lombardy, veneto, and piedmont regions is depicted in figures - , respectively. lombardy was strongly impacted by sars-cov- , as at the end of march, nearly novel infected cases and more than deaths were recorded in a population of million. on the contrary, the veneto region evidenced cases and about deaths in a population of million people. the prediction of the situation in lombardy, according to the deterministic approach ( figure a) , appears rather optimistic, as the trend of the quarantined should start to decrease in a few days (red solid line). this probably does not reflect the evolution of the true situation in that region, even if the rate of the recovered generates positive feelings. if we look at the most probable scenarios predicted according to the pso analysis (dashed lines in figure b) , the wide spreading of the trend of the quarantined indicates how any eventual less-restrictive policy must be evaluated with great care in the next days. the set of most probable solutions from pso presents a wide range of solutions, wider than that for italy (figure ). the final nrmse was . and . for pso and deterministic modeling, respectively (table ) . the prediction of the situation in lombardy, according to the deterministic approach (figure a) , appears rather optimistic, as the trend of the quarantined should start to decrease in a few days (red solid line). this probably does not reflect the evolution of the true situation in that region, even if the rate of the recovered generates positive feelings. if we look at the most probable scenarios predicted according to the pso analysis (dashed lines in figure b) , the wide spreading of the trend of the quarantined indicates how any eventual less-restrictive policy must be evaluated with great care in the next days. the set of most probable solutions from pso presents a wide range of solutions, wider than that for italy ( figure ). the final nrmse was . and . for pso and deterministic modeling, respectively (table ) . approach. in (b) the solid line refers to the best pso solution, the dashed lines refer to the most probable solutions (i.e., the solutions within % of the minimum nrmse). figure a ,b shows the seir model prediction for the veneto region, according to the deterministic and pso approaches, respectively. while the predicted recovered and death cases are in accordance, the curves of quarantined cases present a slightly different estimate of the predicted peak, which is comprised between th and th april. the final nrmse was . and . for pso and deterministic modeling, respectively (table ) . the seir modeling for the piedmont region is shown in figure a ,b, where the solution using the deterministic and pso prediction are reported, respectively. the scenarios predicted from pso are a little worse than those of the deterministic solutions. however, the observed data of piedmont yield a wide range of probable solutions (dashed lines), which can be overlapped to the deterministic solution in some cases. the final nrmse was . and . for pso and deterministic modeling, respectively (table ). figure a ,b shows the seir model prediction for the veneto region, according to the deterministic and pso approaches, respectively. while the predicted recovered and death cases are in accordance, the curves of quarantined cases present a slightly different estimate of the predicted peak, which is comprised between th and th april. the final nrmse was . and . for pso and deterministic modeling, respectively ( table ) . the seir modeling for the piedmont region is shown in figure a ,b, where the solution using the deterministic and pso prediction are reported, respectively. the scenarios predicted from pso are a little worse than those of the deterministic solutions. however, the observed data of piedmont yield a wide range of probable solutions (dashed lines), which can be overlapped to the deterministic solution in some cases. the final nrmse was . and . for pso and deterministic modeling, respectively (table ) . the epidemic situation in spain is shown in figure . the crisis exploded in a few days after the italian collapse, as the direct consequence of the delay in undertaking restrictions in business and social activities to limit the spreading of the infection. at this stage of the evolution of the phenomenon in spain, after one month, the results obtained by the deterministic approach forecast a trend of the recovered that seems very optimistic if compared with the italian situation. we can assume that the spanish health system will react promptly to the last round of the infectious. the result of pso modeling is shown in figure b . the final nrmse was . and . for pso and deterministic modeling, respectively (table ). the epidemic situation in spain is shown in figure . the crisis exploded in a few days after the italian collapse, as the direct consequence of the delay in undertaking restrictions in business and social activities to limit the spreading of the infection. at this stage of the evolution of the phenomenon in spain, after one month, the results obtained by the deterministic approach forecast a trend of the recovered that seems very optimistic if compared with the italian situation. we can assume that the spanish health system will react promptly to the last round of the infectious. the result of pso modeling is shown in figure b . the final nrmse was . and . for pso and deterministic modeling, respectively (table ). the trend of the cases and the predicted response of south korea's situation is presented in figure . the analysis of the data about south korea is useful to look at the italian situation with respect to a country where, for many reasons, the infection was limited, even if the crisis seemed very dramatic at the early stage. the abrupt changes of the recovered and quarantined trend required a careful analysis of the seir coefficients and their temporal variation. the final data fitting was indeed not ideal because of the marked oscillations in both the time-series of quarantined and recovered cases. the final nrmse was . and . for pso and deterministic modeling, respectively (table ). the trend of the cases and the predicted response of south korea's situation is presented in figure . the analysis of the data about south korea is useful to look at the italian situation with respect to a country where, for many reasons, the infection was limited, even if the crisis seemed very dramatic at the early stage. the abrupt changes of the recovered and quarantined trend required a careful analysis of the seir coefficients and their temporal variation. the final data fitting was indeed not ideal because of the marked oscillations in both the time-series of quarantined and recovered cases. the final nrmse was . and . for pso and deterministic modeling, respectively (table ) . we adopted a generalized seir model to offer a quantitative overview of the complex analysis of the sars-cov- epidemic, meanwhile the disease is still running. the parameters were fitted in a least-square sense with a deterministic approach, and then with a stochastic approach, using a particle swarm optimization (pso) algorithm, a novelty in the field of epidemiological studies. the analysis of the results from the stochastic approach gives an overview of the most probable scenarios selected among the solutions within % of the normalized root mean square (nrmse) of the best solution. for each investigated area, we performed trials of pso simulations and from to trials belonged to the most probable set. it is noticeable that the predicted model responses led to an approximately equivalent the data fitting (normalized with respect to the mean value within an l -norm < . ). the probable scenarios sometimes presented a wide range of possible solutions because of the intrinsic setting of the stochastic approach. the different scenarios were achieved thanks to a deeper investigation of the model-space domain where the solutions are not driven and influenced by the initial guess of the seir model coefficients. one of the main limits of the we adopted a generalized seir model to offer a quantitative overview of the complex analysis of the sars-cov- epidemic, meanwhile the disease is still running. the parameters were fitted in a least-square sense with a deterministic approach, and then with a stochastic approach, using a particle swarm optimization (pso) algorithm, a novelty in the field of epidemiological studies. the analysis of the results from the stochastic approach gives an overview of the most probable scenarios selected among the solutions within % of the normalized root mean square (nrmse) of the best solution. for each investigated area, we performed trials of pso simulations and from to trials belonged to the most probable set. it is noticeable that the predicted model responses led to an approximately equivalent the data fitting (normalized with respect to the mean value within an l -norm < . ). the probable scenarios sometimes presented a wide range of possible solutions because of the intrinsic setting of the stochastic approach. the different scenarios were achieved thanks to a deeper investigation of the model-space domain where the solutions are not driven and influenced by the initial guess of the seir model coefficients. one of the main limits of the deterministic approach, instead, is that the results are biased by the selection of the starting point of model parameters. the data seem to confirm that while lombardy and piedmont applied similar approaches to social distancing and retail closures, veneto's strategy applied a much more proactive effort to limit the contagion, by means of extensive testing of symptomatic and asymptomatic cases early on, jointly with an effective tracing of potential positives. the different actions undertaken by the regions are well depicted in the future trend of the model, with evident advantages in an earlier end of the infection spreading in veneto (compare figures and with figure ). in fact, the peak of quarantined in veneto lies before those of lombardy and piedmont. the descending curve in veneto has a sharper trend than that of the other two regions. moreover, in veneto, the predicted fatalities are ten times lower and the recovered are five times lower than those in lombardy. the behavior of piedmont ( figure ) deals with a peculiar trend, introducing a time-delay of the recovered (green curve) with respect to the death cases (black curve), since the number of the recovered in the month of march is always lower than the deaths. this is because the intersection of the trends of recovered and death cases is reached later than the other regions herein analyzed, i.e., th april. this probably resulted from the regional testing policy that tested (and counted as confirmed) only patients with severe symptoms or at high risk. the high rate of fatalities that occurred in march was also due to the unexpected stress on the health system and the scarcity of intensive-care units. differently, lombardy and veneto experienced a higher rate of recovered patients at the early stage of the epidemic outbreak. a recent analysis [ ] has pointed out how, according to the guidance from public health authorities in the central government, lombardy's actions involved a more conservative approach mainly focusing on the symptomatic cases. they also suppose that the set of policies enacted in veneto minimized the burden on hospitals and minimized the risk of spreading in medical facilities. veneto's strategy tried to prevent the diffusion by capillary actions at the local scale, to limit the contagious with additional measures in the hot spots of the infection at the early stage of the epidemic. the expected trend of these regions was controlled by many factors outside the control of policymakers, including lombardy's greater population density and a higher number of cases at the explosion of the crisis. nevertheless, the different public health policies at the early stage of the epidemic phenomena also had an impact, and it seems that tailored capillary actions, as in the example of veneto, obtained better results than applying only a regional lockdown. the difference in the approaches can be underlined by observing that many municipalities or provinces declared "red zones", where, due to high transmission of the infection, additional restrictive measures were introduced, compared to the rest of the regional territory. in the red zones, the different policies acted in response to local epidemiological situations. instead, in piedmont and lombardy, no red zone was established, but restrictive individual distancing measures were regulated on a regional scale. according to the evidenced results of different policies, in the next phase of governmental policies, the reopening of business and activities should be tailored to the local situations, focusing on the organization and integration of all figures of the health system. in particular, the central government should require from the regions an effort to provide local epidemiological data in real-time, to lockdown only limited areas, while the reopening of regional-scale business can be eased. the estimated parameters that regulate the equations of the seir model are reported in table . for the parameters obtained with stochastic approach, the best-solution is shown in bold, while in brackets the mean and the variance of the solutions within % of the minimum nrmse in brackets. in table we compare the parameters among different regions, and between the stochastic and deterministic approach. both approaches provided models that fitted the observed data with good accuracy, although the stochastic approach has, in general, a slightly lower nrmse. the parameters calculated with pso are reported with the best-solution value, mean, and variance. we can observe that α, β, γ and δ had a high variance due to the intrinsic variability due to the stochastic approach. sometimes the best solution is not aligned with the mean value. λ and κ values, instead, are strictly gathered around the mean in almost all pso solutions, hence the low variance. this is explained by considering that, since the number of parameters is higher than the available data series (q, r, and d), the problem is underdetermined, so that the stochastic approach can find more than one series of parameters which fits the data within an acceptable misfit. then, λ and κ do not show large variability among the most-probable scenarios because they govern the equations that correlate q with r and d, that is the official data series. therefore, the estimated λ and κ were always found in the same region of the search space of solutions. south korea and veneto show the highest recovery rates (λ ) with values around . , followed by spain ( . ). this confirms the reports which praise the veneto model, because its administration had the capabilities of testing more quickly than other italian regions, and the family doctors worked in a stronger synergy with the health structures. it also evidences a lower death rate (κ ), probably due to the better health system efficiency to treat patients, but also to the greater number of tests. in both data and policies, the veneto region is more like south korea than other parts of italy. these aspects had an impact on the outbreak of the epidemic, as can be seen comparing figures - : the veneto region is more likely to reach the peak of active cases (q) before the other regions. even though the pso results may seem to provide a wide range for the seir parameters, we stress two important aspects: • as we already stated, the problem is underdetermined, so it is preferable to have an acceptable range of values than a unique point value, that could result in being uncertain, as could happen considering only a deterministic approach solution; • the set of possible predicted scenarios, although related to different solutions with different sets of parameters, are quite similar, thus offering an acceptable level of variability of future predictions. while it would be very useful to estimate a more narrow range of parameters like the infection rate β or the latent time ( /γ), this is beyond the goals of our study, and the topic is being explored by researchers who focus also on the clinical aspects of the disease. the model has some limitations, as previously discussed. we summarize them to highlight possible needs in the further development of the modeling. we have currently not sufficient information to say that, after recovery, an individual becomes totally immune to the disease, but we made this assumption: the model did not allow the passage from the recovered category to the susceptible category. the model does not consider that the exposed category may have a partial infection ability, as described in shi ( ) [ ] , nor distinguishes symptomatic from asymptomatic people, as studied in shaikh ( ) [ ] . the model does not consider the testing differences between different health system structures and country policies. • while italian and spanish data are well fitted, the south korean data fitting presents some issues. this evidences that different policies between countries can induce different trends in the spread of the epidemic and that the models should be adapted to different situations, with the introduction or removal of parameters. this would be especially valid in analyzing the situation of the least developed countries, that are not able to afford strict lockdown policies like the developed countries. except for the death rate parameter, the model does not have a strong link to the health resiliency of citizens. the death rate parameter could also be related to external factors like air pollution, which makes people more sensitive to respiratory diseases [ ] . the introduction of google's covid- community mobility report represents a constraint that was easily implemented in the model. further studies on the quality of those data and a rigorous implementation could represent a novel and interesting research topic. we think that many of these issues could still remain open, but the critical point of the study is not to determine exactly in which way each external factor influences the trend of infectious cases, because we are analyzing a multifaceted problem from a global point of view. moreover, the official data we consider are suspected to be not enough accurate to be the basis of a very detailed study. we applied two different approaches for solving the equations of the seir model to describe the evolution of the epidemic phenomenon in italy and in the most impacted regions of the north of italy (lombardy, veneto, and piedmont). we considered all the possible available data on the april . the main findings indicate that the deterministic approach is not appropriate to explore the possible solutions of the space-domain because the mathematical problem is underdetermined. we recommend fitting the data of this epidemic using a stochastic approach, such as the pso method. taking advantage of the pso approach, we estimated different scenarios for a -day epidemic evolution. every scenario refers to a different set of parameters estimated by the algorithm. the predicted scenarios are fairly similar and suggest that every italian region will reach the peak of the epidemic by mid-may. the influence of the time-varying infection rate βon the model prediction may open interesting discussions about the effect of lockdown policies on the evolution of the epidemic in the near and far future. because the model was provided rapidly and the study was performed during the international emergency, we did not explore further the implications of different "reopening" scenarios. we can say that, if the β parameter remains at current values, e.g., if the lockdown policies are maintained or, better, the reopening of business is done with particular attention to health safety procedures, the prediction of the trend of the recovered and deaths could be considered reliable, with the approximations and the uncertainties that the pso model has pointed out. at the italian level, despite the great dispersion in the prediction of the quarantined and recovered cases, the number of deaths will reach a number of around , - , cases at the end of may and the number of active cases will gradually decrease. this prediction cannot consider the impact of future decisions on social distancing. the data and the model predictions confirm that some valuable lessons should be learned from the approaches of south korea, which was able to contain the contagion very soon before a wide spread of the infection. the veneto region was one of the best examples in italy about how integrated and synergic regional policies in social distance and the health system can tackle the epidemic, and its epidemiological scenario is now more optimistic than those of lombardy and piedmont. we stress that tailored actions provide much better epidemiological outcomes than wide lockdowns, keeping in mind also the example of south korea. the main purpose of our work was to provide a fresh discussion and new tools able to support the policymakers in their decision about the action to minimize the impact of the disease. the analysis demonstrates that, because the italian health care system is highly decentralized, different regions managed different policies, which highly influenced the evolution of the epidemic in its first months: the data and the model prediction well reflected the different approaches taken by lombardy and veneto, two regions with similar socio-economic tissue. the overall lesson that could be learned from this analysis goes beyond the mathematical modeling itself, and will require a wider evaluation on all the possible socio-economic and political factors, even if the data analysis of the veneto situation could be used to revisit regional and central policies early on. if so, the regions are going to emulate the virtuous approach of veneto, including more demanding requests to improve their diagnostic capacity that will weight on the central government. outbreak properties of epidemic models: the roles of temporal forcing and stochasticity on pathogen invasion dynamics early transmission dynamics in wuhan, china, of novel coronavirus-infected pneumonia epidemic analysis of covid- in china by dynamical modeling a short history of mathematical population dynamics a modified sir model for the covid- contagion in italy covid- community mobility report particle swarm optimization seir transmission dynamics model of ncov coronavirus with considering the weak infectious ability and changes in latency duration generalized seir epidemic model (fitting and computation) who novel coronavirus: overview of the state of the art and outline of key knowledge gaps/slides. available online neural network aided quarantine control model estimation of global covid- spread a mathematical model of covid- using fractional derivative: outbreak in india with dynamics of transmission and control a conceptual model for the coronavirus disease (covid- ) outbreak in wuhan, china with individual reaction and governmental action a simulation on potential secondary spread of novel coronavirus in an exported country using a stochastic epidemic seir model self-organizing hierarchical particle swarm optimizer with time-varying acceleration coefficients particle swarm optimization of d magnetotelluric data lessons from italy's response to coronavirus exposure to air pollution and covid- mortality in the united states: a nationwide cross-sectional study author contributions: all authors provided a contribution to the development of the methodology and conceptualization. code implementation and modeling: f.p. and a.g; validation and interpretation, a.g. and a.v.; writing-original draft preparation, a.g.; writing-review and editing, f.p. and a.v.; supervision, a.g. all authors have read and agreed to the published version of the manuscript.funding: this research received no external funding. we are grateful to three anonymous reviewers for their useful comments, which have enriched this work. we are indebted to the team of meisino's boys for their useful discussion on the epidemic evolution in italy. computational resources provided by hpc@polito (http://hpc.polito.it). the authors declare no conflict of interest. key: cord- - omocd q authors: fronza, r.; lusic, m.; schmidt, m.; lucic, b. title: spatial-temporal variations of atmospheric factors contribute to sars-cov- outbreak date: - - journal: nan doi: . / . . . sha: doc_id: cord_uid: omocd q the global outbreak of severe acute respiratory syndrome coronavirus (sars-cov- ) infection causing coronavirus disease (covid- ) reached over two million confirmed cases worldwide, and numbers are still growing at a fast rate. the majority of new infections are now being reported outside of china, where the outbreak officially originated in december in wuhan. despite the wide outbreak of the infection, a remarkable asymmetry is observed in the number of cases and in the distribution of the severity of the covid- symptoms in patients with respect to the countries/regions. in the early stages of a new pathogen outbreak, it is critical to understand the dynamics of the infection transmission, in order to follow contagion over time and project the epidemiological situation in the near future. while it is possible to reason that observed variation in the number and severity of cases stem from the initial number of infected individuals, the difference in the testing policies and social aspects of community transmissions, the factors that could explain high discrepancy in areas with a similar level of healthcare still remain unknown. here we introduce a binary classifier based on an artificial neural network that can help in explaining those differences and that can be used to support the design of containment policies. we propose that air pollutants, and specifically particulate matter (pm) . and ozone, are oppositely related with the sars-cov- infection frequency and could serve as surrogate markers to complement the infection outbreak anticipation. the pandemic of the new coronavirus sars-cov- causing covid- disease is testing the resilience of all the vital components of our communities, and specifically healthcare systems. sars-cov- belongs to a family of coronaviridae, the common spillover pathogens that have recently caused two viral outbreaks, sars and the middle east respiratory syndrome (mers). as sars-cov and mers-cov, new sars-cov- most probably originated from bats, however the intermediate host of the new coronavirus remains unknown to date (wu et al. ) . like other coronaviruses, sars-cov- is a spherical enveloped virus with the positive-sense rna genome. in general, coronaviruses have a relatively slow viral multiplication rate with the in vitro maximal replication efficiency at o - o c, and a rapid decrease in infectivity at higher temperatures (dulbecco and ginsberg ) . moreover, coronaviruses can remain infectious for several days on inert surfaces and in the external environment (firquet et al. ) , (van doremalen et al. ) . general indicator of transmissibility of a viral infection, basic reproduction number (r ), which indicates how many secondary infections are caused by a primary infected person, has been estimated for sars-cov- to be between and mostly using data from china (liu et al. ) . however, these estimations might not be accurate for what seems to be the faster-spreading european epidemic. the main transmission mode of sars-cov- is person-to-person contact (ghinai et al. ) . besides physical contact, transmission can occur through physiological aerosol airborne droplets composed mainly of water in the act of breathing, talking, coughing and sneezing (atkinsons j, et al. ) , (morawska and cao ) . the droplets that compose aerosol include various physiological constituents (cells, proteins, salts, small molecules) but also contain exogenous particles, such as viruses (atkinsons j, et al. ). the dimension of these droplets varies, ranging from millimeters to microns, leading to a substantial difference in the detrimental effect in the case of infective particles (stetzenbach, buttner, and cruz ) , (wong, leung, and fok ) . large droplets have a low probability to pass the upper respiratory tract, whereas the smaller particles have the capacity to reach the bronchi and lungs (atkinsons j, et al. ) . it is considered that droplets independently from the dimension, fall promptly to the ground as subordinated to gravity force, explaining the reason for the heuristic rule of maintaining a distance of a few meters to avoid the infection. however, the gravity force is counteracted by stock's friction and in the case of particles up to μm the two contrasting forces are equilibrating and the particles are floating in the air (shaman and kohn ) . air pollutants, pm . and pm are atmospheric aerosols that are classified on the basis of the particulate size (ie < . μm and < μm). toxicological studies suggest that pm . are especially harmful as smaller particles are more prone to penetrate deeper into the lungs (chan and lippmann ) . in line with this, several studies showed that exposure to particulate matter causes respiratory diseases (xing et al. ) , (rajagopalan s. et ) . similarly, atmospheric pollutant ozone (o ), an oxidant present in the environment, also contributes to the risk of respiratory illness (turner et al. ) . however, unlike particulate matter, ozone is commonly used as a method for air, water and object disinfection. indeed, in the case of an airborne influenza virus type a infection, o was found to rapidly inactivate the virus and to reduce morbidity in infected mice ( hiroshi et al., ) (wolcott, zee, and osebold ) , (jakab and bassett ) . in this study we analyzed particulate matter and ozone ambiental concentration in relation to sars-cov- infected cases from italy and three other european countries, france, germany and spain. we propose that the same atmospheric conditions that elevate the air concentration of pm and o , are also contributing to the modulation of the viral outbreak. we provide a qualitative model that predicts the outbreak severity taking into account only a reduced set of atmospheric factors. we propose that ) atmospheric conditions that elevate the pm concentration are also involved in the sars-cov- viral outbreak, and that ) the environmental ozone concentration can be a repressing factor that attenuates sars-cov- infection. the hourly concentration (μg/m ) of pm . , pm , o and nh in regional capitals of europe, and major italian cities, was retrieved in the period from th february to th of march . for seasonal analysis, the first days of march, june, september, and december were taken from the year . the files are in crib or netcdf format retrieved from http://macc-raq-op.meteo.fr/. the analysis dataset used is the ensemble multi-model that combines the values of other seven models: chimere metnorway, emep riuuk, euradim knmi/tno, lotos-euros smhi, match fmi, silam météo-france, and mocage ukmet ( https://www.regional.atmosphere.copernicus.eu ). the european domain is defined within the coordinates - e/ n and e/ n, with . degrees of horizontal and vertical resolution. the values were extracted and manipulated into a linux environment using bash scripts, cdo . . (doi . /zenodo. ), and the r packages raster (robert j. hijmans, ) and ncdf (david pierce, ). the concentration of pm . , pm , o , and nh are expressed in μg/m within an area of about km where the geographical centre of the province capital corresponds to the geometric center of the quadrant. the latitude and longitude of italian cities were derived directly from the epidemiological data provided by the protezione civile. the latitude and longitude of european cities were obtained manually. inside the italian province dataset we subdivided the data in three macroregion: north italian provinces (latitude . ), central italian provinces ( . latitude . ), and south ≥ ≤ < italian provinces (latitude . ). < to express the overall quantity in the study period for pm . , pm , nh , and o , we developed and tested two indexes, the average daily maximum (adm) and the average daily value (adv). considering the starting day of the atmospheric factors sampling ( . . ), the incubation period ( days, (backer, klinkenberg, and wallinga ) ) and the day at which we started the survey ( . . .), d was imposed as . considering that there are l geographical d = areas, and hours a day ( h ), the at a day d is a vector computed as: where is the meteorological factor value at hour h , day d , and location l . the is x hdc dv a computed as: the two indexes are similar with being more sensitive to rapid changes of the factors dm a whereas being more robust to spikes and responses to temporal consistent changes. for dv a . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . ozone, the concentrations were converted from μg/m to ppm assuming standard conditions ( atm) and using the relation: where c is the ozone concentration , mv is the molar volume approximated with an ideal μg/m ) ( gas ( . l/mol) and mm is the molar mass ( g/mol). for each species, we obtained a xn vector at a final day d, where n is the number of areas considered. given the open data policy of the italian government, for italy it was possible to obtain the daily data of the sars-cov- infected people per province and region. the demographic data were collected from the site http://www.comuni-italiani.it/provincep.html . the italian sars-cov- case data are retrieved from https://github.com/pcm-dpc/covid- in the form of daily comma-separated files. for each day we collected a set of matrices where m is , the m × n number of days (from th of february to th of march) and n is the number of the provinces or regions. in the case of the other three european countries, france, germany and spain (fgs), the data were not centralized or specified at the time of the analysis and were manually imported without the possibility to separate in subgroups of cases. the total number of cases for french, german and spanish regions, were obtained from https://www.statista.com/ referring to th of march. the pairwise spearman correlation coefficients were computed for all the atmospheric factors. conditioning plots were then applied to graphically assess the dependence of the number of infected cases per million of one factor conditioned to different levels of another factor. a generalized poisson model was fitted to estimate the association among the data showing the number of infected cases per million and the atmospheric factors. the infection data used per province are the total number of patients on th of march, whereas for the regions we used the number of total, not hospitalized and hospitalized cases. we calculated , where l= pm . , adm l pm , o and nh considering d= (endpoint th march). all the regression models were tested and adjusted for overdispersion. binary classifier based on an artificial neural network (ann) was implemented to test the capacity of the atmospheric variables to predict the epidemic escalation of the number of positive cases per million on the basis of a combination of where l= pm . , pm , nh dm a l and o . all the values were pre-scaled to be homogeneous. we defined as epidemic dm a l escalation, when the number of cases per million (at the time of the present analysis) is more than one standard deviation ( ), above the mean ( ) of the number of cases per σ = μ = . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . million in the first twenty nations listed in the coronavirus survey https://www.worldometers.info/coronavirus/ . we call this value escalation threshold ( = t e cases per million). the binary classifier was developed using a feed-forward ann with n inputs, one output and h hidden components using the neuralnet r package. the number of inputs n depends on the number of regressors used as input, whereas the number of the hidden components was selected during the training phase with the italian province datasets. the dimensionality of the input parameters was finally reduced to the usage of two variables, pm . and o . to validate the models we selected a monte carlo cross-validation strategy. the observed number of data was split randomly multiple times (n= ) into two datasets, the training ( r ) and the test ( t ), containing and provinces respectively. for each combination of regressors, we varied the number of hidden layers from to . we defined as true negative ( tn ) all the provinces in t that have a number of cases that is lower than the selected threshold. the true positives ( tp ) are provinces where the number of cases is higher than the threshold. the false positives ( fp ) are all the provinces predicted with epidemic escalation but where the number of cases is lower than the threshold. the false negatives ( fn ) are all the provinces predicted without epidemic escalation but where the number of cases is higher than the thresholds. the random assignment to r or t was performed times for each combination of the parameters. a null predictor was implemented with a random swapping of the number of cases. to measure the performances of the binary classifier and the null predictor, we used four indices: the sensitivity ( ), the specificity ( ), the accuracy ( ), to evaluate the capacity of the classifier we collected the total number of sars-cov- cases for three european nations, france, germany and spain from th of march. the atmospheric factors pm . and o were extracted in the same way as described in the study setting and data section. the data were then scaled following the same procedure, and the conditioned factors were used to feed the classifier to measure the performances on completely unseen data. the expected number of infected cases in the total of italian provinces were predicted for the months of march (spring), june (summer), september (autumn) and december (winter) using the real measured values for pm . and o atmospheric factors from seasonal datasets. the results of the predictions and the previous results were condensed into a simplified boolean model. the explaining continuous factors, pm . ( ), and ozone ( ), where and are p o concentrations, are seen as boolean variables that refer to two states, present ( or not present, depending on the unknown thresholds ( , ). as the . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . atmospheric factors considered are limited source that influences the severity of infection, all the hidden factors are summarized by a single hidden variable with unknown characteristics. (u) v the effect of the combination of the three variables on the epidemic has two states, escalating and non escalating, on the basis of the described epidemic escalation threshold . this model t e is qualitative, and the activation thresholds are not explicitly calculated. the observation t p, o that the modality of the interactions among factors derived from the regression step is not known is justifying the usage of an ann as a general approximator. the concentration of four air pollutants in italian and fgs areas were sampled over a period of days from am on the st of february until the am of the th of march. the average population of an italian province is , inhabitants with an average number of cases per million of at the time of the study. on the same day the average population and number of cases per million in the fgs regions was , , and , , respectively. in the first countries, ranked by the total number of cases at the day of the study, the average number of cases per million was . to measure the strength of association between all the variables in the provinces we used the spearman coefficients (scs) ( table ) correlation matrix for pm . , pm and nh shows a strong positive inter-variables correlation (r> . ). the number of sars-cov- cases per million (cas) shows a significant positive correlation with three of the four atmospheric variables pm . , pm , and nh ( ). . . ≤ r ≤ o instead, shows a good negative correlation with cas ( ). we also introduced . − . − ≤ r ≤ the total population of the province (pop) to assess a correlation between the number of inhabitants and atmospheric pollution. a weak correlation was found between the population, pm . and pm ( . =<r<= . ). moreover, no or little correlation was found among the population, nh and o . overall, scs show that all the tested atmospheric variables display an extremely high degree of intra-variable correlation, allowing us to half the dimensionality of the variable space. high inter-variable correlation among pm . , pm , and nh makes these variables redundant. to select the variables that mostly contribute to explain the escalation in the number of cases, we applied the generalized poisson model with the log-transformed infected counts. we reduced the number of variables selecting only the interactions among the atmospheric dm a factors, excluding the as redundant. hence, from now on we refer to the index only, dv a dm a l so that the factor species name l , as for example pm . , will be used throughout the manuscript. the latitude of the province capital was introduced as an explaining factor in the regression model. the location of the province in respect to the latitude was found as a major factor (p< . ) with a positive trend that interacts significantly with pm and o . the amount of pm, in the south of italy (green dots, lat< . ), occupy the left part of the scatter plot, . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . remaining under µg/m (pm . ) and µg/m (pm ) ( figure a and b). the rest of the provinces show a more extended range of values. the distribution of o ( figure c ), is instead more uniform in a range that goes from . to . ppm. in the south italian provinces, this factor shows no association with the total number of cases. in central italy ( . <lat< . ) and in northern italy (lat> . ) the tendency of the o concentration has clearly a negative relation with the number of sars-cov- positive cases. the validity of the regression model passed the check for normality, heteroscedasticity, influence and scale-location (supplementary figure s ). to correct for overdispersion we removed systematically all the explaining variables from the model and performed test to assess the significance of the deleted variables. as expected χ after the correlation analysis, pm . , pm and nh were strongly positively correlated (p< . ) with the number of cases whereas o showed a significant (p= . ) inverse correlation. to visually detect relations that may be obscured by the effects of other variables, a collection of conditional plots (cleveland ) figure a) and o at concentrations lower than around . ppm (first and second scatter plot in figure b ). the number of cases per million starts to be dependent on o concentration only when the pm . are higher than about µg/m (fifth and sixth scatter plot in figure c ). in summary, conditional plots show that the number of cases per million appears to be non linearly linked with the o data due to a threshold effect on the ozone linked to the pm . concentration. the same threshold effect seems to hold true for the other two analyzed variables, pm and nh (supplementary figure s ) . the ann classifier with three hidden neurons ( figure a ) returned the highest score in se, sp, acc and prc ( figure a and supplementary figure s a ), so this topology was selected to classify the datasets. the prediction on the italian provinces using pm . , o and both variables are summarized in supplementary figure s b and figure a , left panel. the classifier with both pm . and o has sensitivity, specificity, accuracy, and precision that are highly significant in respect to the null predictor. the usage of only pm . or o is still significant but with a decreased capacity to predict the escalated provinces (supplementary figure s b) . for the fgs data the results show a lower capacity of the combination of pm . and o in classifying performances (supplementary figure s a) . when only pm . was used ( figure a , right panel), the predictor behaved better than the null model for specificity, accuracy and precision. remarkably, the classification efficiency of o alone was sub-performing for all the four performance indexes. when we combined the data from italian provinces and fgs regions, the distribution of the pm ( figure b , left and middle panels) in the fgs regions was clearly lower, similar to the value observed in the south of italy (< µg/m ). the distribution of the ozone concentration is narrower in the fgs regions, with a shorter tail at the lower concentrations ( figure b , right panel). . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / next, we tested the classifier seasonal predictive behaviour taking the from the italian vm a provinces, extracted from the seasonal recordings of march (spring), june (summer), september (autumn), and december (winter) ( figure c ). zero outbreaks were predicted in june and september (summer and autumn), whereas / outbreaks were found in march (spring) and / in december (winter). taken together, classification results show that the low concentration of the particulate is a reliable predictor for the absence of epidemic escalation. at medium-high values, the model predicts well the epidemic escalation in combination with low (high epidemic escalation likelihood), or high (low epidemic escalation likelihood) concentration of ozone. this observation makes the ozone a possible predictor only if the quantity of particulate is significantly high. for all the conditions where the pm . is low, the unknown factors explained by unknown variables (v u ) are prevailing in explaining the epidemic escalation. finally, we explore the hypothesis that atmospheric conditions that favor the formation of pm and possibly viral micro droplets differentially influence the severity of sars-cov- infection. by separating hospitalized and not hospitalized cases in the italian regions, we found a significant positive correlation (r = . , p= . ) between the particulate matter quantities and the number of hospitalized patients ( figure a ), whereas for infected but non hospitalized population, this was not the case (r = . , p= . ). ozone instead, showed a significant negative correlation with the hospitalized population (r = . , p= . ) and a not significant negative correlation (r = . , p= . ) with the not hospitalized population ( figure b ). to continuously monitor the capacity of our framework to predict the spread of the sars-cov- in the areas, we regularly update the pm . and o data and perform a real-time forecasting of the infection. the results of this analysis is provided as a pdf document in https://github.com/covid upcome/europe . in the present study, we built a predictive model for the sars-cov- viral outbreak as a function of atmospheric pollutants, pm . , pm , nh and o . our hypothesis is based on the correlation between viral outbreak and physicochemical factors that contribute to the enrichment of the atmospheric particulate. based on the spearman, conditional and regression correlation analysis, we chose to favor the use of the average daily maximum values of pm . and o to build the ann classifier for prediction of sars-cov- outbreaks. the ann model with two input variables, pm . and o , and three hidden layers was trained with italian provinces, which resulted in a significant predictive ability for all four considered performance values ( figure a and supplementary figure s b ). the same model however, tested on fgs regions, showed a limited capacity to classify the epidemic escalation in those regions. a more in-depth analysis taking into account predictors with only a single atmospheric input variable resulted in the amelioration of the predictive capacity with only pm . in the fgs regions. however, systematic underperformance with respect to the null model was observed when only the o parameter was used. these results allowed us to build an informal logic approximation that accounts also for the unknown . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . causative variables ( figure b ). when the pm . concentration is low, the capacity of the pm . model to catch non escalating regions (true negatives) is high as few of them show outbreaks (sp> . ). we thus introduced the unknown causative variables to explain the rising in the cases number when pm . concentration is low ( figure b , left branch of the model). however, when the pm . concentration is high, the o is contributing positively to the capacity of the model in predicting the outbreaks ( figure a, se= . ) . this implies the existence of a variable threshold, in this case pm . concentration, that acts on the o predictive capacity. this threshold is exemplified in the right branch of the model in figure b . sars-cov- is an emerging pathogen, and our study is with this regard subjected to possible limitations, primarily access to data and time constraints. our model was validated based on publicly available data, different sampling policies and unknown fractions of untested/asymptomatic infected individuals, all of which could impact the accuracy of the collected datasets. moreover, data access to the broader set of atmospheric factors with more powerful modeling strategies could be applied to establish explicit causal relation with the infection dynamic. finally, the findings of our study should be seen in the light of the ongoing new pandemic, which makes our analysis short time windowed. on this point, temporal synchronization based on epidemiological parameters, was not implemented to synchronize the infection dynamics. nevertheless, our data support the concept that the atmospheric conditions can both ) promote the formation of persisting forms of airborne droplets charged with sars-cov- (pm . ), and ) reduce the activity of the virus (o ). we thus hypothesized that the increase of the concentration of pm . may reflect the rise of infective droplets with a diameter inferior to microns. we speculate that fast evaporation of droplets emitted by talking or sneezing increases the sustained circulation of micro-droplets that carry a higher viral load, particularly in closed spaces. in line with this, it has been shown for an airborne influenza virus that % of the viral particles are present in the droplets with a dimension between and μm. in general, droplets with a dimension bigger than μm fall immediately to the ground, whereas particles with a dimension of μm take more than an hour to reach the ground from a height of m (tellier ) . moreover, particles with an aerodynamic diameter smaller than μm remain suspended in the air for hours or days and are more able to reach the alveolar regions in the lungs (shaman and kohn ) . the droplet size can be modulated by the difference in vapour pressure, reducing it in a few seconds, at a rate of μm/sec (wang et al. ) . these observations are remarkably significant as they show that ) the distribution of the dimension of the droplets changes immediately so that the diameter at the emission source (infected individual) is consistently bigger than the diameter at the destination (susceptible individual), and ) the viral concentration in the droplet (viral particles/μl) increases, as the volume decreases quadratically respect to the radius. these evidence suggest that in unfavorable ambiental conditions, monitored by pm . formation, the amount of persistent droplets with large viral load reflects on a significant increase of the hospitalized cases ( figure a ). this supports the findings that face mask usage can be more beneficial in avoiding the spreading of the virus from an infected . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . individual than to block the viral particles at the susceptible (healthy) individual (leung et al. ) . finally, the environmental factors that influence the particulate size/concentration, may be also responsible for the seasonal dynamics of the airborne infections. our data are in agreement with this possibility (figure c) , however, a more detailed analysis is required to prove this causation for sars-cov- infection. to continue exploring seasonal and daily trends, the updated predictions in the regions of our model with pm . and o as input variables, can be found at https://github.com/covid upcome/europe . in conclusion, our analysis supports the hypothesis that the atmospheric conditions that increase the particulate matter formation, are also contributing to the severity of the sars-cov- infection. an appealing possibility that ozone might act to counteract/sterilize viral charge is to be further investigated. finally, monitoring spatial-temporal variations of atmospheric particulate and o could be used as an aid to estimate upcoming trends for the sars-cov- transmission impact. tables table . the correlation matrix between pm . , pm , nh , o , population and number of sars-cov- cases in italian provinces. suffix s in the columns and rows names (e.g. pm s) corresponds to the whereas the suffix m (e.g. pm m), represents . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . figure . correlation between sars-cov- cases in italian provinces using pm . , pm and o . the scatterplots display the values of three atmospheric factors and the number of cases per million in provinces (left panels) and selected provinces with regional capitals (right panels). a) pm . ; b) pm ; c) o . different colours represent italian provinces at different latitudes. red dots: provinces with a latitude bigger than . n; black dots: provinces with a latitude comprise between . n and . n; green dots: provinces with a latitude lower than . n. a) the schematic design of the aan classifier. a set of n input nodes are filled with a combination of the measurement of atmospheric factors v , v , …, v n . the output of the ann is a function e(v , v , …, v n ) on the atmospheric variables v , v , …, v n . b) intuitive boolean diagram of the model suggested by our conclusions. v p , v o , and v u are three input signals that correspond to pm . , o , and an unknown generic variable. the three variables are mapped to { , } using a threshold function where if the concentration of x is bigger than a threshold v x = t x and otherwise. the threshold t x is implicitly determined by the neural network. the v x = notation indicates the negation of . represents an or and an and gate. v v ⊕ ⊗ is the binary output of our model given v p , v o , and v u . . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . multiple comparisons (* p-value< . ). black asterisk indicates that the classifier performs better than the null classifier. red asterisk indicates that the classifier performs worse than the null classifier. is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . using a single atmospheric value (pm . , o ). fgs (blue dots), and random dataset (black dots). statistical analysis were performed using multiple t test corrected with sidak-boneferroni method for multiple comparisons (* p-value< . ). red asterisk indicates that the classifier performs worse than the null classifier. . cc-by . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . incubation period of novel coronavirus ( -ncov) infections among travellers from wuhan experimental measurements and empirical modelling of the regional deposition of inhaled particles in humans coplots, nonparametric regression, and conditionally parametric fits aerosol and surface stability of sars-cov- as compared with sars-cov- survival of enveloped and non-enveloped viruses on inanimate surfaces first known person-to-person transmission of severe acute respiratory syndrome coronavirus (sars-cov- ) in the usa raster: geographic data analysis and modeling influenza virus infection, ozone exposure, and fibrogenesis respiratory virus shedding in exhaled breath and efficacy of face masks the reproductive number of covid- is higher compared to sars coronavirus airborne transmission of sars-cov- : the world should face the reality ncdf : interface to unidata netcdf format data files" r package version absolute humidity modulates influenza survival, transmission, and seasonality detection and enumeration of airborne biocontaminants aerosol transmission of influenza a virus: a review of new studies long-term ozone exposure and mortality in a large prospective study evaporation and movement of fine water droplets influenced by initial diameter and relative humidity exposure to ozone reduces influenza disease severity and alters distribution of influenza viral antigens in murine lungs outdoor air pollution and asthma genome composition and divergence of the novel coronavirus ( -ncov) originating in china the impact of pm . on the human respiratory system the authors declare that they have no conflict of interest. key: cord- -pw odm authors: moccia, lorenzo; janiri, delfina; pepe, maria; dattoli, luigi; molinaro, marzia; de martin, valentina; chieffo, daniela; janiri, luigi; fiorillo, andrea; sani, gabriele; di nicola, marco title: affective temperament, attachment style, and the psychological impact of the covid- outbreak: an early report on the italian general population date: - - journal: brain behav immun doi: . /j.bbi. . . sha: doc_id: cord_uid: pw odm the outbreak of covid- is severely affecting mental health worldwide, although individual response may vary. this study aims to investigate the psychological distress perceived by the italian general population during the early phase of the covid- pandemic, and to analyze affective temperament and adult attachment styles as potential mediators. through an online survey, we collected sociodemographic and lockdown-related information and evaluated distress, temperament, and attachment using the kessler psychological distress scale (k ), the temperament evaluation of memphis, pisa, paris and san diego-autoquestionnaire short version (temps-a) and the attachment style questionnaire (asq). in our sample (n= ), % of the individuals reported no likelihood of psychological distress, whereas . % and . % displayed mild and moderate-to-severe likelihood. cyclothymic (or: . ; p< . ), depressive (or: . ; p< . ) and anxious (or: . ; p= . ) temperaments, and the asq “need for approval” (or: . ; p= . ) were risk factors for moderate-to-severe psychological distress compared to no distress, while the asq “confidence” (or: . ; p= . ) and “discomfort with closeness” were protective (or: . ; p= . ). cyclothymic (or: . ; p= . ) and depressive (or: . ; p= . ) temperaments resulted as risk factors in subjects with moderate-to-severe psychological distress compared to mild distress, while the asq “confidence” (or: . ; p= . ) and “discomfort with closeness” (or: . ; p= . ) were protective. our data indicated that a relevant rate of individuals may have experienced psychological distress following the covid- outbreak. specific affective temperament and attachment features predict the extent of mental health burden. to the best of our knowledge, these are the first data available on the psychological impact of the early phase of the covid- pandemic on a sizeable sample of the italian population. moreover, our study is the first to investigate temperament and attachment characteristics in the psychological response to the ongoing pandemic. our results provide further insight into developing targeted intervention strategies. the coronavirus disease pandemic has been spreading across italy for over a month. on march th, , the italian government implemented several emergency containment measures, including strict limitations on movement on the whole national territory, except for proven work and health reasons. these measures are unprecedented and aim to contain the epidemic after an increase in total deaths of nearly % in the hours before the decree (lazzerini and putoto, ) . the covid- outbreak is currently leading to severe mental health burden in worst-hit countries (fiorillo et al., ; kang et al., ) . containment measures, including self-isolation and social distancing, have a strong impact on the population's daily life and may negatively affect psychological well-being (brooks et al., ) . however, mixed evidence is available about the role of interindividual characteristics and demographics in determining the psychological response of a population facing large-scale stressful events. hence, it is crucial to detect possible predictors of the psychological impact during the covid- outbreak, in order to implement prompt intervention strategies (sani et al., ) . temperament refers to early-appearing individual differences in emotional reactivity, is stable across the lifespan, and has strong biological underpinnings. certain affective temperament traits and related personality constructs might, to some extent, mediate adaptive functioning, e.g., by subserving better coping mechanisms to environmental stressors (akiskal and akiskal, ; balestri et al., ) . the attachment theory postulates that the intimate bonds built with caregivers very early during infancy are crucial for social and emotional development and provide a template model for enduring patterns of emotional, cognitive and behavioral strategies in adulthood, i.e., adult attachment style (aas) (bartholomew and horowitz, ) . stressful situations are thought to activate the attachment system, and evidence supports the existence of a relationship between attachment patterns and stress responsivity during adulthood (kidd et al., ) . to the best of our knowledge, no study has investigated the determinants of psychological response to the ongoing covid- outbreak so far. we hypothesized that temperament and attachment may affect the degree of perceived psychological distress during the covid- pandemic. therefore, we aimed to evaluate the psychological impact of the covid- outbreak on the italian general population and to analyze the affective temperament and aas as potential predictive factors influencing the extent of psychological burden. the study was conducted through an online survey between april th and april th, . this timeframe was chosen to assess participants' response during an early phase of the covid- outbreak, following the italian government declaration of lockdown (decree of march th, ) and the world health organization (who) announcement of the covid- as a pandemic (march th, ) . the snowball sampling method was used to recruit participants (goodman, ) . we selected an initial set of five participants, ensuring a broad range of age, gender, occupation, education, and geographical area. each participant was asked to choose five people they consider suitable for the survey and to send them the questionnaire. further participants were reached out in the same way until data saturation. efforts were made to recruit subjects from all italian regions, which had been affected by the pandemic to different extents, so to have a representative sample of the italian population. the survey was anonymous, and data confidentiality was assured. eligible participants were aged - , had lived in italy for at least four weeks from february , were fluent in both written and spoken italian, and had at least five years of education. exclusion criteria were: non-italian language speakers; current hospitalization; a history of mental disorder. the study followed the european survey research association (esra) guidelines. all participants completed the questionnaire online via eusurvey. the study was approved by the ethics committee of the fondazione policlinico universitario agostino gemelli irccs, università cattolica del sacro cuore of rome. a dedicated, self-report questionnaire was set up to collect demographic and epidemiological variables of interest (age, gender, educational level, occupation, marital status, geographical area), medical status (lifetime history of chronic diseases, family history of psychiatric disorders), and information on lockdown conditions (living alone, changes in working activities, working on the frontline, and having direct contact with confirmed cases of covid- infection). the kessler psychological distress scale (k ; kessler et al., ) was used to assess the psychological impact of the covid- outbreak. k is a -item questionnaire intended to yield a global measure of distress experienced in the most recent -week period. we adopted the cutoff scores of > and > to detect the likelihood of mild and moderate-to-severe psychological distress, respectively (andrews and slade, ) . affective temperaments (cyclothymic, depressive, irritable, hyperthymic, and anxious) were assessed through the short version of the validated italian temperament evaluation of memphis, pisa, paris and san diego-autoquestionnaire (temps-a; preti et al., ) . aas was evaluated through the italian validated version of the attachment style questionnaire (asq; fossati et al., ) , a self-report instrument containing items. the asq comprises five subscales: ( ) "confidence", describing secure attachment; ( ) "discomfort with closeness" and ( ) "relationships as secondary", both measuring attachment avoidance; ( ) "need for approval", and ( ) "preoccupation with relationships", both assessing attachment anxiety. previous sensitivity analysis suggested that with n= the power was= . to detect a minimally interesting effect size of δ= . (α= . ; two-tailed). to fit our aims, we subdivided our sample into three groups according to k cutoffs: ) subjects without likelihood of psychological distress, ) subjects with likelihood of mild psychological distress, and ) subjects with likelihood of moderateto-severe psychological distress. analyses used standard univariate/bivariate comparisons of continuous measures (anova) and categorical measures (contingency table/χ ) to compare factors of interest (including sociodemographic, aas, and temperament characteristics) in the three groups. we used a statistical model corrected for multiple comparisons according to the bonferroni procedure (p< . /number of comparisons) to minimize the likelihood of type i statistical errors. factors significantly associated with mild or moderate-to-severe psychological distress in bivariate analyses subsequently underwent a multiple multivariate logistic regression to generate odds ratios (ors) and their % confidence intervals (cis), with no psychological distress risk/ mild psychological distress/ and moderate-to-severe psychological distress as dependent outcome measures. we examined possible multicollinearity between variables of interest by ensuring that the variance inflation factor (vif) indicator obtained from linear regression analysis was < . we used the statistical routines of spss statistics . for windows (ibm co., armonk, new york, usa). in our sample (n= ), subjects ( %) reported no likelihood of psychological distress, whereas distress, respectively. sociodemographic and epidemiological characteristics, as well as results of the univariate/bivariate analysis of temperament and aas features of the sample, are summarized in table . the three groups differed only in gender (χ = . ; p= . ) and age (χ = . ; p= . ). anovas revealed significant differences among the three groups regarding cyclothymic (f= . ; p< . ), depressive (f= . ; p< . ), irritable (f= . ; p< . ), and anxious temperaments (f= . ; p< . ). the three groups also differed in several asq dimensions, including "confidence" (f= . ; p< . ), "discomfort with closeness" (f= . ; p< . ), "need for approval" (f= . ; p< . ), and "preoccupation with relationships" (f= . ; p< . ). multinomial logistic regression identified anxious temperament (or: . ; p= . ) as a risk factor for mild psychological distress compared to no psychological distress, whereas male gender (or: . ; p= . ) was protective. cyclothymic (or: . ; p< . ), depressive (or: . ; p< . ) and anxious (or: . ; p= . ) temperaments, and the asq "need for approval" (or: . ; p= . ) were risk factors for moderate-to-severe psychological distress as compared to no distress, while the asq "confidence" (or: . ; p= . ) and "discomfort with closeness" subscales were protective (or: . ; p= . ). lastly, cyclothymic (or: . ; p= . ) and depressive temperaments (or: . ; p= . ) were identified as risk factors when comparing subjects with moderate-to-severe psychological distress to individuals with only mild distress, whereas both the asq "confidence" (or: . ; p= . ) and "discomfort with closeness" (or: . ; p= . ) subscales were protective ( the documented connection between viral epidemics and psychological distress dates back more than years ago, when menniger linked the spanish flu pandemic with psychiatric complications (menninger, ) . people's emotional responses during massive infectious disease outbreaks are likely to include feelings of extreme fear and uncertainty that, along with the separation from loved ones and the limitations of freedom, may eventually lead to dramatic mental health burden (brooks et al., ) . hence, we conducted a survey to investigate the italian population's psychological response during an early phase of the epidemic. our findings indicate that % of the general population is currently perceiving a form of psychological distress. similar results were observed both in online surveys conducted on the chinese population during the covid- pandemic (li et al., ) and among the italian general population following previous natural disasters (dell'osso et al., ) . however, the majority of subjects in our sample displayed no relevant distress. this might be due to the still relatively short exposure to the pandemic, as well as to individual features promoting resilience (mukhtar, ) . anxious temperament and male gender represented, respectively, a predictive and protective factor for mild psychological distress. on the one hand, anxious temperament, as a trait-like phenotype, is characterized by increased behavioral and physiological reactivity to mildly stressful stimuli and is more prevalent in women (akiskal and akiskal, ) . on the other hand, gender is an important biological determinant of vulnerability to psychosocial stress, in addition to genetic, socio-cultural, hormonal, and developmental factors (wang et al., ) . our results indicate that males are, to a certain degree, less likely to develop psychological symptoms in the face of a stressful event. similarly, a recent survey conducted in china one month after the covid- outbreak reported higher post-traumatic stress symptoms in women (liu et al., ) . when comparing subjects with likelihood of moderate-to-severe psychological distress to individuals with no risk, depressive, anxious, and cyclothymic temperaments, as well as the insecure-anxious attachment dimension "need for approval", appeared to be risk factors. conversely, the asq "confidence", as well as the asq "discomfort with closeness", dimensions of secure and avoidant patterns of attachment respectively, were protective. the same asq subscales were protective also for mild psychological distress, compared to moderate-to-severe distress, whereas cyclothymic and depressive temperaments proved to be predictors. depressive temperament is characterized by being pessimistic, highly self-critical, gloomy, prone to excessive worrying and striving to please others, whereas cyclothymic temperament is outlined by shifts in mood, energy, behavior, and thinking. both cyclothymic and depressive temperaments display increased stress reactivity in daily life, as well as enhanced desire for social contact (walsh et al., ) . our results suggest that cyclothymic/depressive individuals may be more likely to perceive the covid- outbreak and related containment measures as distressful and to experience increased negative affect in response to social isolation. in our sample, features of both secure and avoidant aas appeared to be protective for the risk of higher psychological burden during the covid- outbreak, compared to anxious style. a function of attachment is to regulate distress (bartholomew and horowitz, ) and evidence suggests that quality of early caregiving experiences and aas may affect stress responsivity, both at a physiological and psychological level (kid et al., ) . anxiously/avoidantly attached individuals are less able to regulate their emotions, as opposed to securely attached subjects, so that several strategies have developed internally to reduce or manage any distress experienced. bartholomew and horowitz ( ) described anxiously-attached individuals as overly dependent on others and in constant need of attention, in contrast to those high in avoidance who may feel uncomfortable in social interactions. a possible explanation to our results is that while subjects with anxious style overreport distress to ensure care will be provided, individuals with an avoidant attachment may appear as if they are very calm in a distressing situation while their internal experience may be quite the opposite. alternatively, individuals with prominent avoidant attachment features, who tend to be self-directed, and often do not exhibit distress upon social separation, might perceive self-isolation, as well as social distancing preventive measures, as less stressful compared to anxiously-attached individuals. some issues might limit the generalizability of our results. the study was carried out throughout four days and lacks longitudinal follow-up. the impact of the covid- outbreak on the italian population's mental health could worsen over time and long-term implications warrant further investigation. the survey design involved an online invitation, thus leaving unexplored the population who does not use network devices. further, we cannot determine the participation rate since it is unclear how many subjects received the survey. finally, the reliability of self-administered questionnaires may be partially biased. to the best of our knowledge, our survey results are the first showing that a relevant percentage of the italian population might have experienced from mild to moderate-to-severe psychological distress symptoms during the early phase of the covid- outbreak, and that both temperament and aas features may predict the extent of mental health burden. interventions promoting mental health among the general population should be rapidly implemented, bearing in mind individual background and characteristics. none. the theoretical underpinnings of affective temperaments: implications for evolutionary foundations of bipolar disorder and human nature interpreting scores on the kessler psychological distress scale (k ) temperament and character influence on depression treatment outcome attachment styles among young adults. a test of a four-category model the psychological impact of quarantine and how to reduce it: rapid review of the evidence gender differences in the relationship between maladaptive behaviors and post-traumatic stress disorder. a study on l'aquila the consequences of the covid- pandemic on mental health and implications for clinical practice on the dimensionality of the attachment style questionnaire in italian clinical and nonclinical participants snowball sampling impact on mental health and perceptions of psychological care among medical and nursing staff in wuhan during the novel coronavirus disease outbreak: a crosssectional study short screening scales to monitor population prevalences and trends in non-specific psychological distress examining the association between adult attachment style and cortisol responses to acute stress covid- in italy: momentous decisions and many uncertainties. lancet glob health vicarious traumatization in the general public, members, and non-members of medical teams aiding in covid- control prevalence and predictors of ptss during covid- outbreak in china hardest-hit areas: gender differences matter psychoses associated with influenza, i: general data: statistical analysis mental health and emotional impact of covid- : applying health belief model for medical staff to general public of pakistan the italian version of the validated short temps-a: the temperament evaluation of memphis, pisa, paris and san diego mental health during and after the covid- emergency in italy the expression of affective temperaments in daily life gender difference in neural response to psychological stress key: cord- -uyuazdhk authors: soraci, paolo; ferrari, ambra; abbiati, francesco a.; del fante, elena; de pace, rosanna; urso, antonino; griffiths, mark d. title: validation and psychometric evaluation of the italian version of the fear of covid- scale date: - - journal: int j ment health addict doi: . /s - - - sha: doc_id: cord_uid: uyuazdhk background: the advent of covid- worldwide has led to consequences for people’s health, both physical and psychological, such as fear and anxiety. this is the case in italy, one of the countries most affected by the pandemic. given the heightened fear concerning covid- in italy., the present study analyzed the psychometric properties of the italian version of the fear of covid- scale (fcv- s). methods: the sample comprised italian participants who were administered italian versions of the fcv- s, the hospital anxiety and depression scale (hads), and the severity measure for specific phobia–adult (smsp-a). several psychometric tests were performed to investigate the validity and reliability of the test including confirmatory factor analysis. results: analysis of the data showed satisfactory psychometric characteristics and confirmed the scale’s unidimensional properties. the seven fcv- s items had acceptable correlations with the test total (from . to . ). furthermore, the loadings on the factor were significant and strong (from . to . ). the internal consistency was very good (α = . ). construct validity for the fcv- s was supported by significant and positive correlations with the hads (r=. ) and smsp-a (r=. ). conclusions: the italian version of the fear of covid- scale is valid and reliable in assessing fear of covid- among the general italian population. during the last few months of and the first quarter of , one respiratory infectious disease has unexpectedly become a worldwide emergency, to the point that it was declared a global pandemic by the world health organization on march , . novel coronavirus disease has affected individuals in countries and territories as of april , (who ) and has resulted in months-long lockdowns of educational and nonessential business activities in many countries, including italy. covid- 's symptoms include fever, tiredness, dry cough, myalgia, and dyspnea , and its mortality rate appears to have fluctuated over the course of the past few months. after initial mortality rates of approximately % being reported (huang et al. ) , the mortality rate was later found to be between . and % chen et al. ). among the affected countries, italy currently has the highest mortality rate ( . %, over , deaths as of april , ), followed by spain ( . %, over deaths) and china ( . %, over deaths) (johns hopkins center for systems science and engineering ). together with the disease's characteristics, such as its being airborne (i.e., perceived as imminent and invisible), further aspects such as uncertainty over patient outcomes, familiarity with infected people, and mandatory change of habits imposed by the governments to protect the population's health have led many individuals across the globe to experience a generalized sense of fear (guan et al. ; huang et al. ) because similar events are unprecedented in the lifespans of many citizens worldwide. medical research globally has rightfully focused on the development of an effective vaccine wang et al. ) , while governments have dedicated themselves to the implementation of strategies for infection control to minimize the spread of the virus. nonetheless, a joint effort by medical and psychological healthcare professionals might lead to a better outcome for the entire population affected. in fact, literature regarding past virus outbreaks has already underlined the role of fear and its negative psychosocial consequences in exacerbating the harm of an infectious disease (e.g., pappas et al. ). often fueled by sensationalistic headlines in the mass media, fear can bring people to oscillate between denial and phobia, while also stigmatizing citizens racially perceived as being the source of the disease (pappas et al. ; falagas and kiriaze ) . in turn, other psychological disorders such as anxiety and depression have been found associated with fear in previous epidemics, further affecting people's quality of life negatively (e.g., ford et al. ; huang et al. ) . such consequences appear to be particularly relevant in the context of the present pandemic because social isolation (in this case, resulting from mandatory social distancing policies issued by governments) has been previously shown to be strongly intertwined with anxiety and depression symptoms in both younger and older populations (e.g., matthews et al. ; santini et al. ) . as the fear of coming into contact with individuals who may have been infected has been reported in the context of covid- (centers for disease control and prevention a, b; lin ), a new psychometric assessment tool assessing an individual's fear of covid- was recently developed, i.e., the fear of covid- scale (fcv- s), a short and valid robust assessment scale (ahorsu et al. ) . given the degree to which italy has been hit by the spread of covid- , the present study tested the scale among individuals in the italian population. the aims of the study were to (i) examine the psychometric properties of the italian the fear of covid- scale (fcv- s) utilizing confirmatory factor analysis (cfa); (ii) assess fear of covid- in an italian sample using the italian fcv- s; and (iii) confirm whether the italian fcv- s is unidimensional as was found in the original validation study by ahorsu et al. ( ) . a total of participants (age to years) volunteered to take part in the study via an online survey posted in italian online forums and social network communities (e.g., facebook). the online survey took around - min to complete. data collection occurred from march to march . inclusion criteria for volunteers were being (i) at least years old and (ii) italian-speaking citizens. all the participants completed the survey anonymously and gave their informed online consent. all procedures conducted were approved by the ethics committee of the group cognitive-behavioral psychotherapy association. socio-demographics parameters questions concerning socio-demographic aspects of the participants (e.g., age, gender, educational level) were included in the online survey. hospital anxiety and depression scale in order to assess the anxiety and depression levels of participants, the italian version of the hospital anxiety and depression scale (hads) (costantini et al. ) was used. the hads (zigmond and snaith ) is a -item scale comprising seven items relating to anxiety and seven items relating to depression. items are answered on a -point response format with a total score ranging from to for each of the two subscales. example items include "i feel as if i am slowed down" (depression) and "i get a sort of frightened feeling like 'butterflies' in the stomach" (anxiety). the higher the score, the more severe the anxiety or depression. cronbach alphas in the present study were very good for the total scale ( . ) and good for the anxiety ( . ) and depression ( . ) subscales. severity measure for specific phobia-adult the severity measure for specific phobia-adult (smsp-a) (knappe et al. ) is a -item scale that assesses the severity of specific phobias in individuals aged and older (italian version: fossati et al. ) . each item asks individuals to rate the severity of their specific phobia during the past days (e.g., "during the past seven days i felt moments of sudden terror, fear, or fright in these situations"). each item is rated on a -point scale from (never) to (all of the time). the total score can range from to with higher scores indicating greater severity of the specific phobia. cronbach's alpha in the present study was very good (. ). fear of covid- scale the fcv- s (ahorsu et al. ) is a seven-item scale that assesses the fear of covid- . the seven items (e.g., "i am most afraid of coronavirus- ") are rated on a point scale from (strongly disagree) to (strongly agree) with scores ranging from to . the higher the score, the greater the fear of covid- . for the italian fcv- s, the items were independently translated by a mother-tongue translator and internationally accepted practices for translation were employed (beaton et al. ) . additionally, the italian fcv- s was piloted on participants of different ages and education levels to investigate if there were any problems in understanding the items themselves (see appendix). to avoid the effect of the order and the sequence, the order of presentation of scales and the items within the surveys was randomized. univariate normality of the data was verified using the guidelines proposed by muthén and kaplan ( ) before the analysis of the sample's results (i.e., to check if acceptable values for asymmetry/ asymmetry and kurtosis were in the range from − to + in the case of normal univariate data distribution). the statistical analyses carried out were as follows: (i) descriptive statistics of the fcv- s items (i.e., means and standard deviations of the main items); (ii) construct and criterion validity of the italian fcv- s; (iii) the reliability of the scale, examined via composite reliability (cr) (e.g., cr values greater than . are associated with good test reliability; fornell and larcker ; netemeyer et al. ) . the internal consistency of the overall score was calculated using the cronbach alpha coefficient. in addition, the composite reliability was also taken into consideration, to validate the goodness of the test. the model fit was examined with the following: the goodness of fit index (gfi), chi-square test (χ ), degrees of freedom (df), root mean square error of approximation (rmsea), confirmatory fit index (cfi), standardized root mean square residuals (srmr), and tucker-lewis index (tli) fit indices. a good model should have the following characteristics: gfi > . , cfi and tli > . , rmsea < . , and srmr < . (i.e., browne and cudeck ) the analysis was carried out using the following statistical packages: the sample ( participants) comprised % female participants (n = ) and % male participants (n = ), with a mean age of . years (sd = . ). in terms of education level, . % had a university-level degree, % had a high-school degree, and . % had a lowerlevel educational degree. the present study analyzed the distribution of the seven fcv- s items. most items (see fig. ) were distributed asymmetrically, with the highest frequencies in the lowest values. as for asymmetry and kurtosis, most of the items were distributed in a non-normal way (the items do not fall within the range of ± , see muthén and kaplan ; tabachnick and fidell ) . more specifically, using the shapiro-wilk normality test, all items were distributed in a non-normal way (p < . ). moreover, the italian fcv- s appeared to have a unidimensional structure (i.e., a single factor). additionally, by analyzing the italian fcv- s, it had eigenvalues > in a single factor model (see gorsuch ) which suggests one factor as the optimal usable model (more specifically, the eigenvalues = . with a proportion of variance of . ). to investigate and analyze the factorial structure, since there is no unequivocal consensus in the literature (see bollen and long ; boomsma ) , different goodness of fit (gof) adaptation fig. histograms of the seven fcv- s items (note: × is item ; × is item ; the red line refers to the normal distribution curve) indices were used to confirm the dimensionality of the fcv- s. in this specific case, since the items (see table ) were distributed in a non-normal way (some items out of the range of ± [see muthén and kaplan ; tabachnick and fidell ] ), parallel analysis/diagonally weighted least squares method (dwls, polychoric correlation) was used (mindrila ) . the results showed the following: χ = . (df = , n = ), p = . with χ /df = . , the ratio of χ to degrees of freedom [df] < to consider the data-model fit as acceptable (kline ) ; chi-square test is very sensitive to the size of the sample, so several indices were used (kline ) , tli = , cfi = , rmsea = . ( % confidence interval, lower limit . , upper limit . , p = . , i.e., not significant for p = . ) and srmr = . (hu and bentler ) . the explained common variance (ecv) was . % (fornell and larcker ) . the factorial validity of fcv- s was supported by the results (cerny and kaiser ; kaiser ) . the obtained indices were sufficient and all factor loadings were high on all items (min = . , max = . ; i.e., λij ≥ . , [ferguson and cox ] ). furthermore, the modification indices suggest that there was covariance between item ("i am most afraid of coronavirus- ") and item ("when watching news and stories about coronavirus- on social media, i become nervous or anxious") and between item ("it makes me uncomfortable to think about coronavirus- ") and item ("my heart races or palpitates when i think about getting coronavirus- "). moreover, a positive relationship between all items (table ) was observed (min = . , max = . , all items are statistically significant for p = < . , polychoric correlations). these results indicate that the fcv- s presented a good fit to the data. in addition, the total score of the fcv- s on all participants produced the following statistics: mean = . , % ci [ . - . ], standard deviation = . , skewness = . , kurtosis = − . . after the confirmatory factor analysis, different types of reliability (i.e., internal consistency) and validity (i.e., construct and convergent validity) were investigated. first, construct and criterion validity were tested (cronbach and meehl ) . the total score of the fcv- s was correlated with variables that have been formerly associated with fear of covid- (i.e., anxiety and depression; ahorsu et al. ) . concurrent validity was supported by the hads (anxiety and depression levels among participants) and smsp-a (phobia among participants) as indicated by the significant positive correlations for both scales (p < . ). more specifically, the fcv- s positively correlated with the hads (r = . ) and the smsp-a (r = . ). to analyze the reliability of the measure and internal consistency, cronbach's alpha, factor determinacy index, and composite reliability (raykov ) were used. cronbach's alpha in the present study was . and could not be improved by removing any items. the factor determinacy index was . and the composite reliability was . . furthermore, age was negatively correlated with the fcv- s test but was non-significant (r = − . , p = . ). the present study investigated the psychometric properties of the italian fear of covid- scale . results indicated a stable unidimensional structure of the italian fcv- s, confirming the findings of the original validation study (ahorsu et al. ) . psychometric analyses showed the italian fcv- s's good internal reliability and consistency. construct validity was confirmed by the significant correlation with hads (which assesses the general level of anxiety and depression) and smsp-a (which assesses the level of severity of the specific phobia). this is in accordance with previous literature (e.g., brannon and schuyler ; american psychiatric association ) according to which specific phobias are often comorbid with anxiety (which is unsurprising given that phobias can be considered as a defense by individuals from anxiety; greenson ) . in fact, a significant positive correlation between fcv- s scores and scores on the hads and smsp-a was observed, confirming the scale's convergent validity. further observations demonstrated that the fear response pattern was not significantly influenced by the participant's age. this suggests that the italian fcv- s can be used to assess psychological issues caused by covid- diffusion among all ages. although no formal diagnoses concerning mood disorders were obtained (e.g., anxiety, depression), scores on the fcv- s were significantly and positively related to scores assessing depression and anxiety (hads) and the severity of the specific phobia (smsp-a); therefore, individuals with severe fear of covid- may be affected by these disorders co-morbidly. this is in accordance with previous literature indicating that, during long periods of infectious epidemics, individuals' psychophysical health can be affected by negative psychological states (e.g., anxiety, depression, and phobias; duncan et al. ; pappas et al. ; ropeik ) . besides generating concerns on a physical and psychological healthcare level, covid- has also resulted in social issues, which have been previously associated with an acceleration of an infectious disease's spread (centers for disease control and prevention a, b; bloom and cadarette ). for instance, an epidemic can overload a nation's healthcare system, especially in poor public health contexts (bloom and cadarette ) , limiting the ability of operators to deal with the problem and increasing the stress levels of both citizens and healthcare workers. in turn, stress has been previously shown to worsen both the physical and mental health of individuals, often resulting in increased use of alcohol, tobacco, or other drugs (centers for disease control and prevention c) and negatively affecting the immune system, making people more vulnerable to disease (e.g., morey et al. ) . moreover, fear experienced during an epidemics has been previously associated with the stigmatization of citizens perceived as being the source of the disease, with the risk of resulting in scuffles or, in extreme cases, civil conflicts (pappas et al. ; falagas and kiriaze ) . therefore, the italian version of fcv- s could help the general public to better understanding emotional factors correlated with the pandemic and (with a joint effort by those working in medical and psychological healthcare) ultimately lead to a better health outcome for the entire population affected. the findings of the present study should be viewed in light of some limitations. first, the participant pool only comprised a self-selected sample from the general italian population with a majority being female (and therefore, tests for gender differences were not possible). second, no formal diagnosis of mood disorders was undertaken. third, it cannot be excluded that social desirability factors might have influenced participant responses to the questionnaire. further investigation on bigger and more representative samples of italian participants is needed to confirm the preliminary results provided by the present study (e.g., a nationally representative sample with more male participants). however, total scores on the fcv- s were comparable across all ages, which suggests that the italian fcv- s is a good psychometric instrument to be used in assessing fears of covid- among italian individuals. future studies should also evaluate if individuals with underlying medical conditions associated with a higher risk of death from covid- (e.g., diabetes, hypertension, coronary heart disease, pre-existing respiratory conditions) may experience increased levels of covid- fear. moreover, future research including a larger and more representative sample should further investigate the apparent covariance between item and item and between item and item . preliminary covariance results may, in fact, be due to different factors, such as (i) an underlying factor that was not considered in the present model, (ii) the sample size, (iii) the imbalance of the sample's gender composition, (iv) idiosyncratic characteristics of the sample, and (v) the semantic similarity between the items in italian context (whittaker ) . nonetheless, the psychometric testing of the italian fcv- s demonstrates that the instrument is psychometrically robust and assesses a unidimensional construct. in short, the italian fcv- s is a reliable and valid tool for assessing the severity of fear of covid- among italian adults. open access this article is distributed under the terms of the creative commons attribution . international license (http://creativecommons.org/licenses/by/ . /), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the creative commons license, and indicate if changes were made. conflict of interest the authors declare that they have no conflict of interest. ethical approval all procedures performed in this study involving human participants were in accordance with the ethical standards of the research team's organizational ethics board and with the helsinki declaration. informed consent informed consent was obtained from all participants. appendix. italian version of the fear of covid- scale . ho molta paura del coronavirus- . mi rende inquieto (ansioso/nervoso) pensare al coronavirus- . le mie mani iniziano a sudare quando penso al coronavirus- . ho paura di perdere la vita a causa del coronavirus- . quando guardo le notizie e le storie sul coronavirus- sui social media, divento nervoso o ansioso. . non riesco a dormire perché mi preoccupo di contrarre (o avere) il coronavirus- . il mio cuore batte forte o palpita quando penso di poter contrarre il coronavirus- the fear of covid- scale: development and initial validation diagnostic and statistical manual of mental disorders guidelines for the process of cross-cultural adaptation of self-report measures infectious disease threats in the twenty-first century: strengthening the global response reporting analyses of covariance structures comorbid generalized anxiety disorder, phobia, and panic disorder single sample cross-validation indices for covariance structure covid- ): manage anxiety & stress coronavirus disease (covid- ): reducing stigma coronavirus disease (covid- ): stress and coping a study of a measure of sampling adequacy for factor-analytic correlation matrices epidemiological and clinical characteristics of cases of novel coronavirus pneumonia in wuhan, china: a descriptive study detecting psychological distress in cancer patients: validity of the italian version of the hospital anxiety and depression scale construct validity in psychological tests discovering drugs to treat coronavirus disease (covid- ) perceived vulnerability to disease: development and validation of a -item self-report instrument reaction to the threat of influenza pandemic: the mass media and the public exploratory factor analysis: a users' guide program factor at : origins, development and future directions reduced immunity to measles in adults with major depressive disorder evaluating structural equation models with unobservable variables and measurement error scala di valutazione della gravità della fobia specifica -adulto -raffaello cortina editore -ebook raffaello cortina editore phobia, anxiety, and depression clinical characteristics of coronavirus disease in china cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives generalized anxiety disorder, depressive symptoms and sleep quality during covid- epidemic in china: a web-based cross-sectional survey ibm spss statistics for windows coronavirus covid- global cases an index of factorial simplicity methodology in the social sciences. principles and practice of structural equation modeling methodology in the social sciences. principles and practice of structural equation modeling dimensional anxiety scales for dsm- : sensitivity to clinical severity social reaction toward the novel coronavirus (covid- ). social health and behavior lonely young adults in modern britain: findings from an epidemiological cohort study maximum likelihood (ml) and diagonally weighted least squares (dwls) estimation procedures: a comparison of estimation bias with ordinal and multivariate non-normal data current directions in stress and human immune function a comparison of some methodologies for the factor analysis of non-normal likert variables los angeles: muthén & muthén scaling procedures. issues and applications psychosocial consequences of infectious diseases r: a language and environment for statistical computing. vienna: r foundation for statistical computing estimation of composite reliability for congeneric measures the consequences of fear lavaan: an r package for structural equation modeling social disconnectedness, perceived isolation, and symptoms of depression and anxiety among older americans (nshap): a longitudinal mediation analysis. the lancet public health using multivariate statistics clinical characteristics of hospitalized patients with novel coronavirus-infected pneumonia in wuhan using the modification index and standardized expected parameter change for model modification coronavirus disease (covid- ): situation report- the hospital anxiety and depression scale key: cord- -syy okhi authors: dettori, marco; deiana, giovanna; balletto, ginevra; borruso, giuseppe; murgante, beniamino; arghittu, antonella; azara, antonio; castiglia, paolo title: air pollutants and risk of death due to covid- in italy date: - - journal: environ res doi: . /j.envres. . sha: doc_id: cord_uid: syy okhi the present work aims to study the role of air pollutants in relation to the number of deaths per each italian province affected by covid- . to do that, specific mortality from covid- has been standardised for each italian province and per age group ( groups) ranging from - years to > years, based on the national population figures. the link between air pollutants and covid- mortality among italian provinces was studied implementing a linear regression model, whereas the wide set of variables were examined by means of lisa (local indicators of spatial autocorrelation), relating the spatial component of covid- related data with a mix of environmental variables as explanatory variables. as results, in some provinces, namely the western po valley provinces, the smr (standardized mortality ratio) is much higher than expected, and the presence of pm( ) was independently associated with the case status. furthermore, the results for lisa on smr and pm( ) demonstrate clusters of high-high values in the wide metropolitan area of milan and the po valley area respectively, with a certain level of overlap of the two distributions in the area strictly considered milan. in conclusion, this research appears to find elements to confirm the existence of a link between pollution and the risk of death due to the disease, in particular, considering land take and air pollution, this latter referred to particulate (pm( )). for this reason, we can reiterate the need to act in favour of policies aimed at reducing pollutants in the atmosphere, by means of speeding up the already existing plans and policies, targeting all sources of atmospheric pollution: industries, home heating and traffic. air pollution is at present one of the main problems for public health. according to reports from the world health organization (who) [ ], it is responsible for million deaths worldwide every year. in europe, an estimated , people die prematurely each year from exposure to air pollutants [ ] . the main air pollutants taken into consideration are particulate matter (i.e. pm . and pm ), carbon monoxide (co) and carbon dioxide (co ) and nitrogen-based components (e.g. nox). these substances mainly derive from anthropic activities (combustion, traffic, industry), agriculture and cattle breeding which alter the atmosphere's composition [ ] . the damage caused to human health varies and depends both on the concentration of pollutants, and on individual subjectivity and the duration of exposure. several conditions can be attributed to exposure and they vary from mild and transitory to chronic forms. furthermore, it has been shown that a continuous exposure to levels of pollutants above the regulatory limits causes a chronic inflammatory and hyperergic state which can lead to a greater predisposition to infections and to the symptomatic development of disease [ ] , as well as other pre-existing immune alterations [ ] . thus, living in an area with high levels of pollutants could lead a person to be more prone to developing chronic respiratory conditions and consequently susceptible to infectious agents. in relation to the above, since who has drawn up guidelines setting exposure limits for both the short and long term. despite this, regulatory limits are not the same worldwide and vary from country to country [ ] . the recent spread of the sars-cov- virus, responsible for the covid- pandemic, has raised various questions in the academic community in relation to why some countries were primarily subject to a greater spread of the virus and suffered higher lethality [ ] . italy, in particular, was the country most affected by covid- immediately after the appearance of the virus in wuhan, china. the disease has assumed dramatic characteristics throughout the nation, with , infections and , deaths recorded to date, a higher lethality than china and other european countries and numerous differences also within the same national territory [ , ] . different hypotheses have been put forward among the possible explanations for the primary spread of the virus within the country, also in relation to the recent observations showing various similarities from the geographical and social point of view between hubei province and northern italy [ ] . notwithstanding the commercial links between italy and china, and the various opportunities for contact and importation of the virus into our country, understanding why the virus had such a sudden spread in some italian territories is one of the aspects that currently stimulates the scientific debate on an international level. moreover, some other factors may be involved in facilitating the spread of viruses into the community. actually, a previously-proven phenomenon regarding the spread of other viruses (i.e., measles) [ ] attests that the levels of atmospheric pollution and, above all, of particulates, could act as a vehicle for the spread of the virus throughout the territory. nonetheless, setti et al. recently demonstrated the presence of the sars-cov- rna on particulate matter [ ] . therefore, a combination of factors related to air quality, such as pollution and, in general, a wide range of environmental conditions, could be considered responsible for targeting the respiratory tract and weakening the population at risk, at the same time increasing their likelihood of being affected by respiratory diseases like covid- [ , ] . in italy, the areas most affected are found within the po valley, known for its high levels of air pollution, apparently sparing a large part of central italy and most of southern italy. on the basis of these premises, and of a series of recent researches, that show a higher case fatality rate in the regions of the po valley [ , ] , the present work aims to evaluate the relationship between air pollution and covid- at a provincial level. in particular, it has aimed to study the role of air pollutants and a set of environmental variables, selected from recent observations [ , ] , in relation to the number of deaths per each italian province affected by covid- . the present study did not require ethical approval for its observational design according to italian law (gazzetta ufficiale n. dated . . ). italy is located in the southern part of the european peninsula, in the mediterranean sea, and has coasts on the tyrrhenian, ionian and adriatic seas. it covers a surface area of , . km² and its population numbers , , inhabitants [ ] for an average population density of inhabitants per square kilometre. from an administrative point of view, italy is divided into in regions -one of which, trentino alto adige, is split into autonomous provinces with regional competencies. most of the population is concentrated within the geographical area of the po valley, surrounded by the alpine and apennine mountains, and to the east by the adriatic sea towards the po delta and the area represents italy's economic "core". this geographical area includes the provinces of: turin; venice; vercelli; novara; milan; bergamo; brescia; verona; vicenza; padova; asti; alessandria; piacenza; parma; reggio emilia; modena; bologna; forlì-cesena; rimini; pavia; lodi; cremona; mantova; rovigo; ferrara; and ravenna. almost million people live in this area of approximately , km , with a density ( inhabitants per km ) double that of the rest of the peninsula, with a higher concentration in the main urban areas of the greater milan metropolitan area. covid- data takes into account the number of total infected people and the number of deaths as at th june at a provincial level, as reported by the italian ministry of health, and as collected by the national institute of statistics (istat) [ ] . environmental data come from istat, ispra (higher institute for environmental protection and research), il sole ore (an economic and business newspaper, which provides constant reports on economical facts), aci (automobile club d'italia), and legambiente (non-profit association for environmental protection) [ , , [ ] [ ] [ ] . the data relating to the environmental pollutants investigated,(i.e., pm , pm . and no ), come from the official italy-wide monitoring and refer to the average pollution values obtained from detection by the control units. the sources of data acquisition relating to environmental variables are shown in table . the spatial units selected are italian provinces, the intermediate levels between municipalities and regions, still in use as statistical units also by istat. while many provinces lose or change their administrative role, those selected are spatial units provided by istat as of . specific mortality from covid- has been standardised for each italian province and per age group ( groups) ranging from - years to > years, based on the national population figures. the indirect standardisation process initially provided for the calculation of national specific mortality by age group, obtained by dividing the number of covid- deaths to june and confirmed by the national institute of health (iss) [ ] by the defined age groups. thus, the number of deaths expected in the italian provinces for the age groups previously identified and based on the provincial populations, was calculated according to the formula: where n i is the specific age group population in each observed area (province); r i is the national mortality rate for the specific age group. the standardized mortality ratio (smr) was obtained by comparing the number of events observed in each province with the respective number of expected events: where d is the number of observed deaths for covid- ; e the number of expected deaths. finally, the % confidence intervals ( % ci) were calculated as proposed by vandenbroucke [ ] . in order to evaluate the correlation between the environmental variables a pairwise correlation analysis was performed. the link between air pollutants and covid- mortality among italian provinces was studied implementing a linear regression model. the analysis included, as independent variables, important air pollutants such as pm , no and pm . and various factors specific to the urban environment such as cycle paths, pedestrian streets, trees, soil, public and urban green areas, motorcycles and cars (table ) . these variables, previously studied in a recent observation [ ] , and selected on the basis of the direct (vehicular traffic) and indirect (urban green, cycle paths, land use and pedestrianised areas) role in relation to the presence of the air pollutants investigated, were compared with the dependent variable (smr) outcome. in particular, linear regression was carried out using the periodic averages of the atmospheric pollutants in question, obtained from the detections carried out by the official monitoring systems (control units). the level of significance was established at p < . with a type i error of %, the confidence intervals were calculated at %. statistical analysis was performed using stata . (stata corp, college station, tx, us), and medcalc (medcalc software ltd., ostend, belgium). for comparison reasons, some analyses on spatial autocorrelation among variables have been taken into consideration [ , , , ] . the wide set of variables were examined by means of lisa (local indicators of spatial autocorrelation), relating the spatial component of covid- related data (i.e., cases and deaths per province) with a mix of environmental variables as explanatory variables, such as annual average of pm , and pm , no , numbers of trees per inhabitants and urban green areas, number of vehicles and cycle paths, as reported in table . lisa enables an evaluation of the similarity of various observations to their neighbouring provinces, for each location [ , , , ] . in particular, lisa produces results as clusters of areas characterised by varying levels of similarity. high-high values indicate the presence of both strong values of the phenomenon and high similarity with its neighbouring provinces. low-low values represent low values of the phenomenon and low similarity with its neighbors. high-low values represent high values of the phenomenon and low similarity; at the same time, low-high values indicate low values and high spatial similarity. non-significant values represent situations of low importance of attribute and spatial closeness. the standardized mortality ratio (smr) compared the covid- mortalities with that which was expected, basing on national official data. in provinces, namely the western po valley provinces, including the mountainous ones, and on the adriatic coast of the regions of emilia romagna and marche, the standardized mortality ratio is much higher than expected (table ) . table shows the average annual values of pm by reference province. the data relating to smr for the selected provinces are shown graphically in figure . the correlation coefficients between the environmental variables investigated are shown in table . table shows the results of the linear regression analysis for the association between environmental variables (pm , no , pm . , cycle paths, pedestrian areas, trees, soil, urban green spaces, motorcycles and cars) and smr. the presence of pm (p= . , % ci: . - . ) was independently associated with the case status. no significant association with case status was found with the other variables (p > . ). considering only the pm variable, the relationship with the smr is shown in figure . furthermore, figure shows the nationwide situation, divided by province, in relation to the smr, and at the same time highlights the geographical distribution in relation to the data on pm . the map of smr distribution in italy considers provinces grouped in classes. the two classes below unity present covid- related mortality values lower than those expected. the intermediate class presents the value around unity and slightly higher than unity, thus showing a mortality in line with the expectations, or slightly higher, while the other remaining two classes present the values where covid- related mortality is much higher than that expected. with reference to the pm map, the ranges are based on the who and italian limits for particulate, based on multiple of (i.e., μg/mc is the who limit; μg/mc is the italian limit, recently reduced from μg/mc, etc.). regression analysis has been compared also in terms of spatial autocorrelation, for comparison purposes. the results of the spatial autocorrelation analysis are summarised and shown in figure and refer only to the smr and pm . figure shows [ , ] the results for lisa on smr and pm . the results demonstrate clusters of high-high values in the wide metropolitan area of milan and the po valley area respectively, with a certain level of overlap of the two distributions in the area strictly considered milan. covid- mortality highly depends on the prevalence of the disease and transmission rate, and human contacts among the population represent a key factor. nevertheless, other factors could play a role in favouring the infection, such as genetic factors [ ] , or environmental factors [ , ] . the existence of relationships between air pollution and covid- mortality was first hypothesized, taking into consideration air pollution from the particulates (pm . and pm ) and nitrogen based components deriving from human activities. the idea was that the presence of air-related pollutants can put pressure on the health conditions of the populations at risk and offer preconditions for the development of respiratory related diseases and their complications, including some that are life-threatening, which may explain the high case fatality rate observed in the po valley area [ ] . table ranks the italian provinces where the smr as of th june was higher than , indicating a higher increase in mortality than expected. this is a set of provinces, where, by means of example in the po valley area, a city like lodi, with a value of . , presents an increase times more than expected. it can be seen that the most affected areas are found in the po valley and are particularly characterised by a high average yearly value of pm . a similarity in the density classes, which mainly include between and people per square kilometer is also noticeable. the analysis of covid- related mortality shows quite a clear divide between northern italy on one side and central and southern italy on the other, along the apennine mountain chain, with values higher (much higher than expected) in the north, and values in line with the expected mortality in other italian regions, particularly in the south. thus, based on observations recently published by the authors, a set of environmental variables was selected in order to evaluate the association with mortality from covid- in the italian provinces. bearing in mind that the main source of some air pollutants may be the same, we performed a pairwise correlation analysis between the air pollutants, and some of these variables showed a statistically significant correlation. in particular, the highest correlation index (coefficient = . ) was found between pm and pm . . this aspect is attributable to the fact that the particulate detection systems currently in place return the average concentration of pm including the fraction relating to pm . . therefore, the two variables are strongly correlated. nevertheless, as uncovered by the linear regression, only pm was found to have a strong relation to the number of deaths distributed for each italian province affected by covid- . it is important to consider this factor for risk stratification, as close monitoring of air emissions and appropriate measures implemented at a local level may help to reduce air pollution. moreover, although, as mentioned, pm is strongly correlated with pm . , multivariate linear regression analysis has enabled us to highlight how the variable pm has a dependence relationship with specific mortality for covid- such as to be the only statistically significant variable, to the point that it can be considered as an independent predictor of mortality for covid- , and an early indicator of epidemic recurrence as suggested by setti et al. [ ] . furthermore, some spatial autocorrelation can be found regarding pm and smr, showing a certain level of similarity between the most affected provinces in northern italy by covid- and the highest recorded values of pm . numerous studies hypothesise a correlation between the presence of air pollutants and the mortality. in particular, in addition to a recent systematic review of the literature [ ] , which highlights the important contribution of pm , pm . and no as triggers for the spread and lethality of covid- , one us study has shown that a small increase in long-term exposure to pm , leads to a large increase in the mortality rate for covid- [ ] while, currently, a study by the iss and the ispra called pulvirus aims to investigate the controversial link between air pollution and the spread of the pandemic and the physical-chemical-biological interactions between fine dust and viruses [ , ] . in any case, these are observational studies, it will therefore be necessary to carry out further studies for an evaluation especially from an etiological point of view [ , ] . what has emerged therefore is that a profound reflection on the monitoring of air emissions is required, in particular of pm , which did not substantially decrease during the lockdown. in fact, through monitoring it is possible to verify the effectiveness of the measures implemented at a local level to reduce air pollution. this precaution should be included in the agreement of the po valley for the improvement of air quality, signed in bologna during the g environment ministers' meeting of th june , by the minister of the environment and the presidents of lombardy, piedmont, veneto and emilia -romagna. moreover, the pandemic could represent an opportunity to contrast the community outrage due to the perception of the environmental risks [ , ] . as far as the environmental pollution and both the spread of the virus in generating outbreaks and case fatality rate are concerned, this research appears to find elements to confirm the existence of a link between air pollution, this latter referred to particulate (pm ) and covid- mortality. particular atmospheric conditions in the early weeks of may have aggravated the environmental situation in the po valley area. quite an evident divide between northern italy on one side and central and southern italy on the other is clear to see, with provinces north of the apennine mountains presenting values higher in the north, and lower values of mortality -also lower than those expected -in the south, in all the periods considered. a spatial distribution of covid- related deaths through smr presents similarities in the spatial patterns drawn especially with particulates, as represented by the regression analysis thus far presented. as regards to suggestions in terms of policies, we can reiterate the need to act in favour of policies aimed at reducing pollutants in the atmosphere, by means of speeding up the already existing plans and policies, targeting all sources of atmospheric pollution: industries, home heating and traffic [ , ] . investment in clean transport and building should therefore be reinforced, starting also from rapidly applicable measures, i.e. road washing, pollution eating paints, façades and plants, paying particular attention to the housing sanitary conditions to assure healthy and salutogenic environments for the population [ ] [ ] [ ] [ ] . stato dell'ambiente can atmospheric pollution be considered a co-factor in extremely high level of sars-cov- lethality in northern italy? geographical overlap between alpha- antitrypsin deficiency and covid- infection in italy: casual or causal? air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide report sulle caratteristiche dei pazienti deceduti positivi a covid- in italia -report basato su dati aggiornati al luglio covid- italia. monitoraggio della situazione deaths in sars-cov- positive patients in italy: the influence of underlying health conditions on lethality why italy first? health, geographical and planning aspects of the covid- outbreak the effects of air pollution and meteorological factors on measles cases in lanzhou sars-cov- rna found on particulate matter of bergamo in northern italy: first evidence green soap. a calculation model for improving outdoor air quality in urban contexts and evaluating the benefits to the population's health status covid- and living space challenge. well-being and public health recommendations for a healthy, safe, and sustainable housing the covid- infection in italy: a statistical study of an abnormally severe disease case fatality rate analysis of italian covid- outbreak geographical analyses of covid- 's spreading contagion in the challenge of global health risks impatto dell'epidemia covid- sulla mortalità totale della popolazione residente primo quadrimestre qualità della vita epidemia covid- a shortcut method for calculating the per cent confidence interval of the standardized mortality ratio spatial autocorrelation, concepts and techniques in modern geography local indicators of spatial association-lisa the role of air pollution (pm and no ) in covid- spread and lethality: a systematic review exposure to air pollution and covid- mortality in the united states et alpotential role of particulate matter in the spreading of covid- in northern italy: first observational study based on initial epidemic diffusionbmj searching for sars-cov- on particulate matter: a possible early indicator of covid- epidemic recurrence airborne transmission route of covid- meters/ feet of inter-personal distance could not be enough environment and health: risk perception and its determinants among italian university students environmental risks perception among citizens living near industrial plants: a cross-sectional study covid- and cities: from urban health strategies to the pandemic challenge. a decalogue of public health opportunities walkable environments and healthy urban moves: urban context features assessment framework experienced in milan housing demand in urban areas and sanitary requirements of dwellings in italy healthy design and urban planning strategies, actions, and policy to achieve salutogenic cities attendees of the th course urban health. strategies for disease prevention and health promotion in urban areas: the erice charter j o u r n a l p r e -p r o o f ☒ the authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☐the authors declare the following financial interests/personal relationships which may be considered as potential competing interests:j o u r n a l p r e -p r o o f key: cord- - vucm s authors: franzo, giovanni; tucciarone, claudia maria; moreno, ana; legnardi, matteo; massi, paola; tosi, giovanni; trogu, tiziana; ceruti, raffaella; pesente, patrizia; ortali, giovanni; gavazzi, luigi; cecchinato, mattia title: phylodynamic analysis and evaluation of the balance between anthropic and environmental factors affecting ibv spreading among italian poultry farms date: - - journal: sci rep doi: . /s - - - sha: doc_id: cord_uid: vucm s infectious bronchitis virus (ibv) control is mainly based on wide vaccine administration. although effective, its efficacy is not absolute, the viral circulation is not prevented and some side effects cannot be denied. despite this, the determinants of ibv epidemiology and the factors affecting its circulation are still largely unknown and poorly investigated. in the present study, ibv qx (the most relevant field genotype in italy) sequences were obtained between and from the two main italian integrated poultry companies. several biostatistical and bioinformatics approaches were used to reconstruct the history of the qx genotype in italy and to assess the effect of different environmental, climatic and social factors on its spreading patterns. moreover, two structured coalescent models were considered in order to investigate if an actual compartmentalization occurs between the two integrated poultry companies and the role of a third “ghost” deme, representative of minor industrial poultry companies and the rural sector. the obtained results suggest that the integration of the poultry companies is an effective barrier against ibv spreading, since the strains sampled from the two companies formed two essentially-independent clades. remarkably, the only exceptions were represented by farms located in the high densely populated poultry area of northern italy. the inclusion of a third deme in the model revealed the likely role of other poultry companies and rural farms (particularly concentrated in northern italy) as sources of strain introduction into one of the major poultry companies, whose farms are mainly located in the high densely populated poultry area of northern italy. accordingly, when the effect of different environmental and urban parameters on ibv geographic spreading was investigated, no factor seems to contribute to ibv dispersal velocity, being poultry population density the only exception. finally, the different viral population pattern observed in the two companies over the same time period supports the pivotal role of management and control strategies on ibv epidemiology. overall, the present study results stress the crucial relevance of human action rather than environmental factors, highlighting the direct benefits that could derive from improved management and organization of the poultry sector on a larger scale. a total of qx sequences were included in the final dataset. of those, belonged to "company a" and to "company b". the sampled farm location is reported in fig. . overall, farms were mainly located in the "pianura padana" region (central area of northern italy) and, to a lesser extent, in north-eastern, north-western, central and southern italy. although the two companies tend to operate in different italian regions, a clear overlapping was present in the high densely populated poultry area of northern italy (fig. ). qx-population genetics parameter estimation. all considered field sequences formed a monophyletic group including only italian strains ( supplementary fig. ). tempest investigation revealed that the positive correlation between genetic divergence and sampling time (i.e. r = . ) was suitable for phylogenetic molecular clock analysis . the tmrca of the overall qx population in italy (i.e. qx genotype introduction) was estimated in . [ hpd: hpd: . - . ] using the structured coalescent approach. almost identical results were obtained including a third "ghost" deme (i.e. an estimated deme for which no sequences were available, representative of other unsampled companies and farms) in the analysis or using the "traditional" coalescent approach. when strains collected from integrated poultry companies were considered independently, the tmrca was predicted in . [ hpd: [ hpd: . - for company a and in . [ . - . ] for company b. the viral population dynamics evidenced a substantially constant ne*t (effective population size * generation time, or relative genetic diversity) with the remarkable exception of the period between mid- and mid- , when a sudden fluctuation was observed. however, a quite different scenario was demonstrated between the two integrated poultry companies. in fact, company a was featured by a substantially constant population size, with a minor decrease affecting particularly the period - . however, the ne*t hpd were relatively broad and at odds with the significance of the observed variations. on the contrary, a much more changeable pattern was observed in company b (fig. ) . migration among companies. the structured coalescent model fitted with the two company, evidenced the presence of separate clades (fig. a) for the companies, with only exceptions, represented by strains sampled in company b but clustering in the company a clade, thus suggesting the migration of strains from company a to company b. accordingly, the migration rate from company b to company a was . * − [ hpd: . * − - . * − ], while the one from company a to company b was . * − [ hpd: . * − - . ]. the figure . location of farms from which samples have been obtained. different companies have been color coded. samples collected in company b but clustering with company a clade have been colored in red (herein named "imported"). farm location has been jittered using an internal routine of ggplot library to guarantee anonymity. the map was generated in r (version . . ), using the library ggmap . phylogeographic analysis. all the samples phylogenetically belonging to company a but collected from company b originated from farms located in the high densely populated area of northern italy (fig. ) . the continuous phylogeographic analysis reconstructed a spreading pattern originating from a single introduction in emilia romagna region (company a), followed by a progressive expansion and persistence at high level in the pianura padana region. more rarely, spreading episodes toward other italian regions were observed (fig. ) . after qx introduction, the infection wave front increased slowly approximatively until , when a rapid expansion led to the final distribution range by the middle of (fig. ) . accordingly, the dispersal velocity progressively increased in the first years after qx genotype introduction, peaking in the period - and then remaining essentially constant, despite some fluctuations (fig. ) . the presence of a high dispersal velocity after , when no further increase in wave front was observed, suggests that ibv continued to circulate at high rate after its first establishment in a region. the analysis of the effect of different environmental factors on qx genotype dispersal velocity led essentially to negative results (i.e. absence of significant correlation). the only exception was represented by the poultry density www.nature.com/scientificreports www.nature.com/scientificreports/ sl model, which was positively and significantly correlated to viral dispersal velocity: d = . , percentage of d with p-value < . = %. despite the economic relevance, the epidemiology of ibv and the factors affecting its behavior have been only partially investigated. even if a huge amount of knowledge and literature has accumulated over time, most of the reports are anecdotal or based on the analysis of single clinical outbreaks , , . although relevant pieces of information could be obtained, the risk of being biased by personal believes or the particular condition under www.nature.com/scientificreports www.nature.com/scientificreports/ investigation is high. a certain caution is thus required when inferring and extending the same conclusion on a broader/general scale. moreover, most of the available studies are focused on avian influenza and, to a lesser extent, newcastle disease and infectious laryngotracheitis , , . the aim of the present study was to construct an objective and statistically sound framework to understand ibv field strains behavior, the effect of control measures and the factors conditioning their epidemiology. the field of phylodynamics, and all related extensions, provides an invaluable tool for the study of viruses and particularly of rapidly evolving ones, whose evolution can be measured in "real time", over the course of an epidemic . ibv qx genotype is the most relevant field strain in italy , , and despite a relatively long circulation and the efforts devoted to its control, it still remains one of the main menaces for poultry industry profitability. therefore, the understanding of the forces shaping its epidemiology would be of remarkable relevance in order to prevent the induced damages, rather than try to control them. remarkably, the italian ibv strains appear to originate from one introduction events only, as previously reported . therefore, it was possible to reconstruct ibv italian strain evolution and epidemiological pattern without the biasing effect of strains recently introduced from other counties. the implemented approach allowed to reconstruct the migration history of the qx genotype over time. the estimated introduction, in emilia romagna region, shortly predates the first detection, posing in favor of the effectiveness of the italian monitoring and early detection systems. all the analyses, independently of the underlying statistical model, support that company a was the first introduction site (fig. ) . thereafter, the virus circulation was limited to farms belonging to this company for years, until approximatively , when company b became involved. contextually, a progressive increase in diffusion speed was noticed (fig. ) , not unexpectedly considering the rising number of involved farms (especially at the border between veneto and lombardy regions, where most farms are located) and thus the increase in spreading potential and opportunity. the high farm density of this area has been described as a risk factor for different infectious diseases , and ibv seems to be no exception. interestingly, the viral population size remained relatively constant in this time period, evidencing that, even if qx strains were able to effectively spread from farm to farm, their replication was adequately controlled, likely by effective vaccination strategies. actually, a certain slowdown in dispersal velocity was noticed in - , potentially because of a progressive decrease in naive populations availability. a dramatic change was observed in , when a new spreading wave (fig. ) and an increase in diffusion rate (fig. ) and population size (fig. ) were detected. a more detailed analysis demonstrated that this variation affected company b only (fig. ) . a previous study has ascribed this episode to a change in the vaccination scheme adopted by this company, which moved from a heterologous mass+ b based vaccination to a mass only vaccination leading to an increased viral circulation and clinical outbreaks number . moreover, experimental studies demonstrated a significant reduction in r in vaccinated groups compared to unvaccinated ones . it can therefore be speculated that the increase in infectious pressure within-farm and the higher flock susceptibility to infection could have enhanced the risk of ibv spreading to other farms and regions. in support of this hypothesis, the geographical spreading affected mainly northern italian farms (where company b is located). moreover, when a new double vaccination was implemented, the decrease in viral population size was mirrored by a reduction of dispersal velocity. www.nature.com/scientificreports www.nature.com/scientificreports/ continuous phylogeography showed that the areas interested by a more intensive viral circulation were those featured by a higher poultry density, and this evidence was confirmed by a statistically significant correlation between poultry density and dispersal velocity. the association between spatial proximity and farm infection is probably the most consistently reported risk factor for poultry infectious diseases , , . although an airborne transmission has been proposed for ibv, its occurrence has rarely been demonstrated experimentally . however, the spatial proximity likely increases the likelihood of a greater number of horizontal contacts between farms, including the movement of people, vehicles and fomites between farms, as well as sharing similar risk factors (e.g. environmental conditions, climate, presence of wild animals, etc.) , . based on these premises, the presence of segregated poultry companies should represent an effective obstacle to viral shedding and the obtained results partially confirm these evidence. the strains from different poultry companies formed two independent clusters, which suggests the effectiveness of independent production flow/chain in protecting farms from exogenous introductions. additionally, the application of adequate biosecurity measures, enforced also by the italian legislation, likely contributed in limiting new strain introduction. the exceptions to this general rule were farms located in the high densely populated poultry area of northern italy, where an overlap between the two companies occurs. the unidirectionality of the viral flux from company a to company b implies that other factors, besides spatial proximity, must be in place. a detailed survey could shed some insights into relevant factors like different biosecurity measures, structural factors, vaccination strategy etc.. the mediation of other "actors" cannot also be excluded. in fact, the analysis of just two companies, however predominant they are on the italian poultry sector, cannot be considered an accurate depiction of the italian situation. remarkably, the inclusion of a third deme (representative of other unsampled companies and farms) in the analysis model highlighted that several transmission events could be mediated by smaller entities operating in the same region. actually, the high migration rate estimated between company b and this ghost deme poses in favor of its pivotal role in maintaining an active ibv circulation. even if the idea of modeling demes for which no sequences are available could seem counterintuitive, previous studies showed that the structured coalescent can provide meaningful estimates even in absence of samples from one population and this approach has already been applied and proven effective for other diseases, including ebola . since also company a was evaluated in the same analysis run, the absence of relevant links between this company and the ghost deme further supports the analysis reliability, posing in favor of an actual interaction between company b and the ghost deme rather than a mere low specificity of the method. a less effective control of ibv infection could be speculated for small companies, whose management capability and resources are limited compared to big-integrated companies. in fact, all italian farms have to follow national legislation dictating the minimum biosecurity measures to be applied. however, integrated poultry farms, part of major companies, enforce additional managerial practices to increase biosecurity levels. personnel and veterinarian formation, internal audits and periodic controls guarantee a higher level of application of the required standards, compared to most of small non-integrated farms. the higher spatial overlap and the likely sharing of some infrastructures (e.g. streets, accessory personnel, services and infrastructures) could nevertheless have a negative indirect effect on the major companies, especially in northern italy where company b is located. however, differences between company a and company b in the application of biosecurity measures and production flow management could also explain the different ibv epidemiology, as demonstrated by the dissimilar patterns in viral population fluctuations in the two companies (fig. ) . a further risk factor that would deserve further investigation is the presence of the rural sector, which is highly concentrated in the densely populated poultry area of northern italy. this sector is characterized by a complex mix of growers, dealers and backyards flocks, often applying poor biosecurity measures and linked together by a poorly traceable contact network . although interactions with industrial poultry farming is hardly discouraged, illegal/indirect interactions have been documented and multiple epidemiological connections could result in a bidirectional transmission between the two sectors, as demonstrated in the italian low pathogenicity avian influenza (ai) outbreaks occurred in - . after these episodes, a stricter legislation has been developed, imposing limits to animal movements and more active surveillance in the rural sector. nevertheless, no measures were taken for the monitoring and control of ibv in these enterprises, and therefore their role as sources of encroachment in intensive farming cannot be excluded. other environmental factors do not seem to play a relevant role in affecting viral dispersal. while climatic conditions like temperature, humidity and wind could actually affect viral viability and spreading, their effect could be circumvented by a transmission mediated by "fast-moving" vectors like trucks, personnel and, potentially, wild species , . more surprising could be the non-significant role of road density. however, it must be stressed that the available raster reported the overall density of roads, which could significantly differ from those preferentially used for live animal or their byproduct transportation, hindering the detection of an otherwise plausible risk factor. therefore, the mapping of the live animal transportation pathways could provide remarkable benefits in ibv (and other infectious diseases) epidemiology understanding and control. the present study demonstrates that ibv spreading potential is mainly affected by farm and poultry density overall, which can be reasonably claimed as a major risk factor. other environmental/climatic variables do not seem to affect ibv epidemiology, stressing the pivotal role of human action and thus highlighting the direct benefits that could derive from an improved management and organization of the poultry sector on a larger scale. actually, the integration of poultry production seems to provide a relevant constrain to ibv circulation, even though some differences were noted between the two considered companies. in fact, despite differences in management and applied control strategies likely playing a role, the presence in the same area of other minor poultry companies seems to represent a major issue, probably due to the less effective infection control ascribable to the sometimes lower organization capability and resources of small enterprises. the present study results emphasize the need of an active sharing of sequences and related molecular epidemiology data originating from all the actors in poultry production, allowing a proper depiction of the viral exchange dynamics, based on actual data rather www.nature.com/scientificreports www.nature.com/scientificreports/ than estimations. the obtained information would represent a fundamental substrate for the implementation of effective and shared efforts for the infection control on a broad regional scale. ibv strain sampling, diagnosis and sequencing. samples were collected for routine diagnostic purpose in the period - from poultry flocks belonging to the two main poultry companies (here named company a and company b) operating in italy, which account together for about % of italian poultry production. samples were obtained mainly from outbreaks of respiratory disease, following a standard protocol that enforced the collection of a pool of tracheal swabs from randomly selected birds. for each sampling, collection date and farm localization were recorded. all considered samples had been performed in the context of routine diagnostic activity and no experimental treatments or additional assays were implemented during the study. therefore, no ethical approval was required to use specimens collected for diagnostic purpose. additionally, several samples from company a were already sequenced using the same protocol and published in franzo et al. . when detailed information on sampling farm and time could be traced back, these samples were included in the study. the permission to use the collected samples for research purpose was obtained from each company. swab pools were resuspended in ml of pbs and vortexed. thereafter, rna was extracted from µl of the obtained eluate using the high pure viral rna kit (roche diagnostics, monza, italy) kit. diagnosis was performed by amplification and sanger sequencing of the hypervariable region of the s region using the primer pair described by cavanagh et al. . obtained chromatograms quality was evaluated using finchtv (http://www. geospiza.com) and consensus sequences were generated using cromaspro (cromasproversion . ). sequence dataset preparation. all obtained sequences plus the reference dataset provided by valastro et al. ( ) were aligned using mafft and a phylogenetic tree was reconstructed using iq-tree selecting as the best substitution model the one with the lowest akaike's information criterion, calculated using jmodeltest . the strains clustering with the gi- lineage (previously known as qx genotype) were selected and further evaluated for the presence of recombination in the considered region using rdp and gard : to limit the computational burden the sequences were clustered using a % identity threshold using cd-hit and a single representative sequence for each cluster was selected. these sequences plus the valastro et al. ( ) references were re-aligned and recombination analysis was performed. recombinant sequences, including the ones belonging to the same cluster, were removed from the dataset. finally, the dataset was re-expanded to the original size and sequences identical or closely related (p-distance < . ) to the qx-based vaccines administered in italy were also excluded. to evaluate the distribution of italian gi- strains in the international scenario, an extensive dataset of s ibv sequences was downloaded from genbank and a phylogenetic tree was reconstructed as previously described. to reduce computational complexity and increase interpretation easiness (without losing information), only one sequence representative of all identical ones was selected using cd-hit and included in the analysis. the presence of an adequate phylogenetic signal was assessed by a likelihood mapping analysis performed with iq-tree. tempest was used to preliminarily evaluate the temporal signal of the italian qx phylogeny and therefore the applicability of molecular clock-based methods . strain migration among integrated poultry companies. ibv qx strain migration among companies was evaluated using the structured coalescent-based approach implemented in the multitypetree extension of beast . according to this model, the considered population is divided in a series of demes, which can be imagined as different islands, featured by their own populations size and interconnected by a certain migration rate among them. in the particular italian qx scenario, the serially sampled (i.e. with known collection date) strains were used to infer the migration rate and history between the two integrated poultry companies (i.e. considered as different demes) over time. additionally, the bayesian approach implemented in beast allowed to contextually estimate other population parameters, including the time to most recent common ancestor (tmrca), evolutionary rate and population size. accounting for the presence of other farms and companies operating in the italian poultry sector, which could take part in or mediate the viral transmission among the investigated major companies, a third "ghost" deme (a deme for which no sequences were available) was added to the model . the priori of the ghost deme size was set to one tenth of the other demes, according to the estimated poultry population distribution. however, broad priori distribution (i.e. relatively uninformative priori) was chosen to avoid constrains or biases in the parameter posterior estimation. for all analyses, the best substitution model (tn + g ) was selected based on the bayesian information criterion, calculated using jmodeltest , while the relaxed lognormal molecular clock model was selected based on marginal likelihood calculation and comparison using the path sampling and stepping stone method . the final estimations were obtained performing a million generation markov chain monte carlo run, sampling parameters and trees every twenty thousand generations. results were visually inspected using tracer . and accepted only if mixing and convergence were adequate and the estimated sample size was greater than for all parameters. parameter estimation was summarized in terms of mean and % highest posterior density (hpd) after the exclusion of a burn-in equal to % of the run length. maximum clade credibility (mcc) trees were constructed and annotated using treeannotator (beast package). results consistency was also evaluated performing a "traditional" serial coalescent analysis in beast . . . the same substitution and clock model of the structured coalescent analysis were selected, while a nonparametric skyline population model was chosen to reconstruct the viral population dynamic over time . independent ( ) : | https://doi.org/ . /s - - - www.nature.com/scientificreports www.nature.com/scientificreports/ analysis for each integrated company were also performed using the same approach but generating two new datasets including only the sequences collected from a specific company. however, sequences introduced from one company to the other were excluded from the company-specific analysis since they did not share a common evolution history. continuous phylogeography and determinants of ibv spreading. the history of qx dispersal was reconstructed over time using the continuous phyogeographic approach described by lemey et al., using beast . . . substitution and clock models were selected as previously described. similarly, the gamma relaxed random walk was preferred over the other phylogeographic continuous diffusion models based on the marginal likelihood calculation and comparison using the path sampling and stepping stone method , . the final estimations were obtained performing a million generation markov chain monte carlo run, sampling parameters and trees every twenty thousand generations. results were visually inspected using tracer . and accepted only if mixing and convergence were adequate and the estimated sample size was greater than for all parameters. the reconstruction of qx movements over time within italian borders was obtained using spread , summarizing and visualizing the full posterior distribution of trees obtained in continuous phylogeographic analyses . pattern and determinants of viral spreading were evaluated as described by (dellicour et al.) , using the seraphim r library . the history of lineage dispersal was recovered from the posterior trees generated using beast and annotated with ancestral longitude and latitude reconstruction. particularly, the distance, duration and velocity of spatial dispersal were recoded as vectors and used to generate different summary statistics of viral spreading, including dispersal velocity and maximal wave front distances (measured from the location of the tree root). several environmental/social variables were considered to determine if they were associated with the dispersal rate of ibv lineages. the environmental rasters describing the variables of are shown in supplementary fig. . more in detail, the values in the raster (i.e. altitude, population density, poultry density, temperature, etc.) were used to associate a weight to the abovementioned vector. two models of spatial movements were considered: ( ) "straight line (sl) path" model, assuming a straight movement between the starting and ending locations of each branch (i.e. the branch weight is computed as the sum of raster cells through which the straight line passes); ( ) "least cost (lc) path" model, using a least cost algorithm (i.e. the branch weight is computed as the sum of the values of cells transition values between adjacent cells along the least-cost path). in this model, the analyzed environmental variable can be considered both as a conductance (i.e. enhancing viral dispersal through the cells with higher values) or resistance factor (i.e. allowing an easier dispersal through cells with lower values). both instances were evaluated for each considered factor. the obtained "environmental" weights were used to calculate a regression with the branch duration and the corresponding coefficient of determination (r env ) was obtained. a null coefficient of determination (r null ) was also calculated assuming the null raster (i.e. when only the spatial distance of each movement is assumed to affect branch duration). the statistic d = r env -r null was selected as final outcome, and describes how much the regression is strengthened when the spatial variation in the environmental variable is included. to account for the phylogenetic uncertainness, the d statistic was calculated for each tree of the posterior distribution. however, for computational constraints, the number of posterior trees was down-sampled to after discharging a % burn-in. only the environmental variables with more than % of d statistics > were considered for further analysis. particularly, the significance of d statistic of those variables was assessed against a d null distribution obtained by randomizing times the phylogenetic nodes location under the constraint that branch length remained equal. a p-value was generated for each initial tree, therefore a percentage of the trees with p-value < . could be calculated, which can be interpreted as a posterior probability of observing a significant correlation between lineage movements and considered environmental variable. according to , a percentage of p-value < . greater than % was considered a strong evidence that the environmental variable is associated to viral movement speed . molecular evolution and emergence of avian gammacoronaviruses s gene-based phylogeny of infectious bronchitis virus: an attempt to harmonize virus classification review of infectious bronchitis virus around the world effect of different vaccination strategies on ibv qx population dynamics and clinical outbreaks sjaak) & cook, j. k. a. factors influencing the outcome of infectious bronchitis vaccination and challenge experiments infectious bronchitis virus variants: a review of the history, current situation and control measures age-dependent immune responses and immune protection after avian coronavirus vaccination a novel variant of the infectious bronchitis virus resulting from recombination events in italy and spain porcine circovirus type (pcv ) evolution before and after the vaccination introduction: a large scale epidemiological study genetic data provide evidence for wind-mediated transmission of highly pathogenic avian influenza modelling the wind-borne spread of highly pathogenic avian influenza virus between farms avian influenza virus infections. ii. experimental epizootiology of influenza a-turkey-wisconsin- virus in turkeys characterisation of influenza a viruses isolated from turkeys in england during avian influenza in caged laying chickens isolation of avian influenza virus in texas assessing the probability of introduction and spread of avian influenza (ai) virus in commercial australian poultry operations using an expert opinion elicitation transmission parameters of highly pathogenic avian influenza (h n ) among industrial poultry farms in northern italy in - risk maps for the spread of highly pathogenic avian influenza in poultry explaining the geographic spread of emerging epidemics: a framework for comparing viral phylogenies and environmental landscape data exploring the temporal structure of heterochronous sequences using tempest (formerly path-o-gen) risk factors for the introduction of high pathogenicity avian influenza virus into poultry farms during the epidemic in the netherlands in risk factors for highly pathogenic h n avian influenza virus infection in poultry during the - epidemic in italy a cross-sectional survey of australian chicken farms to identify risk factors associated with seropositivity to newcastle-disease virus wind-borne transmission of infectious laryngotracheitis between commercial poultry operations phylogenetic and epidemic modeling of rapidly evolving infectious diseases continued use of ibv b vaccine needs reassessment after its withdrawal led to the genotype's disappearance think globally, act locally: phylodynamic reconstruction of infectious bronchitis virus (ibv) qx genotype (gi- lineage) reveals different population dynamics and spreading patterns when evaluated on different epidemiological scales transmission of infectious bronchitis virus within vaccinated and unvaccinated groups of chickens control of avian influenza in poultry studies on australian infectious bronchitis virus. iv. apparent farm-to-farm airborne transmission of infectious bronchitis virus estimating population parameters using the structured serial coalescent with bayesian mcmc inference when some demes are hidden new routes to phylogeography: a bayesian structured coalescent approximation proroga e modifica dell' ordinanza agosto e successive modificazioni, concernente: «misure di polizia veterinaria in materia di malattie infettive e diffusive dei volatili da cortile epidemiology and control of low pathogenicity avian influenza infections in rural poultry in italy coronaviruses in avian species -review with focus on epidemiology and diagnosis in wild birds coronaviruses in poultry and other birds longitudinal field studies of infectious bronchitis virus and avian pneumovirus in broilers using type-specific polymerase chain reactions mafft multiple sequence alignment software version : improvements in performance and usability w-iq-tree: a fast online phylogenetic tool for maximum likelihood analysis jmodeltest : more models, new heuristics and parallel computing rdp : detection and analysis of recombination patterns in virus genomes gard: a genetic algorithm for recombination detection cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences efficient bayesian inference under the structured coalescent improving the accuracy of demographic and molecular clock model comparison while accommodating phylogenetic uncertainty bayesian coalescent inference of past population dynamics from molecular sequences phylogeography takes a relaxed random walk in continuous space and time spread : interactive visualization of spatiotemporal history and trait evolutionary processes seraphim: studying environmental rasters and phylogenetically informed movements spatial visualization with ggplot this research was partially founded by the grant (bird / ) from the department of animal medicine, production and health, university of padua. g.f., a.m. and m.c. planned the study, g.f., c.m.t., m.l., t.t., r.c., p.p. performed laboratory work and generated the sequences obtained in the present study, c.m.t. and m.l. curated the sequences dataset, g.f. analyzed the data, m.c., a.m., p.m., g.t., g.o., l.g. supervised the respective research groups, g.f. wrote the manuscript, c.m.t., m.l., m.c. revised and improved the manuscript. all authors reviewed and agreed on the current version of the manuscript. the authors declare no competing interests. supplementary information is available for this paper at https://doi.org/ . /s - - - .correspondence and requests for materials should be addressed to g.f.reprints and permissions information is available at www.nature.com/reprints.publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.open access this article is licensed under a creative commons attribution . international license, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the creative commons license, and indicate if changes were made. the images or other third party material in this article are included in the article's creative commons license, unless indicated otherwise in a credit line to the material. if material is not included in the article's creative commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. to view a copy of this license, visit http://creativecommons.org/licenses/by/ . /. key: cord- -l wjwn authors: fiorillo, andrea; sampogna, gaia; giallonardo, vincenzo; del vecchio, valeria; luciano, mario; albert, umberto; carmassi, claudia; carrà, giuseppe; cirulli, francesca; dell’osso, bernardo; nanni, maria giulia; pompili, maurizio; sani, gabriele; tortorella, alfonso; volpe, umberto title: effects of the lockdown on the mental health of the general population during the covid- pandemic in italy: results from the comet collaborative network date: - - journal: european psychiatry : the journal of the association of european psychiatrists doi: . /j.eurpsy. . sha: doc_id: cord_uid: l wjwn background: the coronavirus disease (covid- ) pandemic is an unprecedented traumatic event influencing the healthcare, economic, and social welfare systems worldwide. in order to slow the infection rates, lockdown has been implemented almost everywhere. italy, one of the countries most severely affected, entered the “lockdown” on march , . methods: the covid mental health trial (comet) network includes italian university sites and the national institute of health. the whole study has three different phases. the first phase includes an online survey conducted between march and may in the italian population. recruitment took place through email invitation letters, social media, mailing lists of universities, national medical associations, and associations of stakeholders (e.g., associations of users/carers). in order to evaluate the impact of lockdown on depressive, anxiety and stress symptoms, multivariate linear regression models were performed, weighted for the propensity score. results: the final sample consisted of , participants. among them, . % of respondents (n = , ) reported severe or extremely severe levels of depressive symptoms, . % (n = , ) of anxiety symptoms and . % (n = , ) reported to feel at least moderately stressed by the situation at the dass- . according to the multivariate regression models, the depressive, anxiety and stress symptoms significantly worsened from the week april – to the week april to may (p < . ). moreover, female respondents and people with pre-existing mental health problems were at higher risk of developing severe depression and anxiety symptoms (p < . ). conclusions: although physical isolation and lockdown represent essential public health measures for containing the spread of the covid- pandemic, they are a serious threat for mental health and well-being of the general population. as an integral part of covid- response, mental health needs should be addressed. there is no doubt that the coronavirus disease (covid- ) pandemic, and its related containment measures such as lockdown, is affecting mental health of the general population worldwide [ ] [ ] [ ] . this is an unprecedented event, which is influencing the healthcare, political, economic, and social systems [ ] . given the high level of contagiousness, as well as the lack of appropriate treatments and vaccines, almost all countries have adopted confinement measures, including lockdown, home isolation and physical distancing [ ] . while most of the clinical and research efforts have been directed to reduce the effects of the virus on physical health [ ] [ ] [ ] , its short-and long-term effects on mental health are causing a second wave of pandemic, which has been mostly neglected [ ] [ ] [ ] . furthermore, the pandemic represents a traumatic event which has differential effects at individual and population levels. at the individual level, high rates of depression, anxiety, fear, panic, anger, and insomnia have been documented in studies mainly carried out in china or from short-term reports [ ] [ ] [ ] . at the population level, the pandemic is associated with a range of psychosocial adversities, including economic hardship and financial losses (due to unemployment and reduced income), school closures, inadequate resources for medical response, domestic violence, and deficient distribution of basic good necessities [ ] . the psychopathological consequences include the fear of contracting the disease and of dying, losing livelihoods and loved ones, uncertainty and worries about the future, social discrimination, and separation from families and caregivers [ ] [ ] [ ] [ ] . this is why the current pandemic represents a new, complex and multifaceted form of psychosocial stressor [ ] , being completely different from other natural disasters [ ] , such as earthquakes or tsunamis [ , ] , wars, terroristic attacks, mass conflicts, or economic crisis [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] , and also from previous epidemics, such as severe acute respiratory syndrome (sars), middle east respiratory syndrome (mers) and ebola [ , ] . italy has been the first western country heavily affected by the pandemic, and it has been the country with the highest number of infected and dead people for many weeks [ ] . on march , , the italian prime minister has placed million people under lockdown. this measure has been prolonged for weeks, until may , . this period is known as "phase one," during which all not necessary activities have been closed, more than , people have died and almost , people have been homeisolated. during the initial phase of the pandemic, the outbreak in italy seemed to have a greater severity of the disease, with a higher case fatality rate (cfr) than previously observed in china ( . vs. . %) [ ] . the excess in covid- mortality was higher in men than in women living in northern cities versus in central and southern italy (men: + % and + % and women: + % and + %, respectively), with an increasing trend by age [ ] . from may , a gradual reopening of financial and commercial activities has taken place (known as "phase two" of the national sanitary emergency). a few, short-term studies have already shown the impact of lockdown on the mental health of the italian general population in the first days of "phase one" [ ] [ ] [ ] . we have decided to carry out an online survey using several validated assessment instruments in order to evaluate the impact of the lockdown on the mental health of italian population throughout the different weeks of phase one [ ] . in particular, in this paper we aim to: (a) report the levels of depressive, anxiety and stress symptoms in a large sample of the italian general population; (b) explore the levels of depressive, anxiety and stress symptoms during the different weeks of lockdown; and (c) identify possible risk and protective factors for mental health outcome. the covid mental health trial (comet) is a national trial coordinated by the university of campania "luigi vanvitelli" (naples) in collaboration with nine university sites: università politecnica delle marche (ancona), university of ferrara, university of milan bicocca, university of milan "statale", university of perugia, university of pisa, sapienza university of rome, "catholic" university of rome, university of trieste. the center for behavioral sciences and mental health of the national institute of health in rome has been involved in the study by supporting the dissemination and implementation of the project according to the clinical guidelines produced by the national institute of health for managing the effects of the covid- pandemic. the comet trial includes three phases: phase one consists in the dissemination of a survey on the impact of lockdown and its related containment measures on the mental health of the italian general population; the second phase consists in the development of a new psychosocial online supportive intervention [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] for the management of the consequences on mental health of the pandemic; the last phase consists in the evaluation of the efficacy and feasibility of the experimental psychosocial intervention in a randomized control trial. the results of phase are described in this paper. the study has been approved by the ethical review board of the coordinating center (protocol number: /i). the primary outcome of the study is the severity of depressive-anxiety symptoms evaluated with the depression, anxiety, stress scale (dass- ) [ ] . secondary outcomes include the levels of global mental health status, of obsessive-compulsive and post-traumatic symptoms, presence and severity of insomnia, the levels of perceived loneliness and the presence of suicidal ideation/suicidal thoughts. furthermore, exploratory variables include coping strategies, levels of post-traumatic growth, perceived social support and resilience. the dass- evaluates the general distress on a tripartite model of psychopathology [ ] and is a reliable and valid measure in assessing mental health in the general population [ ] , which has been already adopted in previous research on sars [ ] and covid- [ , ] . the dass consists of items grouped in three subscales: depression, anxiety, and stress. each item is rated on a -level likert scale, from (never) to (almost always). the total score is calculated by adding together the response values of each item, with higher scores indicating more severe levels of depressive, anxiety, and stress symptoms. the score at the dass-depression subscale (e.g., "i felt that i had nothing to look forward to") is divided into normal ( - ), mild ( ) ( ) ( ) , moderate ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) , severe ( ) ( ) ( ) ( ) ( ) ( ) ( ) , and extremely severe depression ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) . the score at the dass-anxiety subscale (e.g., "i was worried about situations in which i might panic and make a fool of myself") is divided into normal ( - ), mild ( - ), moderate ( ) ( ) ( ) ( ) ( ) , severe ( ) ( ) ( ) ( ) ( ) , and extremely severe anxiety . the score at the dass-stress subscale (e.g., "i tended to over-react to situations") is divided into normal ( - ), mild ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) , moderate ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) , severe ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) , and extremely severe stress ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) . the general health questionnaire (ghq)- items version explores participants' mental health status through six positively worded items (e.g., "have you been able to concentrate"?) and six negatively worded items (e.g., "have you lost much sleep over worry?"). the standard scoring method recommended by goldberg for the need of case identification is called "ghq method." scores for the first two types of answers are " " (positive) and for the other two are " " (negative). threshold ≥ at ghq identifies people with a probability > % of having a mental health problem [ ] . the obsessive-compulsive inventory-revised version (oci-r) consists of items rated on a -level likert scale, ranging from to . the total score is calculated by adding all single items. scores above the threshold of are indicative of an ocd diagnosis [ ] . the insomnia severity index (isi) includes seven items rated on a -level likert scale (from to ), with a total score ranges from to [ ] . the suicidal ideation attributes scale (sidas) consists of five items assessing frequency, controllability, closeness to attempt, level of distress associated with suicidal thoughts and impact on daily functioning. each item is assessed on a -level likert scale, with a total score ranging from to . in case of scoring " -never" to the first item, all other items are skipped, and the total score is zero. the presence of any suicidal ideation is considered indicative of risk for suicidal behavior, while a cut-off of is used to indicate high risk of suicidal behavior [ ] . the severity-of-acute-stress-symptoms-adult scale (sass), which consists of nine items rated on a -point scale (from = not at all to = extremely), has been used to assess the presence of traumatic stress symptoms. the total score ranges from to , with higher scores indicating a greater severity of acute stress disorders [ ] . the impact of event scale (ies)-short version measures the traumatic reactions in people who have experienced traumatic events. each item is rated on a -point scale ranging from (not at all) to (often). the ies evaluates the dimensions of intrusion, avoidance, and alteration in arousal [ ] . the ucla loneliness scale-short version is an eight-item scale designed to measure subjective feelings of loneliness, as well as feelings of social isolation. each item is scored on a -level likert scale from = never to = often [ ] . the brief-cope consists of items grouped in subscales [ ] . each item is rated on a -level likert scale from = "i have not been doing this at all" to = "i have been doing this a lot." coping strategies are divided in maladaptive strategies, including denial, venting, behavioral disengagement, self-blame, self-distraction and substance abuse, and adaptive coping strategies, which include emotional support, use of information, positive reframing, planning and acceptance. two other subscales include religion and humour. the short form of post-traumatic growth inventory (ptgi) is a -item assessment instrument grouped into five dimensions: relating to others, new possibilities, personal strengths, spiritual change, and appreciation of life. items are rated on a -point likert scale, from = "i did not experience this change as a result of my crisis" to = "i experienced this change to a very great degree as a result of my crisis". higher scores indicate higher levels of post-traumatic growth [ ] . the connor-davidson resilience scale (cd-risc), which includes items rated on a -level likert scale, is subdivided into the following five factors: (a) personal competence, high standards, and tenacity; (b) trust in one's instincts, tolerance of negative affect, and strengthening effects of stress; (c) positive acceptance of change and secure relationships; (d) control; and (e) spiritual influences. higher values indicate higher levels of resilience [ ] . the multidimensional scale of perceived social support (mspps) consists of items rated on a level-likert scale, from = "absolutely false" to = "absolutely true". items are grouped into three dimensions: family support, support by friends and support by significant others. higher values correspond to higher levels of perceived support [ ] . the maslach burnout inventory (mbi) has been used to evaluate the levels of burn-out in medical personnel [ ] . data regarding healthcare professionals are not included in this paper since they are out of the aims of the study and will be reported in subsequent analyses. respondents' socio-demographic (e.g., gender, age, geographical region, working and housing condition, etc.) and clinical information (e.g., having a previous physical or mental disorder, using illicit drugs or medications, etc.) have been collected through an ad-hoc schedule. the phase one of the comet trial consists in an online survey carried out between march and may in the italian adult population. the survey has been implemented through a multistep procedure: (a) email invitation to healthcare professionals and their patients; (b) social media channels (facebook, twitter, instagram); (c) mailing lists of universities, national medical associations and associations of stakeholders (e.g., associations of users/carers); and (d) other official websites (e.g., healthcare or welfare authorities websites). the online survey has been set up through eusurvey, a web platform promoted by the european commission ( ). the survey has been officially launched on march , , and it takes approximately min (range - min) to be completed. the full study protocol is available elsewhere [ ] . descriptive statistics were performed in order to describe the sociodemographic and clinical characteristics of the sample. the time points of data collection were recorded and coded using the variable "week" . therefore, geographical regions of respondents were recoded using a binary variable "severely impacted area." this variable has been entered in the regression model in order to evaluate the direct impact of living in an area with a higher risk of being infected rather than the impact of geographical area per se. we hypothesized that individuals living in the most affected areas should have presented more severe symptoms compared with those living in less affected areas. by order of the italian health authority, persons subject to quarantine are forbidden to move from home or residence for days, with the aim to separate persons exposed (or potentially exposed) to the infectious agent from the general community for reducing the contagion rate. people who have been subjected to those restrictions were coded using the binary variable "quarantine." in order to adjust for the likelihood of participants of being exposed to covid infection in each week, a propensity score was calculated [ ] . this methodological choice was due to the fact that the propensity score produces a better adjustment for differences at baseline, rather than simply including potential confounders in the multivariable models. the propensity score was calculated using as independent variables age, gender, socioeconomic status and living in a severely impacted area [ ] . in the final regression model, the inverse probability weights, based on the propensity score, were applied in order to model for the independence between exposure to the infection, outcomes and estimation of causal effects. in order to evaluate factors associated with the severity of depressive, anxiety and stress symptoms at dass- (primary outcomes), multivariate linear regression models were performed, including as independent variables: being infected by covid- , having a pre-existing mental disorder, being a healthcare professional. furthermore, in order to evaluate the impact of the duration of lockdown and of other related containment measures on the primary outcomes, the categorical variable "week" was also entered in the regression models. the models were adjusted for the rate of new covid cases and of covid-related mortality during the study period, as well as for several socio-demographic characteristics, such as gender, age, occupational status, having a physical comorbid condition, hours spent on internet, levels of perceived loneliness, health status, number of cohabiting people, level of satisfaction with one's own life, with cohabiting people, with the housing condition. missing data have been handled using the multiple imputation approach [ ] . statistical analyses were performed using the statistical package for social sciences (spss), version . and stata, version . for all analyses, the level of statistical significance was set at p < . . the final sample consisted of , participants, % female (n = , ), with a mean age of . ( . ) years; half of respondents were in a stable relationship, living with the partner ( . %, n = , ) ( table ). the vast majority of participants were employed ( . %, n = , ) and . % (n = , ) shifted to smart working during the pandemic. (n = , ) of respondents lost their job during the pandemic. % spent more time on internet than usual, more frequently for instant messaging ( . %, n = , ), searching for information ( . %, n = , ), or using social networks ( . %, n = , ). about . % of cases (n = , ) suffered from a pre-existing physical illness, mainly cardiovascular diseases ( . %), osteo-articular disorders ( . %), thyroid dysfunctions ( %), and diabetes/dyslipidaemia ( . %). . % (n = , ) reported to have a pre-existing mental disorder, more frequently anxiety ( . %) and depressive disorders ( . %). % of respondents (n = , ) were healthcare professionals. almost all participants ( . %, n = , ) scored above the threshold of at the ghq, indicating the risk of having any mental health problem. in particular, depressive symptoms were moderate in . % of respondents (n = , ) and severe or extremely severe in . % (n = , ); anxiety symptoms were moderate in . % (n = , ) of respondents and severe or extremely severe in . % (n = , ); stress symptoms were at least moderate in . % (n = , ) ( table ) . moderate to severe levels of insomnia were found in . % of respondents (n = , ). about . % (n = , ) of the sample scored above the threshold for clinical relevance of obsessivecompulsive symptomatology, with a global severity of obsessivecompulsive symptoms of . (ae . ) at oci-r. suicidal ideation is reported by . % (n = , ) of the sample, with a mean score of . ( . ) at the sidas. participants showed high levels of avoidance and hyperarousal symptoms ( . ae . and . ae . , respectively), with lower levels of intrusive symptoms ( . ae . ) at the ies-r. . % (n = , ) reported to feel alone, . % (n = , ) to feel excluded by others and . % (n = , ) feel that "other people are around them, but not together with them", at the ucla. at the brief-cope, we found that respondents more frequently used adaptive coping strategies, such as planning ( . % of participants, n = , ), acceptance ( . %, n = , ), and active coping ( . %, n = , ). as regards maladaptive coping strategies, . % (n = , ) of the sample used venting, % (n = , ) self-blame and . % (n = , ) self-distraction. moreover, a relatively high proportion of respondents ( . %; n = , ) reported to use psychoactive medications in order to cope with the situation. at the ptgi, participants reported that they found "something positive" out of this situation, with high levels of "appreciation for life" ( . %, n = , ), feeling closer to other people ( . %, n = , ), being more satisfied of everyday life ( . %, n = , ) and increased ability to handle difficult situations ( . %, n = , ). furthermore, respondents reported a good level of resilience with a mean score of . ae . at the cd-risc. finally, the majority of participants declared to feel supported by family ( . %, n = , ) and friends ( . %, n = , ), with a mean score of . ae . at the mspps family support subscale and of . ae . at the mspps friend support subscale (table ) . the levels of depressive symptoms increased over the period of the lockdown. in particular, depressive symptoms changed from . ae . in the week march to april to . ae . in the week april to may (p < . ). anxiety symptoms increased from . ae . in the week march to april to . ae . in the week april to may (p < . ). furthermore, the levels of stress symptoms increased from . ae . in the week march to april to . ae . in the week april to may (p < . ). these increases were higher in female participants compared to males (figures - ; p < . ). the multivariate regression analyses are reported in table . according to the multivariate regression models, weighted for the propensity score, weeks of exposure to the pandemic and to the related containment measures were significantly associated with worsening of depressive symptoms, with beta coefficient ranging from . ( % confidence interval, ci: . - . protective factors against the development of psychiatric symptoms included higher levels of satisfaction with one's own life and with cohabiting people, and living with a higher number of family members (p < . ). the comet is the first trial evaluating the global impact of the covid- pandemic and its related containment measures on several dimensions of mental health in a large sample of the italian population. one of our main findings is the presence of moderate to severe levels of depressive, anxiety, and stress symptoms which are higher than those found in china [ , , ] . this difference could be due to the type of immediate health response in the two countries, with clear lockdown measures from the beginning of the pandemic in china [ ] and a more fragmented preventive approach in italy, which may have increased the levels of fears and uncertainty in this country [ ] [ ] [ ] ] . in fact, the uncertainties about the pandemic progression, the "hypochondriac concerns" [ ] and fear that the epidemic is difficult to control represent triggering factors for the development of mental health problems [ , ] . moreover, studies carried out during natural disasters, war, fires and terroristic attacks found high levels of depressive/anxiety-related symptoms in the general population [ ] [ ] [ ] [ ] [ ] [ ] [ ] , but nevertheless they were significantly lower compared to those we found in our study. these data confirm that the current pandemic is an unprecedented event in terms of its impact on the mental health of the general population. a second interesting finding of our survey is that the levels of anxiety, depressive and stress symptoms increased over time, being more severe in the last weeks of the lockdown, as also found in our regression models controlled for all socio-demographic characteristics of respondents. this finding confirms the hypothesis that the duration of containment measures significantly influences mental health and well-being of the general population, as also found by sibley et al. [ ] in a sample of the general population in new zealand. moreover, this trend has not been influenced by the rate of covid cases and covid mortality rates in italy, highlighting that these public measures-although being necessary for infection control-should be removed as soon as possible in order to safeguard public mental health. female participants are at higher risk of developing depressiveanxiety symptoms, as already shown in small italian samples [ , ] and in previous outbreaks [ ] . this finding can be due to the higher incidence in women of anxiety-depressive disorders [ ] [ ] [ ] [ ] and of anxious, cyclothymic and depressive temperaments in women [ ] , also in community-based samples [ ] . moreover, being affected by a pre-existing mental health problem represents an independent significant risk factor for the development of depressive, anxiety and stress symptoms, as already reported by plunkett et al. [ ] and hao et al. [ ] . this finding suggests the need to provide as soon as possible adequate and tailored supportive interventions to mentally ill patients, who represent fragile and at-risk individuals that have been overlooked during the initial phases of the pandemic [ ] [ ] [ ] [ ] [ ] . during the lockdown participants reported an increased time spent on internet, which was associated with a higher risk of developing mental health problems, thus not confirming our hypothesis of a protective effect played by internet on mental health. this finding may be due to the diffusion through internet of uncontrolled and unreliable information and fake news, which may have increased the levels of anxiety and depressive symptoms in people who are alone and with lower levels of education [ ] . this finding highlights the need for media professionals to receive an appropriate training, in order to provide unbiased and nonsensationalistic information during catastrophic events. being unemployed, retired or housewife was significantly associated with higher levels of anxiety-depressive symptoms [ ] . in the uk, belonging to a socio-economic disadvantaged group increased the risk of developing mental health problems, according to a gradient across the different weeks of the lockdown [ ] . this finding highlights the need for global, multi-level socio-economic initiatives aiming to reduce the negative effect of the pandemic on the society [ ] . these data should also be interpreted considering the high rate ( . %) of suicidal ideation/suicidal thoughts found in our sample. the rate of suicidal ideation found in our sample is quite impressive, compared with the % found in a previous epidemiological study carried out in italy [ ] . several factors may contribute to the increased rate of suicidal ideation in the italian general population, including uncertainty about the future, loneliness, physical distancing, unemployment, economic recession and interpersonal violence [ ] . all these risk factors should be taken into account in the implementation of actions aiming to prevent suicide [ ] [ ] [ ] . participants reported several disturbances in sleep quality and patterns, as already found in other studies carried out in china and in other european countries [ , ] . the public health containment measures implemented worldwide have markedly changed daily routines and may have had an impact on sleep pattern and on the risk of developing other mental health problems [ , ] . in order to develop tailored innovative preventive and/or therapeutic strategies, the specific socio-demographic and clinical predictors of sleep problems should be identified. finally, good levels of perceived social support and of posttraumatic growth in the aftermath of the pandemic have been reported from the italian general population participating in our survey. it may be that the italian socio-cultural context, with strong family ties and social relationships, may have positively impacted on the perception of mutual social support [ ] . however, longitudinal studies may help to evaluate changes in the levels of posttraumatic growth, resilience, and social support in the subsequent phases of the ongoing health crisis [ ] . our study has several strengths. this is the first study carried out in different geographic italian regions with a large sample from the general population during the lockdown period. validated and reliable assessment instruments have been used in order to investigate several domains of mental health and psychological wellbeing according to a propensity score analysis. moreover, as primary outcome we have selected the same assessment tool (the dass- ) used in studies carried out in china in order to allow direct comparisons between the two countries. although the dass- scores in the italian general population prior of the pandemic are not available, the comparison of our findings with national statistics (https://www.epicentro.iss.it/mentale/epidemiologia-italia) document higher levels of anxiety, depressive and stress symptoms during the pandemic. therefore, the increased frequency of depressive-anxiety symptoms in our sample could be interpreted as covid- related, although this causal association should be further investigated. in any case, we believe that the analysis of dass- over the different weeks of lockdown provide an important contribution to the field in order to clarify the direct impact of the pandemic on the mental health. we are aware that the use of an online tool is not the best methodological choice, since it may have excluded elderly people or those living in socially disadvantaged contexts [ ] . however, this choice was necessary in order to reach a large portion of italian population in a short time and in a pandemic situation, when faceto-face contacts are forbidden [ ] . finally, it must be acknowledged that collected data are related to depressive or anxiety symptoms, which cannot be considered as sufficient to formulate a diagnosis of depressive/anxiety disorders. therefore, this survey represents an initial step for the promotion of appropriate screening procedures in the general population for the early detection of full-blown mental disorders. the present study has several clinical implications: (a) to promote mass screening campaigns for the general population in order to identify the presence of subthreshold mental disorders; (b) to disseminate informative intervention on how to deal with the mental health consequences of the pandemic; and (c) to support at-risk population-mainly people with pre-existing mental health problems and covid- patients-with tailored innovative psychosocial interventions. in conclusion, there is the need to address mental health needs as an integral part of covid- response. in fact, although physical isolation and lockdown represent essential public health measures for containing the spread of the covid- pandemic, they are a serious threat for mental health and well-being of the general population. it is necessary to get prepared if a next emergency will come, in order to provide appropriate community-based mental health service responses to the population. financial support. this research received no specific grant from any funding agency, commercial or not-for-profit sectors. conflicts of interest. the authors have no conflicts of interest to disclose. data availability statement. the dataset is not available for sharing. ethical standards.the authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the helsinki declaration of , as revised in . the ethical review board of the university of campania "l. vanvitelli" has approved the study. the consequences of the covid- pandemic on mental health and implications for clinical practice psychological adjustment during the global outbreak of covid- : a resilience perspective the relevance of covid- pandemic to psychiatry update: use of quarantine to prevent transmission of severe acute respiratory syndrome-taiwan the term "physical distancing" is recommended rather than "social distancing" during the covid- pandemic for reducing feelings of rejection among people with mental health problems case-fatality rate and characteristics of patients dying in relation to covid- in italy risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease pneumonia in wuhan, china redfeld rr covid- -navigating the uncharted principles of mental health care during the covid- pandemic the impact of the covid- outbreak on the medico-legal and human rights of psychiatric patients traumatic stress in the age of covid- : a call to close critical gaps and adapt to new realities covid- pandemic and mental health consequences: systematic review of the current evidence unprecedented disruption of lives and work: health, distress and life satisfaction of working adults in china one month into the covid- outbreak do psychiatric patients experience more psychiatric symptoms during covid- pandemic and lockdown? a case-control study with service and research implications for immunopsychiatry mental health and the covid- pandemic addressing mental health needs: an integral part of covid- response are we all in this together? longitudinal assessment of cumulative adversities by socioeconomic position in the first weeks of lockdown in the uk impaired interferon signature in severe covid- psychiatry in the age of covid- the psychological impact of quarantine and how to reduce it: rapid review of the evidence prevention of child maltreatment: strategic targeting of a curvilinear relationship between adversity and psychiatric impairment mental health consequences for survivors of the fukushima nuclear disaster: a systematic review tsunami: understanding mental health consequences and the unprecedented response mental health and politics since the eurozone crisis: the role of mental health professionals / -related ptsd among highly exposed populations: a systematic review years after the attack health outcomes during the financial crisis in europe: systematic literature review the nascent empirical literature on psychopathology and terrorism mental health of children living in war zones: a risk and protection perspective international consortium to predict ptsd. estimating the risk of ptsd in recent trauma survivors: results of the international consortium to predict ptsd (icpp) who are the european youths willing to engage in radicalisation? a multidisciplinary review of their psychological and social profiles income inequality and depression: a systematic review and meta-analysis of the association and a scoping review of mechanisms sars epidemic unmasks age-old quarantine conundrum control measures for severe acute respiratory syndrome (sars) in taiwan the covid- infection in italy: a statistical study of an abnormally severe disease the importance of patients' case-mix for the correct interpretation of the hospital fatality rate in covid- disease mortality impacts of the coronavirus disease (covid- ) outbreak by sex and age: rapid mortality surveillance system affective temperament, attachment style, and the psychological impact of the covid- outbreak: an early report on the italian general population mental health during and after the covid- emergency in italy covid- outbreak in italy: are we ready for the psychosocial and the economic crisis? baseline findings from the psycovid study the impact of quarantine and physical distancing following covid- on mental health: study protocol of a multicentric italian population trial scaling up psychological treatments for common mental disorders: a call to action the efficacy of app-supported smartphone interventions for mental health problems: a meta-analysis of randomized controlled trials telemental health: why the revolution has not arrived three questions to consider before developing a mental health app bridging the dichotomy of actual versus aspirational digital health the "online brain": how the internet may be changing our cognition towards a consensus around standards for smartphone apps and digital mental health digital interventions in severe mental health problems: lessons from the actissist development and trial manual for the depression anxiety & stress scales mental health strategies to combat the psychological impact of covid- beyond paranoia and panic immediate and sustained psychological impact of an emerging infectious disease outbreak on health care workers immediate psychological responses and associated factors during the initial stage of the coronavirus disease (covid- ) epidemic among the general population in china the validity of two versions of the ghq in the who study of mental illness in general health care the obsessive-compulsive inventory: development and validation of a short version the insomnia severity index: psychometric indicators to detect insomnia cases and evaluate treatment response the suicidal ideation attributes scale (sidas): community-based validation study of a new scale for the measurement of suicidal ideation american psychiatric association. severity of acute stress symptomsadult (national stressful events survey acute stress disorder short scale brief measure of posttraumatic stress reactions: impact of event scale- a short-form measure of loneliness you want to measure coping but your protocol' too long: consider the brief cope a short form of the posttraumatic growth inventory development of a new resilience scale: the connor-davidson resilience scale (cd-risc) the multidimensional scale of perceived social support maslach burnout inventory the propensity score an evaluation of inverse probability weighting using the propensity score for baseline covariate adjustment in smaller population randomised controlled trials with a continuous outcome multiple imputation for nonresponse in surveys mental health burden for the public affected by the covid- outbreak in china: who will be the high-risk group? anxiety and depression among general population in china at the peak of the covid- epidemic rethinking progress and challenges of mental health care in china a nationwide survey of psychological distress among italian people during the covid- pandemic: immediate psychological responses and associated factors hypochondriacal concerns and somatization in panic disorder effects of the covid- pandemic and nationwide lockdown on trust, attitudes toward government, and well-being mental health outcomes of quarantine and isolation for infection prevention: a systematic umbrella review of the global evidence disaster victims speak: part ii. summary and implications of the disaster mental health research weighing the costs of disaster: consequences, risks, and resilience in individuals, families, and communities the risk perceptions of sars and sociopsychological behaviors of urban people in china the psychological burden experienced by hong kong midlife women during the sars epidemic stress and psychological impact on sars patients during the outbreak years of life lost due to the psychosocial consequences of covid- mitigation strategies based on swiss data psychological distress and sleep problems when people are under interpersonal isolation during an epidemic: a nationwide multicenter cross-sectional study dysfunctional personality features, non-scientifically supported causal beliefs, and emotional problems during the first month of the covid- pandemic in italy the psychological impact of preexisting mental and physical health conditions during the covid- pandemic gender differences in depression, anxiety, and stress among college students: a longitudinal study from china gender differences in anxiety disorders: prevalence, course of illness, comorbidity and burden of illness gender differences in major depressive disorder: results from the netherlands study of depression and anxiety the global prevalence of common mental disorders: a systematic review and metaanalysis - mental health problems and social media exposure during covid- outbreak impact of the covid- pandemic on patients with pre-existing anxiety disorders attending secondary care mental health policy in the era of covid- a global needs assessment in times of a global crisis: world psychiatry response to the covid- pandemic how might the nhs protect the mental health of health-care workers after the covid- crisis? global mental health: how are we doing? mental health first aid training: lessons learned from the global spread of a community education program covid- -related web search behaviors and infodemic attitudes in italy: infodemiological study prevention of depression will only succeed when it is structurally embedded and targets big determinants prevalence of suicide ideation, plans, and attempts and related risk factors in italy: results from the european study on the epidemiology of mental disorders--world mental health study the effect of minority status and social context on the development of depression and anxiety: a longitudinal study of puerto rican descent youth how will covid- affect the health care economy? preventing suicide in the context of the covid- pandemic suicide prevention: towards integrative, innovative and individualized brief contact interventions the acute effects of the covid- pandemic on insomnia and psychological symptoms escalation of sleep disturbances amid the covid- pandemic: a crosssectional international study the assessment and management of insomnia: an update insomnia and inflammation: a two hit model of depression risk and prevention the family in italy: cultural changes and implications for treatment intergenerational transmission of trauma effects: putative role of epigenetic mechanisms social research . : virtual snowball sampling method using facebook says who? the significance of sampling in mental health surveys during covid- key: cord- - kmvyovm authors: martinotti, giovanni; alessi, maria chiara; di natale, chiara; sociali, antonella; ceci, franca; lucidi, lorenza; picutti, elena; di carlo, francesco; corbo, mariangela; vellante, federica; fiori, federica; tourjansky, gaia; catalano, gabriella; carenti, maria luisa; incerti, chiara concetta; bartoletti, luigi; barlati, stefano; romeo, vincenzo maria; verrastro, valeria; de giorgio, fabio; valchera, alessandro; sepede, gianna; casella, pietro; pettorruso, mauro; di giannantonio, massimo title: psychopathological burden and quality of life in substance users during the covid- lockdown period in italy date: - - journal: front psychiatry doi: . /fpsyt. . sha: doc_id: cord_uid: kmvyovm background: following the development of the covid- pandemic, a rigid public health strategy of reduced social contact and shelter-in-place has been adopted by the italian government to reduce the spread of the virus. in this paper, we aim at evaluating the impact that the covid- pandemic, and the relative containment measures, have had on a real-life sample of patients suffering from substance use disorders (suds) and/or behavioral addictions. methods: an anonymous questionnaire was filled out by addicted patients, both outpatients and residential inpatients, recruited across italy and highly representative of the current italian population suffering from addictions. psychopathological burden (anxiety and depressive symptomatology, somatization, irritability, and post-traumatic symptoms), quality of life, and craving changes in daily habits were assessed. results: in our sample, we found moderate rates of depression ( . %), anxiety ( . %), irritability ( . %), and post-traumatic stress ( . %) symptoms. psychopathological burden was globally higher among residential patients. reported levels of craving were generally low. discussion: this study is the first attempt to collect italian data regarding the effects of the rigid quarantine period, during the covid- pandemic, on patients suffering from a sud and/or behavioral addictions. the presence of a moderate psychopathological burden correlated to poor quality of life and low craving scores represented the main outcomes. long-term studies, with follow-up after the end of the restrictive measures, should be considered to implement our findings. following the advent of the covid- pandemic, which developed between march , and may , , social containment measures were implemented across italy through a series of consecutive ministerial decrees aimed at limiting the spreading of the virus. the lockdown soon proved effective for such purposes, but at the same time, it generated an important series of consequences from both a social and an economic point of view. social distancing, emotional isolation, complete transformation of the daily routine, abrupt adoption of an unhealthy lifestyle (sedentary lifestyle and unbalanced nutrition), and economic difficulties resulting from the interruption of work activities have thus compromised, and could continue to do so, the well-being of each individual and the entire community ( ) . within the general population, problems such as feelings of frustration, aggressive behavior ( ) , post-traumatic stress symptoms (ptss), depression, anxiety, insomnia, perceived stress, and adjustment disorder symptoms (ads) have increased ( ) , with the consequent risk of self-medication through the abuse of alcohol and/or psychoactive substances and with a greater tendency to engage in pathological behaviors (gambling and internet addiction). it is possible that, among patients with pre-existing mental disorders, the symptomatology may flare up or worsen (with important management difficulties for the caregivers); the risk to develop suicidal ideation is also plausible for the most critical cases ( , , ) . the aforementioned effects in terms of mental health can be superimposed on those observed during other major epidemics/pandemics that have occurred in former times. ebola ( ) , human h n avian flu ( ) , middle east respiratory syndrome (mers) ( , ) , and severe acute respiratory syndrome (sars) ( ) ( ) ( ) ( ) have in fact caused a real "mental health catastrophe" ( ) among the affected population, above all amid the frontline workers managing the health emergency and among those who have recovered from the infection, including their relatives. in this context, people with pathological dependencies on psychoactive substances and/or with behavioral addictions are particularly vulnerable. there is a real "collision" between suds and the covid- infection. moreover, drug users exposed to social risk factors, such as belonging to under-privileged social classes or, even worse, being homeless or imprisoned, are more often subject to precarious hygiene and health conditions. they are particularly susceptible to contract the infection, and, by virtue of obstructive and cardiovascular comorbidities of the ischemichypertensive type, they are prone to develop the disease in its most serious forms ( , ) . in patients with alcohol use disorder, the effects of the lockdown are notpredictable: social isolation, restricted freedom, and the resulting difficulties in obtaining the substance could lead to a reduction in the dysfunctional behavior. nonetheless, an increase in withdrawal symptoms, and the possible use of diy alcohol products, might have significant health fallouts and, potentially, even lead to death ( , ) . among active users, a scarce availability of drugs, hence a reduction in their usage, could lead to withdrawal symptoms that are difficult to manage at home ( ) . patients who are recovering from substance use experience psychological discomfort from social isolation, which might increase the risk of relapse. this alarming scenario is exacerbated by a quantitative and qualitative reduction in the addiction services' assistance and in the stretching of their services ( ) : for instance, recovering patients' access to support groups is prevented, and other forms of psychosocial assistance are limited as well ( ) . the handling of the substitution therapies for opiates addiction, in particular methadone and buprenorphine, has proven to be particularly complex, with difficulties in both supplying and distributing the aforementioned drugs ( , , , ) . these critical issues, caused by the rigid regulations that still guide the provision of replacement treatments, are similar to those documented in the past, e.g., following the terrorist attacks of september , on the twin towers and following hurricane katrina and hurricane sandy that hit the united states, respectively, in and ( ) . this implies a greater tendency to resort to illicit trafficking of opiates whenever the replacement drug cannot be found and increases the risk of death from possible overdose of the replacement drug, every so often dispensed to the patient in doses that are suitable to cover a greater period of time ( ) . therefore, it is evident that the covid- health emergency crisis collides with another important public health emergency, which is that of suds ( ) . the aim of this study was to evaluate the impact that the covid- pandemic, and the relative containment measures adopted by the italian government, had on patients with suds and/or behavioral addictions; to assess the psychopathological burden in terms of depression, anxiety, post-traumatic load; and to evaluate the relevance of craving symptoms and their correlation with psychiatric symptoms and quality of life. from march , to may , , throughout the whole italian lockdown phase, we carried out a survey meant only for adult people with an ongoing and/or previous sud and/ or gambling. disorder (dsm- ) currently in treatment as outpatients and/ or in a residency program as inpatients. two hundred twentyseven patients were recruited and offered the possibility to fill out the questionnaire. one hundred fifty-three patients gave their consent and completed the questionnaire. the survey was conducted in two ways: through a self-administered paper questionnaire and through an online platform where the subjects filled out the questionnaire independently using an url (uniform resource locator) provided by the clinician during an interview. the survey was completed by each subject anonymously only after having read the information sheet and having signed the informed consent form. various centers for recruitment were randomly selected in different regions of italy (abruzzi, calabria, lazio, piedmont, marche, lombardy, and molise) in order to guarantee an equal distribution of the sample's population around the country. in each recruitment center, a psychiatrist gave the survey to all eligible subjects. the presence of a dsm- diagnosis of sud had been assessed and confirmed before the study procedures, representing an inclusion criterion of the study. the survey was organized in three sections. in the first section, we collected anamnestic information and clinical variables that included age, gender, education level, relationship status, days spent in lockdown, primary substance of abuse, substitute and/or support treatments, pathological gambling, support by addictions services, comorbid psychiatric disorders and psychopharmacological treatment, hospitalization, and sars-cov- testing. in the second section, we asked the subjects to indicate the level of craving for the primary substance of abuse and how much their craving and habits have changed since the start of lockdown. we used a visual analogue scale (vas), which ranged from (strongly reduced) to (strongly increased). we investigated the change in quality of life, the consumption of cigarettes, coffee, alcoholic drinks, cannabis, cocaine, opioids, benzodiazepines, food, and the time spent shopping online, instant messaging, and making video calls with friends/relatives on social networks, carrying out old and/or new hobbies, in sport activities, watching tv series or films, and watching pornographic material. in the third section, we investigated the psychopathological variables of interest, from the start of the lockdown to the completion of the survey. irritability was measured using four irritability items from the irritability depression anxiety scale (idas) ( ); five items from the self-rating anxiety state (sas) were employed to investigate anxiety ( ) . somatic symptoms were investigated with a single question about the presence of all possible pathological conditions. the davidson trauma scale (dts) was adopted for the assessment of post-traumatic stress symptoms ( ) , and the beckdepression inventory -ii (bdi-ii) ( ) was utilized to assess current depressive symptoms. according to the scores obtained in the scales, symptomatology was divided into two categories: minimal/ mild and moderate/severe. statistical analysis was performed using statistica . (statsoft inc. usa, ) . quantitative parameters were presented as mean ± standard deviation (sd) and qualitative parameters as number and percentage per class. kolmogorov-smirnov (k-s test) was used to check for the normality of distributions. analysis of variance (anova) and duncan post hoc test were utilized to evaluate the differences among subgroups' means. the associations between variables were measured using pearson's correlation. the p value was considered significant if < . . most patients were males (n = , . %); the mean age was . (± . ) years. at the time of questionnaire completion, the subjects had been in quarantine for an average of . (± . ) days. most subjects (n = , . %) indicated cocaine as the principal substance of abuse, followed by alcohol (n = , . %) and thc (n = , . %). of the entire sample, patients ( . %) were outpatients, living at home during quarantine, while ( . %) were inpatients in residential programs. the full participants' characteristics and the substances' patterns of use are presented in table . sixty-seven ( . %) participants reported a comorbid psychiatric condition, especially mood disorders (depression and bipolar disorder) or anxiety. sixty-three ( %) of those with comorbid psychiatric condition and ( . %) of those without a comorbid psychiatric disorder reported undergoing psychopharmacological treatment. all the information regarding the comorbid psychiatric conditions and their pharmacological treatments remained unchanged. about % of the patients reported a comorbid medical condition. only one subject ( . %) had a covid- related pneumonia ( table ) . we calculated the total score for five psychometric scales (idasirritability, dts, sas-five items, somatization, and bdi-ii) in both the entire sample and in five of the principal categories of substances/behaviors (alcohol, cocaine, gambling, thc, and heroin). anova showed no significant effect on the principal substance of abuse ( table ) . each psychopathological domain was scored into two levels of severity: minimal/mild and moderate/severe. scores are detailed in table . the mean level of craving was generally low ( . ), nonetheless a general low difficulty in finding the substances of abuse was reported. the level of craving was higher in outpatients (mean = . ) compared to inpatients (mean = . , p = . ) ( table ). the association between the level of craving for the principal substance of abuse and the values of the psychometric scales was measured using pearson's correlation. these data about craving will be further elaborated elsewhere. the level of significance (p = . ) was corrected for multiple comparisons using the bonferroni correction: p corr = . /n comparisons = . . we observed a significant positive correlation between the level of craving and the mean total values of dts, sas, (five items) and bdi-ii, and the results remained significant after bonferroni correction ( table ) . when comparing inpatients versus outpatients by means of anova, the idas (irritability) scale resulted in significantly higher levels among inpatients. comparing dual diagnosis participants against non-dual diagnosis participants, bdi-ii, dts, and somatization scores were significantly higher among dual-diagnosis patients. vas quality of life scored higher in the non-dual diagnosis group. results of anova tests are detailed in table . we found an increase of about % of the cases for the amount of time spent on the following daily activities: eating, instant messaging, social networking, video calls to friends/ relatives, watching movies/tv shows, and sleeping. about % of subjects increased their online search to gather information about the ongoing pandemic. this study collects the first italian data regarding patients suffering from suds and/or behavioral addictions during the rigid quarantine period caused by the covid- pandemic. the study, which includes patients recruited in seven different representative italian regions, has the uniqueness of incorporating previously treated patients who were known by local services and who were all given a dsm- diagnosis of sud. in addition, the recruited group represents a real-life sample that reflects the italian addiction scenario ( ) and was homogeneously differentiated into residential and non-residential patients, with some patients reporting a dual diagnosis and others none. the psychopathological burden observed in our sample is in line with recent international data concerning psychiatric patients, subjects with dual diagnosis, and drug addicts. the effects of quarantine on mental health have been highlighted in a recent review that evaluates the psychological distress among the quarantined people during past pandemics and epidemics ( ) . many studies, based on online surveys, have shown an increase in anxiety, depression, and stress among chinese ( ) ( ) ( ) ( ) , italian ( , ) , and spanish ( ) people due to the covid- pandemic. our results are in line with these findings, showing relatively high rates of depression, anxiety, irritability, and post-traumatic stress symptoms among the sample. specifically, . % of our sample reported moderate/severe depressive symptoms, and . % reported moderate/severe anxiety symptoms, similar to what was indicated by another italian survey that rated . % of participants as having high/very high depressive symptoms and . % of them as having high/very high anxiety symptoms ( ) . these results show no substantial psychopathological difference between our sample and the general population. mazza et al. reported a considerable increase in the use of telephones, social networks, and mobile apps toconnect with family and friends during the quarantine period among the italian population. our findings are in line with these results, showing an increase in the use of instant messaging ( . %) and video calls ( . %) to connect with friends and relatives among substance users as well. moreover, we found an increase in the time spent utilizing social networks ( . %), collecting online information about the current situation ( . %), and watching movies or tv shows ( . %). in our study, the level of craving resulted to be overall, lower than real-life samples of italian patients with suds ( ) . craving is one of the key symptoms in addicted patients, closely correlated with the prognosis and progression of the pathology ( ) and lower levels could influence positively the treatment outcome ( ) . this unexpected result could be explained by a perceived lack of availability of the substance that interrupted the development of the craving priming and by the presence of decreased social pressure on a group of subjects that are usually excluded and stigmatized. specific craving variations between the lockdown-period and prior times will be reported and discussed elsewhere. craving was higher among outpatients than inpatients. this data underlines the importance of residential treatment in suds. in fact, numerous studies demonstrate the effectiveness of this approach in increasing the perceived quality of life and in improving executive functions and psychological distress ( , , ) , conditions that lead to a reduction in craving ( ) . such a notion is relevant because substance craving is a known predictor of relapse after treatment for suds ( ) . residential treatment could, therefore, be a fundamental first step in laying the foundations for subsequent long-term outpatient treatment. this is even more true if we take into account that it also causes a change in the perception that the drug addict has of himself, transitioning from a 'substance user' social identity to an 'in-recovery' identity ( ) . in terms of craving intensity, the benefits of the presence of strict limitations on personal freedom, including the impediment to obtain substances, combined with the benefits of carrying out intensive treatment in residential structures, are perhaps the most interesting result of our study and it has relevant therapeutic implications. moreover, our results underline the link between craving and quality of life, defined as the perception that the individual has regarding the effects that a disease, and its treatment, have on his physical, emotional, and social well-being ( ) . more than half of the cohort reported reduced quality of life during covid- lockdown, and the analysis showed a negative correlation between perceived quality of life and reported craving. the association between alcohol craving and quality of life was previously studied by herrold et al. in war veterans demonstrating that high levels of craving were associated with poor perceived quality of life, bothmentally and physically ( ) . at the same time, improving the quality of life, for instance, through physical exercise, can play an important role in reducing craving and, therefore, conducts of abuse ( ) . several studies have demonstrated that stress, negative mood, and craving could expose addicted patients to relapse and dropout from treatment ( ) . these factors are important elements of vulnerability that can be correlated with each other. it is essential to recognize and treat each one of them to improve the outcome. in fact, in our analysis we found a positive correlation between craving and depressive symptoms, anxiety, and traumatic stress. these findings are in line with the study of fatseas et al. that found an association between psychiatric distress, mood and/or anxiety disorders, and higher levels of craving ( ) . moreover, luminet et al. found strong correlations between negative affect and craving in alcohol-dependent patients. in their study, an increase in depressive symptoms was related to increased levels of craving in women ( ) . it is necessary to look for the association between craving and psychopathological conditions because it could present useful information for a successful treatment. specific attention to these clinical parameters could be the basis for a specific strategy to be employed in those populations exposed to the pandemic and to its associated restrictions and could open new scenarios based on possible preventive interventions. in lockdown period, the role of telepsychiatry acquires great importance for careful monitoring of the patient's clinical and psychopathological conditions in order to prevent relapses ( ) . through telematic interview, the clinician can also supervise the patient's family environment, trying to understand if it provides the patient with enough support. this study has some limitations: ) the absence of a long-term follow-up, potentially useful to highlight the consequences of the lockdown; ) in a part of the sample, the survey was completed online directly by the patient without proper verification by the clinician; and ) the assessment of craving, which has always been complex and sometimes difficult to interpret, was carried out with a visual analogue scaling and not with more structured scales. long-term studies, with follow-up at the end of the restrictive measures and after the full development of the psychopathological experience caused by the pandemic and by its socio-economic consequences, may clarify the true impact of the covid- pandemic on those subjects affected by suds. meanwhile, thanks to this study being conducted with a sample of italian drug addicts, it was possible to identify a moderate psychopathological burden correlated with poor quality of life and craving scores. the latter were overall low, especially among patients who are hospitalized in residential structures, opening interesting questions in terms of treatment strategies. the raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. the patients/participants provided their written informed consent to participate in this study. gm mental health and the covid- pandemic danger in danger: interpersonal violence during covid- quarantine covid- pandemic and lockdown measures impact on mental health among the general population in italy. an n = web-based survey lifestyle behaviours during the covid- -time to connect covid- pandemic and addiction: current problems and future concerns post-ebola psychosocial experiences and coping mechanisms among ebola survivors: a systematic review the negative psychology for the public in zhejiang province during the epidemic of human h n avian influenza mental health status of people isolated due to middle east respiratory syndrome mental health of nurses working at a government-designated hospital during a mers-cov outbreak: a crosssectional study stress and psychological distress among sars survivors year after the outbreak mental symptoms in different health professionals during the sars attack: a follow-up study was sars a mental health catastrophe? alcohol abuse/ dependence symptoms among hospital employees exposed to a sars outbreak collision of the covid- and addiction epidemics what about heart and mind in the covid- era? minerva cardioangiol covid- outbreak: challenges for addiction services in india mitigating and learning from the impact of covid- infection on addictive disorders editorial: challenges to opioid use disorders during covid- opioid use disorder and the covid pandemic: a call to sustain regulatory easements and further expand access to treatment a clinical scale for the self-assessment of irritability a rating instrument for anxiety disorders assessment of a new self-rating scale for post-traumatic stress disorder manual for beck depression inventory-ii validation of a substance craving questionnaire (scq) in italian population the psychological impact of quarantine and how to reduce it: rapid review of the evidence epidemic of covid- in china and associated psychological problems comparison of prevalence and associated factors of anxiety and depression among people affected by versus people unaffected by quarantine during the covid- epidemic in southwestern china mental health, risk factors, and social media use during the covid- epidemic and cordon sanitaire among the community and health professionals in wuhan, china: cross-sectional survey a longitudinal study on the mental health of general population during the covid- epidemic in china a nationwide survey of psychological distress among italian people during the covid- pandemic: immediate psychological responses and associated factors mental health consequences during the initial stage of the coronavirus pandemic (covid- ) in spain transcranial direct current stimulation reduces craving in substance use disorders: a double-blind, placebo-controlled study alcohol urges in alcohol-dependent drinkers: further validation of the alcohol urge questionnaire in an untreated community clinical population craving and other transdiagnostic dimensions in addiction: toward personalized neuromodulation treatments psychological stress, drug-related cues and cocaine craving ecological momentary assessment of the effects of craving and affect on risk for relapse during substance abuse treatment breaking good: breaking ties with social groups may be good for recovery from substance misuse medicines and risk/benefit decisions association between alcohol craving and health-related quality of life among veterans with cooccurring conditions effects of exercise on depression, anxiety, cognitive control, craving, physical fitness and quality of life in methamphetamine-dependent patients stress, craving and mood as predictors of early dropout from opioid agonist therapy effects of anxiety and mood disorders on craving and substance use among patients with substance use disorder: an ecological momentary assessment study the association between depression and craving in alcohol dependency is moderated by gender and by alexithymia factors telepsychiatry and other cutting edge technologies in covid- pandemic: bridging the distance in mental health assistance this study was partly found by the european project entitled 'analysis, knowledge dissemination, justice implementation and special testing of novel synthetic opioids' -just- -ag-drug. conflict of interest: gm has been a consultant and/or a speaker and/or has received research grants from angelini, doc generici, janssen, lundbeck, otsuka, and pfizer. mg has been a consultant and/or a speaker and/or has received research grants from angelini, janssen, lundbeck, otsuka, pfizer, and recordati.the remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.the reviewer, jc, declared a shared affiliation, though no collaboration, with one of the authors, gm, to the handling editor.copyright © martinotti, alessi, di natale, sociali, ceci, lucidi, picutti, di carlo, corbo, vellante, fiori, tourjansky, catalano, carenti, incerti, bartoletti, barlati, romeo, verrastro, de giorgio, valchera, sepede, casella, pettorruso and di giannantonio. this is an open-access article distributed under the terms of the creative commons attribution license (cc by). the use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. no use, distribution or reproduction is permitted which does not comply with these terms. key: cord- -vtuy l authors: cascio, antonio; colomba, claudia; di carlo, paola; serra, nicola; lo re, giuseppe; gambino, angelo; lo casto, antonio; guglielmi, giuseppe; veronese, nicola; lagalla, roberto; sergi, consolato title: low bone mineral density in hiv-positive young italians and migrants date: - - journal: plos one doi: . /journal.pone. sha: doc_id: cord_uid: vtuy l background: human immunodeficiency virus (hiv) infected individuals may have osteoporosis. we aimed to evaluate the bone mineral density (bmd) in naïve antiretroviral (arv) treated hiv positive patients comparing native italian group (itg) to a migrants group (mig) upon arrival in italy. methods: we conducted a cross-sectional study on hiv patients less than years old. we used the dual-energy x-ray absorptiometry (dxa) within six months from the hiv diagnosis. participants were categorized as having low bmd if the femoral neck or total lumbar spine z-score was– or less. results: mig showed low bmd more often than itg ( . % vs. . %), especially for the female gender ( . % vs. . %). a low cd rate (< cells/μl) was most often detected in mig than itg. in particular, we found most often male italians with abnormal cd than male migrants ( . % vs. . %) and vice versa for females ( . % vs. . %). we found an abnormal bone mineral density at the lumbar site. low bmd at the lumbar site was more frequently observed in female migrants than female italians. both male and female migrants had a z-score value significantly lower than male and female italians, respectively. by logistic regression low vitamin-d level was positively correlated to low bmd in itg only. all data were verified and validated using a triple code identifier. conclusions: both dxa and vitamin-d evaluation should be offered after the diagnosis of hiv infection. lumbar site low bmd is an initial condition of bone loss in hiv young patients, especially in female migrants. vitamin d levels and supplementation may be considered after hiv diagnosis independently of age to improve bone health. highlights: this study evaluates the frequency of bone mineral density in hiv positive patients naive to antiretroviral therapy. it compares the density of the native italian population with that of hiv migrants upon arrival in italy. the results show that hiv positive migrants, even if younger than years of age, are at risk for osteoporosis, especially if they are female. a a a a a osteoporosis is a process characterized by increased bone resorption without reciprocal bone apposition. the decreased bone mass leads to an increased risk of bone fractures, which are particularly as the subject ages [ , ] . currently, lifestyle factors, including alcohol, diet, hormones, physical activity, and smoking, are influencing the bone mass, and osteoporosis is affecting younger people than before [ ] [ ] [ ] . the incidence of osteoporosis and osteoporosisrelated fractures varies across the european union/european economic area (eu/eea), and especially in the mediterranean area, it is increasing, as it has been evidenced in the literature with a clear link to nutrition [ , ] . among human immunodeficiency virus (hiv)-related comorbidities, the bone disease has emerged as a significant question in this chronically infected population [ ] [ ] [ ] [ ] [ ] [ ] . several studies suggest that fracture rates are higher in communities with hiv than among matched uninfected controls despite anti-retroviral therapy (art) and gender [ ] [ ] [ ] . the causes of a low bone mineral density (bmd) could be different in hiv-infected patients by considering the time of hiv infection [ ] . previous studies conducted in our geographical area established that the prevalence of osteoporosis was significantly higher in hiv-infected than in uninfected subjects, which mirrors a result similar to previous meta-analyses [ ] [ ] [ ] [ ] . the prevalence of osteopenia and osteoporosis in hiv mono-infected patients in our geographical area was about . %, and . % in comparison with % reported in a healthy italian population [ ] . a recent analysis of bone-healthy in the immigrant population conducted in sweden showed that women had low bmd for age according to the american and african-american referents [ ] [ ] [ ] . moreover, a study conducted in german-turkish immigrants osteopenia was diagnosed in % and osteoporosis in % of young migrants [ ] . in the migrant setting, it has been hypothesized that bmd is intriguingly associated with lifestyle, body mass index (bmi), and vitamin d levels. recently, low level of vitamin d were also reported in somali migrant women in sweden and refugees in canada using the calgary refugee health program, an urban family practice that serves newly arrived refugees in calgary, alberta [ ] [ ] [ ] . although the number of migrants and refugees crossing the mediterranean sea has decreased in - and , during the covid- pandemic, this number has been unprecedented in the last ten years compared with the past twenty or thirty years [ ] [ ] [ ] . topical italian data showed that the new diagnosis of hiv infection in % of migrants was late and less than six months before developing aids with half of the subjects coming from sub-saharan africa (ssa, . %) and with an increasing percentage of fertile females [ ] [ ] [ ] [ ] [ ] . further, the incidence of bmd loss and related fractures attributed to a specific eating pattern has also been targeted for guidelines and suggestions within the eu/eea [ ] . our study aims to emphasize the burden of bone health in naïve arv hiv positive patients and compare the bone density of the native italian population group (itg) with that of hiv migrants (mig) upon arrival in italy. this investigation is a retrospective cross-sectional study. we gathered data from arv naïve subjects consecutively admitted between january and may at the sicilian aids center of the university of palermo, italy. we retrospectively analyzed all patients who underwent dual-energy x-ray absorptiometry (dxa) within six months from hiv diagnosis independent of their bmi. in this study, we excluded all patients with the treatment of steroidinduced bone loss and subjects with active tuberculosis. dxa measures bmd utilizing two x-ray beams with different energy levels, which are aimed at the patient's bones (see below). of the subjects, were italian, and were migrant patients living in italy for less than twelve months. table shows the characteristics of enrolled patients. we collected the medical records and entered into an anonymous database, as previously reported [ , ] . cd + t-cell count and plasma hiv-rna levels were assessed as previously reported [ ] . -hydroxy-vitamin d ( (oh)d) was assayed as previously reported [ , , ] . we considered a serum (oh)d concentration of ng/ml as a threshold value for identifying low levels of vitamin d, as previously reported (standard value of our laboratory: - ng/ml) [ ] . all subjects gave their informed consent for inclusion before they started participating in the study, which was conducted by the declaration of helsinki. the ethics committee of the university of palermo approved the study protocol. each participant was free to decline or withdraw from the study at any time, and this was explained in their original language or a language that the migrant knew. bmd was assessed by dxa, using a qdr discovery hologic dxa in the femoral neck and dxa in the lumbar spine by total body dxa. for each scan, bmd, z-scores were recorded as previously reported [ ] [ ] [ ] . dxa measurements were performed in the femur (femoral neck and/or total hip) and lumbar spine in each patient. since the age of our patients ranged from to years, the use of z-scores (defined as an individuals' bmd in comparison to agematched normal individuals) was used for all the analyses, according to world health organisation (who) recommendation [ ] . participants were categorized as having low bmd if the femoral neck or total lumbar spine z-score was- or less. to individualize sample sizes statistically significant in this study, we considered a bernoulli sampling for both italian and migrant groups [ ] . for itg, the minimum sample size for this study was estimated equal to patients affected by low bmd. it was obtained considering a statistical z-score at %, an error ε = % and hypothesizing a prevalence π, about % according to the studies of cavalli et al. [ ] and tomažič et al. [ ] . in this case, we estimated a prevalence range equal to %- %. for mig, the minimum sample size for this study was estimated equal to patients affected by low bmd. it was obtained considering a z-score at %, an error ε = %, and hypothesizing a prevalence π about %, according to varenna et al. [ ] . in this way we estimated a prevalence range equal to %- %. in this case, we considered an error ε greater to an error in itg because of the amount of information that could be gathered in migrants in italy. also, the sample size for itg and mig were increased to and patients, respectively considering the possibility of unexpected events and, consequently, the likelihood of patients' data loss. the statistical analysis was performed by matlab statistical toolbox version (math-works, natick, ma, usa). data are presented as number and percentage for categorical variables. numerical data are expressed as the mean ± standard deviation (sd) or median and confidence interval at % (ci). the χ test and fisher's exact tests were performed to evaluate significant differences of proportions or percentages between two groups. the fisher's exact test was used where the χ test was not appropriate. the mann-whitney test was used to test the difference between two independent samples (itg and mig group). it was the alternative for a t-test dealing with independent samples when the distribution of the samples is not normal. normal distributed independent samples were studied using the d'agostino-pearson test. linear correlation analysis was also performed, and the test on pearson's linear correlation coefficient r was performed with the t-student test under the null hypothesis of pearson's linear correlation coefficient r = . the logistic regression was performed to analyze the relationship between low bmd (dichotomous variable) and the independent variables: gender (dichotomous-m = , f = ), previous fractures (dichotomous-yes = , no = ), bmi (continuous), cd cells (continuous), and -hydroxy-vitamin d (continuous). finally, all tests with p-value < . were considered significant. the patients, composed by . % males and . % females, with ages into range - , mean . years old and standard deviation (s.d.) equal to . years old, was subdivided into two groups (table ) in table , we report the differences between itg and mig group and statistical tests. low bmd was significantly more frequent in mig group in comparison to itg group ( . % > . %, p = . ), particularly it was significantly more frequently found in migrant females than in italian females ( . % > . %, p = . ). significant differences were seen in low-bmd values. in fact, both female and male italians had a measure of bmd significantly greater than migrant females and males respectively (median: - . > - . , p = . ; . > - . , p = . ). in addition, we found major presence of migrant females with abnormal lumbar value in comparison to italian females ( . % > . %, p = . ). conversely, no significant differences between itg and mig were found considering the femoral bmd values. in examining the gender in mono-infected hiv patients, the percentages of hiv patients were significantly greater in italian males in comparison to migrant males ( . % > %, p = . ). on the other hand, hiv was most frequently observed in female migrants in comparison to female italians ( . % > . %, p < . ). for co-infected patients, there were no significant differences in gender between itg and mig. also, for aids patients there was a significant difference between itg and mig ( . % < . %, p = . ). in particular, aids was most frequent in female migrants in comparison to female italians ( . %> . %, p< . ). we found a higher percentage of male italians with abnormal values of cd in comparison to male migrants ( . % > . %, p = . ), and vice versa for the female gender ( in migrant patients, -oh-vitamin d mean values were significantly lower than in italian patients ( . > . ng/ml, p = . ). also, male italians had mean value significantly figs and , the heat maps of lbd and fbd values for itg and mig groups are shown, respectively. every line represents the fbd and lbd values for each patient, and the graduation of the colors are representative of the z-score. in fig , we showed the scatter plots for itg and mig group between cd values with lbd and fbd (a-d). for every scatters plot, the red points on graphs are individuated by cd values and correspondent lumbar bmd and femoral bmd z-scores. the blue line is the best linear fit, and the red lines define the % prediction interval for the regression curve. for any given value of the independent variable (cd ), this interval represents the % probability for the values of the dependent variable (lbd or fbd). by linear correlation analysis, it results that for the itg group, there was a significant positive linear correlation between cd and femoral bmd (r = . , p = . ) and between cd and lumbar bmd (r = . , p = . ). in contrast, the mig group showed no significant correlation between cd and femoral bmd (r = . , p = . ) and between cd and lumbar bmd (r = . , p = . ). in other words, in italian patients, an increase/decrease of cd values implicate an increase/decrease of lumbar bmd or femoral bmd scores. we did not observe these correlations in the mig group. finally, in table , we report the logistic regression analysis between low bmd variable (dichotomous) and the independent variables: gender (dichotomous), bmi (continuous), hydroxy-vitamin d (continuous), cd (continuous), and previous fractures (dichotomous) for the total sample, itg, and mig. for this scope, two models were considered. the null model: - ln(l ), where l was the likelihood of obtaining the observations if the independent variables did not affect the outcome, and the full model: - ln(l ), where l was the likelihood of obtaining the observations with all independent variables incorporated in the model. the difference between these two yields was estimated with the chi-square test to define how well the independent variables affect the outcome or dependent variable. if the chi-square test was positive (p < . ), then there was evidence that at least one of the independent variables contributes to the prediction of the outcome. by logistic regression, it resulted that only hydroxy-vitamin d was negatively correlated to low bmd both in total sample (odds ratio, or = . and p = . ), and in itg (or = . and p = . ). in other words, an increase (decrease) of hydroxy-vitamin d contribute to decreasing (increasing) of low bmd. in other words, a decrease of hydroxy-vitamin d contributes to increasing the probability of osteoporosis or osteopenia in the total sample, and particularly, in the italian patients. in the migrant group, we did not observe significant correlations by regression analysis, and the small sample size of migrants considered in our study may be the reason (see below). our study shows that "low bmd" was significantly more frequent in the mig group in comparison to the itg group (migrants . % vs. . %). in particular, it was significantly more frequently found in migrant females than in italian females. our previous reports [ , ] on the prevalence of low-bmd in hiv mono-infected patients who underwent arv therapy showed higher percentage rates of osteopenia ( . %) and osteoporosis ( . %) than an agerelated healthy italian population ( %) [ ] . our current study showed that the percentage of abnormal bmd was mostly restricted to the lumbar site. the low bmd of the lumbar site in female migrants was more evident than in female italians, and both males and female migrants had a z-score value significantly lower than males and females italians. studies conducted on hiv negative patients have, indeed, shown how osteophyte formation, bone sclerosis, disk space narrowing, and spondylolisthesis are positively correlated with lumbar spine bmd. at the same time, there was no association with femoral neck bmd [ ] . in general, anatomical studies suggest that sex-differences in the lumbar spine appear to be supported by postural differences in sacral bone-orientation and morphological differences in the vertebral bodies of females [ ] . moreover, emerging data show that pregnant and breastfeeding women are associated with early osteoporosis, especially in the thoracolumbar spine [ ] . the involvement of the lumbar site is probably justified by the fact that our sample is younger than in other studies conducted on older populations of hiv negative subjects [ - , , ] . hcv is a risk factor for osteoporosis and fractures in the hiv population. in our studied population, hcv co-infection was found in . % of the enrolled sample. recently, one of the authors identified a more lumbar than femoral "low bmd" in hiv/hcv coinfected patients [ ] . the comparison of aids diagnosis between italian and migrants showed that aids was more frequent in migrants, especially in females. in general, we found more frequently low cd values (< cells/μl) in both male and female migrants in comparison with italian patients. these data are in agreement with the emerging late diagnosis of hiv infection in young italians comparing with age-matched female migrants [ , ] . by linear correlation analysis, an increase/decrease of cd values implicates an increase/decrease of lumbar or femoral bmd scores only in italian patients. in contrast, we did not observe similar correlations in the mig group. this aspect is probably due to the element that all migrants had lower cd values than age-matched italians. indeed, no migrant had values of cd � . although the mean of bmi in migrants was lower than italian patients, the statistical analysis of this variable evidenced that low bmd was significant only in female migrants. on the other hand, the comparison of the percentage of patients with abnormal bmi was significant in male italians. the logistic regression showed a relationship between low hydroxy-vitamin d and low bmd only in the italian group. this data should be validated on a larger sample size. low consumption of dairy products, high consumption of soft drinks, moderate sun exposure, and a high prevalence of vitamin d deficiency have been identified among youth as risk factors for early osteoporosis. the nutrition factor is probably intrinsically associated with the geographical region and climate extremes (droughts and storms) may aggravate the bone mass loss of hiv-infected youth. as reported in other studies, we observed early natural menopause status both in italian than in migrant women. this finding was analogous to both groups. apart from hiv-related immunologic status (cd count and viral load) other socio-demographic variables (marital status, parity, education, or income) and religious clothing of the female migrant sample may have influenced the low bmd observed in our study. to the best of our knowledge, this study is the first dxa-based investigation to study and assess the bmd among the flow of migrants and refugees from countries with a high incidence of hiv infection across the mediterranean area. in this respect, the fact that our study includes only migrants, who were resident in italy for no more than one year, could offer the opportunity to carefully assess the risk factors, which stress the bone composition in the hiv infected population. a recent paper on osteoporosis among hiv positive patients highlighted how an assessment of baseline bone mass loss within the first two years from the start of arv is often lacking in longitudinal studies and claimed that there is some evidence concerning bone loss within the first year of hiv infection and art initiation [ , ] . the findings of a high prevalence of low bmd in hiv infected migrants aged less than years of age could suggest that the first screening for osteoporosis should be carried out as soon as possible and before the start of art. this aspect should help the clinician in arv management. planning hiv therapy to prevent bone loss is crucial. we suggest the use of tenofovir alafenamide (taf) molecule instead of tenofovir to reduce the risk of bone toxicity [ , ] . a recent study enrolled young hiv positive with low bmd showed tenofovir as a key molecule in arv therapy [ ] . taf is a hepatitis b virus (hbv) nucleotide reverse transcriptase inhibitor for the treatment of chronic hbv infection in adults with compensated liver disease and has been approved by the food and drug administration (fda) for the treatment of hiv- in november . the altered bone remodeling in young hiv positive women may suggest some alertness about our current preventive health care services. most of the immigrants diagnosed with hiv are pregnant women who have undergone tests for sexually transmitted diseases. although this study is original and discloses essential steps to improve current management strategies because of new migrations from underdeveloped countries, there are some limitations. our research involves a small size of the groups, but the cross-sectional study design still maintains its validity. this study does not permit us to establish a cause-effect relationship from our results or to evaluate the impact of osteoporosis risk factors and/or lifestyle risk factors on the management of bone disease in the hiv population. thus, a multicentric and prospective study is warranted. in conclusion, our study raises the question of the need to plan preventive strategies in hiv migrants and refugees, also in the light of the great movement of people from southern mediterranean areas to northern european countries with low sun exposure [ , ] . the female migrants must be protected from precocious low bmd with an additional layer of safety [ - , , , ] . the approach requires different and specific intervention strategies in the future. our data confirm that early screening for low bmd and other risk factors associated with bone loss in hiv patients is useful [ , , , [ ] [ ] [ ] . in agreement with the osteoporosis foundation (http://share.iofbonehealth.org/wod/ compendium/ -iof-compendium-of-osteoporosis-press.pdf) the authors stress the strategies for preventing osteoporosis. the role of nutrition in maintaining bone health associated with a personalized anti-hiv treatment and daily vitamin supplementation to reduce early bone loss as well as a baseline dxa are critical steps for hiv migrants and refugees after arrival in the host countries [ , , , , ] . high-risk groups should be first identified, and osteoporosis investigations should be carried out earlier than at years of age [ ] [ ] [ ] . as indicated above, our results should be evaluated with caution due to the retrospective nature of the study. also, despite the estimated minimum sample sizes of itg and mig were correctly applied and the conditions for the sample size according to bernoulli are met, prospective and more extensive studies are needed to confirm our results. assessment of fracture risk and its application to screening for postmenopausal osteoporosis: report of a who study group world health organization dietary patterns, bone mineral density, and risk of fractures: a systematic review and meta-analysis pathological case of the month: classic rickets in a setting of significant psychosocial deprivation english disease": historical notes on rickets, the bone-lung link and child neglect issues strategy and action plan for refugee and migrant health in the who european region. copenhagen, denmark: world health organization nutrition, osteoarthritis and cartilage metabolism endocrinological aspects of hiv infection mapping a method for systematically reviewing the medical literature: a helpful checklist postmortem protocol of human immunodeficiency virus (hiv)-related pathology in childhood decreased apoptosis despite severe cd depletion in the thymus of a human immunodeficiency virus- infected child estimating minimum adult hiv prevalence: a cross-sectional study to assess the characteristics of people living with hiv in italy reduced bone mineral density among hiv infected patients on anti-retroviral therapy in blantyre, malawi: prevalence and associated factors reduced bone mineral density in human immunodeficiency virus-infected individuals: a meta-analysis of its prevalence and risk factors: supplementary presentation dairy calcium intake and lifestyle risk factors for bone loss in hiv-infected and uninfected mediterranean subjects associated factors and liver disease severity for decreased bone mineral density in hiv mono-and hiv/hcv co-infected patients vitamin d levels and il b polymorphisms are related to rapid virological response to standard of care in genotype chronic hepatitis c prevalence of osteoporosis in the italian population and main risk factors: results of bonetour campaign comorbidity and health-related quality of life in somali women living in sweden. scand j prim health care factors associated with bone mineral density in healthy african women site-specific differences in bone mineral density in black and white premenopausal south african women adult lactose intolerance, calcium intake, bone metabolism and bone density in german-turkish immigrants vitamin d status of refugees arriving in canada: findings from the calgary refugee health program high hiv- diversity in immigrants resident in italy impact of an -week exercise and sport intervention on post-traumatic stress disorder symptoms, mental health, and physical fitness among male refugees living in a greek refugee camp behavioral and clinical characteristics of people receiving medical care for hiv infection in an outpatient facility in sicily, italy. patient prefer adherence scientific group on the assessment of osteoporosis at primary health care level estimation of a population total under a "bernoulli sampling" procedure prevalence and risk factors for osteopenia/osteoporosis in an hiv-infected male population prevalence of osteoporosis and fractures in a migrant population from southern to northern italy: a cross-sectional, comparative study insulin/igf- r, sirt , and foxos pathways-an intriguing interaction platform for bone and osteosarcoma morphological and postural sexual dimorphism of the lumbar spine facilitates greater lordosis in females calcium and vitamin d in human health: hype or real? phalangeal quantitative ultrasound: cheaper methods for screening and follow-up of bone pathologies in hiv-infected women? optimizing antiretroviral therapy for women living with hiv planning hiv therapy to prevent future comorbidities: patient years for tenofovir alafenamide osteoporosis risk factors in hiv positive women with osteoporosis: a retrospective analysis vitamin d status and the relationship with bone fragility fractures in hiv-infected patients: a case control study subsequent fracture risk of women with pregnancy and lactation-associated osteoporosis after a median of years of follow-up the authors ac, pdc, and ns, contributed equally to this paper. we are deeply grateful to the patients and their families for participating in the study. preliminary results of this study were presented at the hiv glasgow conference, oct. - , , glasgow, united kingdom. key: cord- -q g uqj authors: commodari, elena; la rosa, valentina lucia title: adolescents in quarantine during covid- pandemic in italy: perceived health risk, beliefs, psychological experiences and expectations for the future date: - - journal: front psychol doi: . /fpsyg. . sha: doc_id: cord_uid: q g uqj since march , many countries throughout the world have been in lockdown in response to the covid- pandemic. in italy, the quarantine began on march , , and containment measures were partially reduced only on may , . the quarantine experience has a significant psychological impact at all ages but can have it above all on adolescents who cannot go to school, play sports, and meet friends. in this scenario, this study aimed to provide a general overview of the perceived risk related to covid- and the psychological experience of quarantine in a large sample of italian adolescents. nine hundred and seventy eight adolescents (males = ; females = ) living in italian regions and attending upper secondary school (age range: – , m = . , sd = . ), responded to an internet-based questionnaire about perceived health risk related to covid- , knowledge and information on measures to control the pandemic, beliefs and opinions on stage two of the quarantine, and psychological experiences related to quarantine. . % of the participants lived in “red zones,” which are places where the government has imposed stricter measures of containment due to exponential and uncontrolled growth in contagion cases compared to other areas in italy. according to our results, italian adolescents had a low perception of risk of covid- . perceived comparative susceptibility and perceived seriousness were also very low. however, they were aware of the restriction measures necessary to contain the spread of the virus, and they agreed with the limitations imposed by the government. females and adolescents living in a “red zone” showed more significant psychological negative feelings about the quarantine experience. however, no significant differences were found about the regions where the teenagers of our sample live and the other variables related to the covid- experience. this is very interesting data, leading us to hypothesize that the participants’ negative feelings may be more related to the adolescent period than to the pandemic itself. since march , many countries throughout the world have been in lockdown in response to the covid- pandemic. in italy, the quarantine began on march , , and containment measures were partially reduced only on may , . people had to stay at home. all social and sporting activities were canceled, and many work activities were forbidden; schools were closed and will reopen only with the new academic year. many countries around the world have temporarily closed educational institutions to contain the spread of the covid- pandemic. according to the united nations educational, scientific and cultural organization (unesco), school closures have impacted over % of the world's student population (unesco, ) . italy and several other countries have used educational technologies, including online platforms, radio, television, and texting, to support access to remote learning during the covid- pandemic, and so guarantee the students' right to education (unesco, ) . changes in life caused by the pandemic were dramatic for people of all ages. however, the revolution of behavioral routines caused by quarantine can have been particularly hard to accept for young people who could not go to school, play sports, and meet friends. usually, adolescents spend much of their waking time in school or other social contexts, such as gyms or recreational spaces (mahoney et al., ). now they have had to stay home all day for months, with online relationships only with peers and adults, such as their teachers, except the persons that live with them. moreover, their home has become a "school." for these reasons, there was an extensive debate between scientists on the effects of quarantine, limitations of freedom, and school closures on adolescents' emotional and affective states. previous studies on the effect of quarantine, which is the separation and restriction of movement of people who can be potentially exposed to a contagious disease, report common psychological effects (brooks et al., ) . research conducted on the severe acute respiratory syndrome (sars) epidemic (hawryluck et al., ; mihashi et al., ; liu et al., ) reported a high prevalence of symptoms of psychological distress, such as insomnia, irritability, anger, and other mood disorders. however, although the risk for psychological disorders related to changes in life caused by an epidemic is largely documented, there is evidence that several sociodemographic and psychological variables influence the emotional responses to behavioral limitations (hawryluck et al., ) . in particular, the perceived health risk for disease affects the emotional responses to the prevention measures and the acceptance of limitations of behavior (tang and wong, ; commodari, ) . health-related perceived risk depends on perceived "seriousness" and perceived "susceptibility" to a disease. risk perception is one of the key drivers of health behavior (brewer et al., ; ibuka et al., ; commodari et al., ) and influences the adoption of precautionary measures. perceived seriousness refers to how at risk a person considers himself to develop a disease, while perceived susceptibility concerns the perceived probability of getting a disease. perceived susceptibility can be differentiated into perceived personal susceptibility, which is the perceived probability that one will be harmed by a hazard (rogers, ) , and perceived comparative susceptibility, which is the perceived probability that a hazard will hurt one compared with other people of the same age and gender. research on health-related risk perception in young people has shown that adolescents engaged in risky behavior do not have a complete appreciation of their exposure to harm (johnson et al., ) . however, there are no previous studies on the health risk perception for pandemic diseases and the psychological experiences related to quarantine in this stage of the life span. based on these considerations, the main goal of this study was to investigate the perceived risk related to covid- and the psychological experiences of adolescents during the pandemic. in particular, the purpose of the study was to analyze the perceived seriousness of and susceptibility to covid- , the beliefs of adolescents in the first phase of quarantine and their opinions on the stage two of quarantine, during which a partial reduction of behavioral measures was hypothesized. moreover, the study explored adolescents' moods, emotions, and feelings, with attention to expectations for the immediate future. more specifically, the study intended to verify the following hypotheses: hypothesis (h ): living in an area with more restrictions than in other areas of the country contributes significantly to increasing the disease's perception of risk. hypothesis (h ): living in an area with more restrictions than in other areas of the country significantly contributes to accentuating the negative psychological impact of the quarantine experience. hypothesis (h ): other sociodemographic variables influence health risk perception and psychological experiences of the adolescents in the sample. in particular, a higher perception of risk and a greater concern of contracting covid- were expected to predict more negative feelings during the quarantine. conversely, it was expected that higher adherence to government measures to contain the infection and greater confidence in the information received on covid- were predictors of positive emotions. participants were adolescents (males = ; females = ) who attended upper secondary school (age range: - , m = . , sd = . ), which corresponds to the international standard classification of education level . the participants lived in of the italian regions which are the first-level constituent entities of the italian republic. five hundred and seventy four of the respondents lived in a provincial seat, while lived in towns that are not the provincial seat. teachers and some students collaborated in the recruitment of the participants, sharing an online survey on the leading social networks and inviting students to respond to the questionnaire. data were collected using an internet-based questionnaire. in total, the questionnaire consisted of multiple-choice and open-ended questions. participation was voluntary, and the questionnaire required approximately - min to complete. the survey collected sociodemographic information, such as age, gender, area in which the respondents live, type of upper secondary school, academic grade, number of persons in the household, and other information. moreover, it explored perceived health risk related to covid- , knowledge and information on measures to control the pandemic, beliefs and opinions on stage two of the quarantine, routines and habits of life that adolescents miss most (such as going out with friends, meeting boyfriend or girlfriend, going to visit their relatives, for a total of six items), and psychological experiences related to quarantine. the survey also collected information on e-learning experiences during quarantine, but this subject is beyond the scope of this article. perceived seriousness, perceived personal susceptibility, and perceived comparative susceptibility to covid- , which are the main dimensions of risk perception related to health, were investigated using an adjustment of the italian version (commodari, ) of the risk perception of infectious diseases questionnaire (brug et al., ) . participants responded to questions using a five-point likert-type scale. the participants were invited to report (a) how serious it would be for them to get the disease, (b) how likely they think they are to contract the disease, (c) whether they would have a smaller or larger chance of getting the disease before summer, compared with their peers of the same age and gender, and (d) if they believed that students could be a category particularly at risk of contracting the virus. the original version of the risk perception of infectious diseases questionnaire was developed during the sars epidemic, and it was translated into several languages (de zwart et al., ) . its psychometric characteristics are good (cronbach's α = . ), and many international studies have used this measure in different contexts (de zwart et al., (de zwart et al., , commodari, ) . in this regard, a recent study by commodari ( ) used the italian adjustment of this questionnaire to investigate the risk perception of flu and the role of sociodemographic and psychological variables on perceived risk. a confirmatory factor analysis was run to assess the validity of this adapted version and a good model fit was obtained [χ ( ) = . ; p < . ; rmsea = . ; srmr = . ; cfi = . ; tli = . ]. the reliability is also good (cronbach's α = . ). regarding their opinions and beliefs, participants were asked to indicate whether they agreed with statements of reported information on covid- and quarantine (e.g., "there are some categories of people at higher risk for covid- than the general population"; "in stage two of the quarantine it is necessary to avoid the use of public transport to reduce the risk of contagion and to avoid a new increase in the epidemic, " and others). regarding feelings, emotions, and moods, participants were asked to complete a likert-type scale that focused on the personal feelings about one's cognitive, physiological, and behavioral state. participants indicated their level of agreement with several statements using a five-point likert-type scale (e.g., "in this period in which i have to stay at home, i feel well physically"; "in this period in which i have to stay home i am tense and i feel tight"). the scale measured two aspects: "negative feelings" and "positive feelings." a high score corresponded to high perception of negative or positive feelings, respectively. a cfa was also performed to assess the validity of these scores. regarding the model for the "negative feelings, " although the chi-square statistic resulted to be statistically significant [χ ( ) = ; p < . ], the other values were indicative of a good model fit (rmsea = . ; srmr = . ; cfi = . ; tli = . ). the same result was obtained also for the model of the "positive feelings" [χ ( ) = . ; p < . ; rmsea = . ; srmr = . ; cfi = . ; tli = . ]. both the scales showed a good reliability (negative feelings: cronbach's α = . ; positive feelings: cronbach's α = . ). the scores were converted into z scores for the purpose of statistical analyses. finally, participants were asked to answer questions about their routines and habits of life they missed most, and to complete, without a word limit, the sentence "in this period in which i have to stay home, i think my summer will be. . .". participants completed the online survey between april and may , . the statistical package for the social sciences (spss) version . (ibm corporation, armonk, ny, united states) was used for the statistical analyses. quantitative data were expressed as frequencies and percentages in the case of categorical and ordinal variables and as mean and standard deviation in the case of continuous variables. independent-samples t-test were run and the magnitude of the differences between the means were assessed using hedges's g formula (hedges, ) to calculate effect size (es), with . indicating a small es, . a medium es, and . a large es (cohen, ) . besides, two multiple regression analyses were calculated to investigate the impact of sociodemographic variables and risk perception on positive and negative feelings experienced during the lockdown. for this purpose, the main sociodemographic variables and risk perception values were the independent variables, and the "positive feelings" and "negative feelings" scores were the dependent variable. qualitative data were coded and analyzed to show emerging themes. the thematic coding structure's development and confirmation was an iterative process involving two researchers conducting individual, recursive reading of the textual data and group meetings to discuss and test the emerging themes. discrepancies were resolved by consulting specific instances in the data, discussing their relationship to establish themes, and reaching consensus as a group (corbin and strauss, ) . the participants' responses to some items were free, and these could be single words or full sentences. codification was realized using "thought unit, " also denoted "sense unit" or "unit of meaning." the units comprised one idea communicated, whether it was expressed as a sentence, a verb-object sequence, or a single word. the responses were first categorized into categories. then, these categories were progressively reduced. finally, the responses were coded in categories (srnka and koeszegi, ) . the study was performed following the ethical standards of the declaration of helsinki and followed the ethical code for italian psychologists (l. . . , n. ), italian law for data privacy (dlgs / ) , and the ethical code for psychological research (march , ) approved by the italian psychologists association. no sensitive data that could identify the participants was collected. the schools involved in the research had previously informed the students' parents to consent to the study's participation. the chair of school and family psychology, disfor, university of catania, approved this study. table shows the sociodemographic characteristics of the participants. participants lived in of the italian regions. covid- was not diffused equally throughout italy. the regions in which participants lived reported on may different levels of contagion, with a broader spread of the virus in the regions of northern italy. of the student participants, . % lived in "red zones, " which were places in which the government has imposed stricter measures of containment due to an exponential and uncontrolled growth in cases of contagion compared to other areas in italy. regarding the number of persons with whom respondents were spending quarantine, . % reported that the size of their household was more than four people, including themselves; . % reported four persons; . %, three persons; . %, two persons; and . % reported living alone during the quarantine. five of the respondents ( . %) had been or was currently suffering from covid- , while . % of the respondents reported that they were uncertain about having had this disease; . % of respondents reported that one or more family members living with them had been or was currently suffering from covid- . furthermore, . % of the sample had at least one family member who worked with people affected by covid- (health care or other essential services). table reports data on the perceived health risk in the sample. in general, adolescents considered the possibility of getting covid- before summer to be low: . % responded that their probability was very low or low, while only . % think that this probability was high or very high ( . % very low; . % low; . % neither low nor high; . % high; and . % very high). furthermore, most subjects believed they had a low or very low probability of contracting the virus before the summer compared to peers of the same age and gender ( . % very low; . % low; . % neither low nor high). only . % of the sample thought that this probability was high and . % very high. regarding the perceived seriousness of the disease, % of the sample believed that contracting the virus could be serious or very serious ( . % serious and . % very serious), and . % believed that it was neither serious nor not serious. only a limited there were no differences by regions in perceived health risk, although some of these regions were the most affected by the disease. moreover, the presence of parents or other family members who had been or was currently suffering from covid- did not influence risk perception. interestingly, both the perceived susceptibility (family member working with people who had covid- : m = . , sd = . ; no family member working with people who had covid- : m = . , sd = . ; t = . ; p = . ; g = . ) and comparative susceptibility (family member working with people who had covid- : m = . , sd = . ; no family member working with people who had covid- : m = . , sd = . ; t = . ; p < . ; g = . ) were higher in those adolescents whose parents or other relatives worked with persons sick with this disease and the es was medium. to better analyze the perception of the risk for covid- , students were also invited to report how frightened they were of getting covid- . the majority of the sample was not particularly afraid of contracting covid- . there were no significant differences in the responses of the students by region. the students who lived in the regions with higher diffusion of the disease did not show greater fear of the disease than their peers. however, the students who lived in a red zone were more fearful of covid- than their peers with a medium es (red zone: m = . , sd = . ; non-red zone: m = . , sd = . ; t = . ; p = . ; g = . ) ( table ) . ninety three percent of respondents believed that there were categories of people more at risk of getting covid- , but, interestingly, . % of the respondents did not consider students as a category at risk for covid- . the remaining participants ( . %) believed that students were an at-risk category for this disease. these students motivated their response with the argument that the school setting does not permit social distancing. the remaining . % did not answer the question. a significant percentage of the respondents reported having confidence in the information that they received on the disease ( . % trust enough; . % trust a lot; . % trust very much). moreover, the most critical information the adolescents would have liked to receive on covid- concerned on how to cure the disease ( . % of the respondents). interestingly, only . % of respondents were interested in how to prevent the infection. furthermore, . % wanted information on the likelihood of contracting the virus in an area of residence, . % how to recognize the symptoms of the disease, % the geographical areas where the virus is most present, and only . % would have liked to have been more informed about how the virus was transmitted. regarding adolescents' opinions of the behavioral measures that could be useful to maintain in stage of quarantine, during which there was a partial reduction of containment measures, a very high percentage of respondents ( . %) agreed on the need to avoid public transport, such as trains or busses, as well as to confined spaces such as bars, restaurants, cinemas, theaters, and school classrooms ( . %). similarly, . % of the respondents agreed with the need to avoid going into shops if not necessary and only with personal protective equipment, such as a face mask. further, . % of the respondents agreed with the need to avoid going to gyms or swimming pools, and . % considered it useful to avoid medical consultations if possible. however, adolescents did not think it will be necessary to maintain social distancing in the second quarantine stage. most respondents did not agree on the need to avoid staying with persons who are not cohabiting ( . %), and . % think that it is not necessary to avoid staying in open places such as parks. data are summarized in table . interestingly, adolescents showed a high awareness of the particularity of the moment in which italy was living. the vast majority of the sample reported having no difficulty complying with the government's restrictive provisions ( . %), and they substantially agreed with the restrictions imposed on citizens due to the pandemic ( . %). the majority of the adolescents interviewed said they had more homework than before due to the remote school activities ( . %), and a significant percentage of them ( . %) reported having little free time. however, a large percentage was coping with quarantine by dedicating at least h a day to a hobby ( . %), watching television or playing video games ( . %), or spending much more time on social networks such as facebook or instagram ( . %). more specifically, males tended to devote themselves to hobbies ( differences were small or approaching to medium. furthermore, adolescents living in a red zone tended to watch tv or play video games more than peers who did not live in a red zone but the difference was small (red zone: m = . , sd = . ; no red zone: m = . , sd = . ; t = . ; p = . ; g = . ) ( table ) . as for what the teenagers in the sample missed most in this time of restrictions, the majority of the participants stated that they especially missed being able to meet friends ( . %) and relatives ( . %) and staying out later in the evening due to the closure of premises such as restaurants, pubs and discos and the prohibition to go out except for reasons of absolute necessity ( %). to confirm this, a large majority of the sample said they found significant support from family ( . %) and friends ( . %) to face this time when they had to stay home. the responses of the adolescents show heterogeneous psychological reactions to the experience of quarantine. to better investigate the specific emotion and feeling they perceived, the responses to some of the more relevant items were first examined. this analysis aimed to capture a snapshot of the emotional state of adolescents during the quarantine. the majority of those interviewed stated that they stayed physically well ( . %). males felt better than females but the difference was small (males: m = . , sd = . ; females: m = . , sd = . ; t = . , p < . ; g = . ). however, quarantine influenced their sense of security and self-confidence: . % of the students reported feeling less secure than in the past. females were less self-confident than males with a medium es (females: m = . , sd = . ; males: m = . , sd = . ; t = . , p < . ; g = . ) while there were no significant differences by age, the area in which the person lived, and other socio-demographic variables. concerning psychological status, about % of students reported feeling tenser and sadder ( . %) and more irritable table . interestingly, the responses of the students showed their great empathy and interest in socialization. a high percentage of respondents said they were not bored listening to others' problems ( . %) and reported being able to manifest their emotions ( . %). these results agree with the findings discussed in the previous section, which showed that the things and situations students missed most were meeting friends, staying with relatives, and being out late in the evening. multiple regression analyses were performed to investigate the impact of sociodemographic and perceived health risk variables on the psychological outcomes (positive and negative feelings z scores). sociodemographic variables, perceived health risk, and adherence to government restrictive measures were used as independent variables while positive and negative feelings z scores were the dependent variable. all regression assumptions were checked. there was linearity as assessed by partial regression plots and a plot of studentized residuals against the predicted values. there was independence of residuals, as assessed by a durbin-watson statistic of . . there was homoscedasticity, as assessed by visual inspection of a plot of studentized residuals versus unstandardized predicted values. there was no evidence of multicollinearity, as assessed by tolerance values greater than . . there were no studentized deleted residuals greater than ± standard deviations, no leverage values greater than . , and values for cook's distance above . the assumption of normality was met, as assessed by a q-q plot. regarding positive feelings, a significant regression equation was found (f = . , p ≤ . ), with an r square of . . more specifically, significant predictors of positive feelings were gender (t = − . , p < . , std β = − . ), region (t = . , p = . , std β = . ), confidence in the information received on covid- (t = . , p = . , std β = . ), perceived susceptibility (t = − . , p = . , std β = − . ), ease in respecting government measures (t = . , p < . , std β = . ), and belief that the government measures were justified (t = . , p = . , std β = . ). according to these results, females reported less positive feelings than males on average as well as adolescents living in northern italy. furthermore, higher confidence in the information received on covid- , higher perceived susceptibility, higher ease in respecting government measures and higher beliefs that these measures are justified were predictive of positive feelings. table shows the significant results of the regression analyses and the contribution of each predictor to the dependent variable. regarding negative feelings, the regression model was significant (f = . , p < . ), with an r square of . . more in detail, significant predictors of negative feelings were gender (t = . , p < . , std β = . ), age (t = . , p < . , std β = . ), living in a red zone (t = . , p = . , std β = . ), perceived seriousness (t = . , p = . , std β = . ), fear of getting covid- (t = . , p = . , std β = . ), and compliance with government measures (t = − . , p = . , std β = . ). according to these results, females and older adolescents reported more negative feelings than males and younger adolescents on average. furthermore, living in a red zone, a higher perceived seriousness, a higher fear of getting covid- and a lower compliance with government measures were predictive of negative feelings. table presents the significant results of the regression analyses and shows the contribution of each predictor to the dependent variable. in summary, the model showed a moderate but significant impact of both the sociodemographic and the health risk perception variables related to covid- experience on the perception of negative and positive feelings. participants' expectations for the immediate future were also investigated through an open-ended question about how they imagined the upcoming summer holidays. data were codified according to the modality described in the procedure section. we first categorized the responses into categories and then progressively reduced these categories. the adolescents were aware that the experience of quarantine would continue to produce effects during the summer period. in this regard, a significant percentage of the adolescents in the sample said that their summer would be "different" or would have "different limitations" ( . %). uncertainty and doubt were widespread feelings, as expressed by these answers: "my summer will be full of anguish, doubts, and perplexity, but my friends are enough for me to feel good"; "i am very worried because i cannot imagine how it will be but i look forward to it as much as every year." in addition, . % of the sample thought that summer would be sad, boring, or horrible, as demonstrated by the following quotes: "if transportation does not reopen, my summer will be wasted, the collapse of different dreams and projects that have so far pushed me to go on and resist a very dense pool of mud in which i will struggle so much not to sink"; "like a prison, locked between the walls and between the screams, as if it were winter or autumn, without being able to see the seawater, which is freezing or boiling." in some cases, quarantine only highlighted pre-existing difficult situations, such as in this case: "it will be the usual summer in which it is amplified that i have no friends and that no one ever invites me to go out and i will spend the day in my pajamas eating food at will and watching netflix and sky until i vomit." however, . % thought summer would still be interesting and fun: "interesting, i have high school exams, a girl, and too many friends to share my life with. i'm curious to see how everything will evolve." in general, the teenagers in the sample looked to the future with the hope of overcoming the difficult period of the pandemic and resuming a normal life, even if different from the previous one: "i don't care what my summer will be like, i just hope that we will be able to get out of this situation with a new unitary spirit, i hope people understand that we are one family in one house"; "meeting again, with the necessary restrictions, my friends, my grandparents, will be difficult as if we had to learn to live in a different way from what we were used to until a couple of months ago. it will be exciting!". this study aimed to provide a general overview of the psychological experience of quarantine in a large sample of italian adolescents. according to other recent studies conducted on this topic in the italian population, the covid- emergency was a very difficult experience from an emotional point of view, and several categories such as health professionals have undergone significant stress with a consequent negative impact on their psychological well-being (ramaci et al., ) . to the best of the knowledge, this is the first study investigating the experience of the covid- pandemics in italian adolescents and with such a large number of subjects. first, the perceived seriousness and susceptibility to covid- were evaluated, as well as the impact of sociodemographic variables on the perception of health risk. according to the study results, italian adolescents had a low perception of risk of covid- . perceived comparative susceptibility and perceived seriousness in italian adolescents were also very low. these results show that young people think that covid- is not a potentially severe disease for them. indeed, there is some evidence that young people are less vulnerable to the effects of the new coronavirus sars-cov- (kolifarhood et al., ) , although the possibility of getting the disease depends on the diffusion within the population. they underestimate the probability of getting the disease and show a very high trust in their good health, neglecting that the probability of being infected, albeit slight, is similar to that of their peers and people of other age groups. interestingly, as hypothesized, teenagers residing in a red zone reported higher perceived seriousness and susceptibility than those who did not reside in these zones. furthermore, females showed a higher perceived seriousness than males. in both cases, the medium effect size suggests a role of these variables in influencing health risk perception. therefore, living in an area with more restrictions than in other areas of the country may have contributed significantly to increase the perception of risk about the disease. also, this information seems consistent with several studies demonstrating that women tend to have a higher perception of risk than men, thus avoiding risky behaviors to a greater extent (harris et al., ) . despite underestimating their risk of infection, however, the italian teenagers who participated in this study were aware of the restriction measures necessary to contain the spread of the virus and they agreed with the limitations imposed by the government. these responses show high awareness of the potential danger of covid- and acquire more value when considering that young people were conscious that they were not at serious risk, but that the risk was high for society as a whole. the study also wanted to investigate the emotional and psychological impact of the quarantine period on the youth population in italy. as underlined in the literature on this topic, prolonged school closure and home confinement during an epidemic can have a detrimental effect on children's and adolescents' physical and psychological well-being (brooks et al., ; wang et al., ) . according to the results, this study also shows that italian adolescents suffered the psychological effects of this quarantine period. indeed, they had more marked negative feelings. more specifically, females and adolescents residing in the red zones with more restrictions showed higher levels of negative feelings related to the quarantine, in accordance with the study's hypotheses. in this regard, effect sizes approaching to medium indicate a possible role of these variables in determining negative feelings in the adolescents of the sample, even if these feelings may be likely influenced also by other variables not considered in the study. however, it is essential to emphasize that these feelings are subjective perceptions rather than a psychopathological state. indeed, quarantine did not reduce the empathy and sociability of young people. in this regard, the adolescents who participated in this study reported that they engaged in school activities remotely and carried out the assigned homework. furthermore, they continued to listen to the problems of others and to express their emotions. no significant differences related to the regions where the teenagers lived were found. this is very interesting data, leading to hypothesize that the negative feelings reported by the participants may be more related to the adolescent period than to the pandemic itself. the results also showed a moderate but significant impact of both the sociodemographic and the health risk perception variables related to the covid- experience in the perception of negative and positive feelings. more specifically, being male, living in a region with less virus spread, reporting low levels of perceived susceptibility and high compliance and agreement with government measures were all variables associated with the perception of positive feelings. on the contrary, being female and older, living in a red zone, reporting high levels of perceived seriousness and fear of getting covid- , and being less compliant with government measures were associated with more negative feelings. as recently underlined by wang et al. ( ) , the adverse effects on psychological well-being are more significant when children and adolescents are confined to their homes without the possibility of carrying out activities outside and meeting peers. in confirmation of these considerations, this study showed that the majority of teenagers interviewed suffered in particular from not being able to meet friends and relatives, as well as from not being able to go out and stay out late in the evening. the results also confirmed the literature data demonstrating that when children and teenagers do not go to school and stay home, they are physically less active, are exposed to much more screen time, and have irregular sleep patterns. similarly, the teenagers in the sample had difficulty falling asleep and spent more time watching television, playing video games, or using social networks. however, a significant percentage stated that they dedicated at least an hour a day to playing a musical instrument, dancing, exercising, acting, or drawing. in light of these insights, it is important to promote healthy habits and lifestyles in adolescents to reduce psychosocial stress and improve the psychological and physical well-being of the young population. finally, the quarantine experience was also associated with a widespread sense of uncertainty about the near future in the adolescents interviewed in this study. although some participants were convinced that the virus would disappear during the summer, allowing a return to normal life, most believed that their near future would be unpredictable or different due to the various health and social distancing rules that must be respected. in this regard, it is important to support the youth population in addressing the uncertainties related to the period following the quarantine to ensure better adherence to the limitations that will have to be faced to avoid a new outbreak of the epidemic. this study has some important strengths. as already underlined, this is one of the first studies conducted in italy about the perceived seriousness and susceptibility for covid- as well as the effects of school closures and home confinement on the physical and psychological well-being of adolescents. another strength is certainly the large sample size, with almost italian teenagers interviewed from north, central, and south italy. however, there are also several limitations. first, this is a cross-sectional study so an exact causal relationship between the variables could not be established. secondly, an internetbased questionnaire with self-reported measures was used, so it was not possible to ascertain the accuracy of the answers to the questions and the possible influence of selfreport bias on the results. finally, not all italian regions are represented in the sample; however, it is representative of the three main areas in which italy is generally divided (north, central, and south). this study has several social and psychological implications. in particular, the results underline that the covid- emergency has undoubtedly had a significant impact on the lifestyle and psychological well-being of italian adolescents. in light of these findings, the physical and mental impact of the covid- epidemic on children and adolescents is a matter of fundamental importance both for governments and families and cannot be neglected, especially in this phase of a progressive resumption of ordinary life. therefore, it is necessary to prepare adequate strategies to support the youth population in addressing the uncertainty associated with the pandemic and the quarantine period to reduce the psychological impact of school closures and home confinement as much as possible and guarantee adequate support to deal with the return to school. the raw data supporting the conclusions of this article will be made available by the authors, without undue reservation. the studies involving human participants were reviewed and approved by chair of school and family psychology, disfor, university of catania. written informed consent to participate in this study was provided by the participants, and where necessary, the participants' legal guardian/next of kin. ec designed the study, analyzed the data, and wrote the first draft. vllr revised the manuscript. both authors contributed to the article and approved the submitted version. risk perceptions and their relation to risk behavior the psychological impact of quarantine and how to reduce it: rapid review of the evidence sars risk perception, knowledge, precautions, and information sources, the netherlands statistical power analysis for the behavioral sciences the role of sociodemographic and psychological variables on risk perception of the flu health risk perceptions in the era of the new coronavirus: are the italian people ready for a novel virus? a cross sectional study on perceived personal and comparative susceptibility for infectious diseases basics of qualitative research : techniques and procedures for developing grounded theory, edn perceived threat, risk perception, and efficacy beliefs related to sars and other (emerging) infectious diseases: results of an international survey monitoring of risk perceptions and correlates of precautionary behaviour related to human avian influenza during - in the netherlands: results of seven consecutive surveys gender differences in risk assessment: why do women take fewer risks than men? sars control and psychological effects of quarantine distribution theory for glass's estimator of effect size and related estimators the dynamics of risk perceptions and precautionary behavior in response to (h n ) pandemic influenza risk involvement and risk perception among adolescents and young adults epidemiological and clinical aspects of covid- ; a narrative review depression after exposure to stressful events: lessons learned from the severe acute respiratory syndrome epidemic adolescent out-of-school activities predictive factors of psychological disorder development during recovery following sars outbreak social stigma during covid- and its impact on hcws outcomes cognitive and physiological processes in fear appeals and attitude change: revised theory of protection motivation from words to numbers: how to transform qualitative data into meaningful quantitative results an outbreak of the severe acute respiratory syndrome: predictors of health behaviors and effect of community prevention measures in hong kong. china covid- educational disruption and response mitigate the effects of home confinement on children during the covid- outbreak the authors wish to thank the united network europe for the fundamental support in the diffusion of the online survey and dr. giulia carnemolla and dr. jessica parisi for the help in the data collection. the authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.copyright © commodari and la rosa. this is an open-access article distributed under the terms of the creative commons attribution license (cc by). the use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. no use, distribution or reproduction is permitted which does not comply with these terms. key: cord- -t b wpe authors: falcone, rino; colì, elisa; felletti, silvia; sapienza, alessandro; castelfranchi, cristiano; paglieri, fabio title: all we need is trust: how the covid- outbreak reconfigured trust in italian public institutions date: - - journal: front psychol doi: . /fpsyg. . sha: doc_id: cord_uid: t b wpe the central focus of this research is the fast and crucial impact of the covid- pandemic on a crucial psychological, relational, and political construct: trust. we investigate how the consequences of the pandemic, in terms of healthcare, state intervention and impositions, and daily life and habits, have affected trust in public institutions in italy, at the time when the contagion was rapidly spreading in the country (early march ). in this survey, addressed to italian citizens, we analyzed and measured such impact, focusing on various aspects of trust. this attention to multiple dimensions of trust constitutes the key conceptual advantage of this research, since trust is a complex and layered construct, with its own internal dynamics. in particular, the analysis focuses on how citizens attribute trust to public authorities, in relation to the management of the health crisis: with regard to the measures and guidelines adopted, the purposes pursued, the motivations that determine them, their capacity for involvement, and their effectiveness for the containment of the virus itself. a pandemic creates a bilateral need for trust, both in public authorities (they have to rely on citizens’ compliance and must try to promote and maintain their trust in order to be effective) and in citizens, since they need to feel that somebody can do something, can (has the power to) protect them, to act at the needed collective level. we are interested to explore how this need for trust affects the attributional process, regarding both attitudes and the corresponding decisions and actions. the most striking result of this survey is the very high level of institutional trust expressed by respondents: % of them trust italian public authorities to be able to deal with the covid- emergency. this is in sharp contrast with the relatively low levels of institutional trust characteristic of italy, both historically and in recent surveys. moreover, the survey allowed the discrimination of several potential predictors for trust, thus emphasizing factors that, during this crisis, are exhibiting an anomalous impact on trust. the great societal challenge presented by the covid- pandemic has prompted extraordinary efforts to meet such a challenge, from public authorities, civil society, and the scientific community. extreme policies for containment, mitigation, and co-existence with the virus have been implemented by the governments of most afflicted countries, as well as by relevant international institutions (e.g., the who and the eu). at the same time, scientific research worldwide has focused on tackling the many facets of this dramatic phenomenon, including its impact on social relationships and psychological well-being, as well as the key socio-cognitive factors in promoting effectiveness of the proposed countermeasures. several of these studies have highlighted the crucial and complex role of trust in dealing with the covid- pandemic. llewellyn ( ) puts it very succinctly and effectively: "in times of crisis, trust is the most important thing to consider if you want to communicate health advice." this blanket pronouncement is well-supported by previous evidence: in their systematic review on the importance of trust when preparing for and during a pandemic, siegrist and zingg ( ) found confirmation that "trust in health agencies positively influenced people's willingness to adopt recommended behavior." in addition, among the five recommendations for crisis communication highlighted by the authors' survey, two directly concern trust management: "the focus should be not only on trust but also on confidence, and establishing trust in health authorities before a pandemic occurs is important." this latter point is also stressed by lewnard and lo ( ) , with reference to the current pandemic: "the effectiveness and societal impact of quarantine and social distancing will depend on the credibility of public health authorities, political leaders, and institutions. it is important that policy makers maintain the public's trust through use of evidence-based interventions and fully transparent, factbased communication." it is worth noting that this emphasis on evidence and transparency, albeit crucial, describes only part of the relevant socio-cognitive dynamics that affect trust in public institutions: in particular, it collapses trust to confidence in information sources and their credibility, while a crucial problem is also trust in the institution's power to intervene, as well as trust in collective compliance with the proposed measures. finally, in specific circumstances, interesting inversions in cognitive causeeffect relationships can occur, as widely studied in cognitive sciences and social psychology (e.g., festinger, ; koller, ; kunda, ; epley and gilovich, ) . in fact, the relevance of trust for dealing with health emergencies is also linked to the limits of direct enforcement of the required behavioral change: without the active cooperation of the population, any drastic intervention is doomed to fail, because the desired behaviors (e.g., frequently sanitizing one's hands, wearing a facemask, and keeping a safe distance from others) cannot be effectively monitored on the required scale and with sufficient frequency. in a broad and comprehensive survey of social and behavioral results to support covid- pandemic response, van bavel et al. ( ) highlight how most measures needed to contain an epidemic are, by their very nature, difficult to enforce directly: this, in turn, makes trust in public authorities all the more relevant. based on scientific evidence gathered during previous outbreaks, van bavel et al. ( ) argue that "trust in institutions and governments (. . .) may play an important role." for example, trust in the liberian government was correlated with decisions to abide by mandated social distancing policies and utilizing clinics for care during the ebola outbreak. trust was also related to decisions to adopt preventive measures such as ebola vaccinations in the drc. conversely, a lack of trust in public health officials may lead to negative effects on utilization of health services. reliable information and public health messages are needed from national leaders and central health officials. however, local voices can amplify these messages and help build the trust that is needed to spur behavioral change (van bavel et al., ) . these expectations on the positive role of trust in promoting adherence and compliance with preventive regulations and guidance are finding ample confirmation also in recent studies on the ongoing crisis, both within and across various countries. in a nationally representative survey conducted in denmark during the covid- pandemic (n = ), olsen and hjorth ( ) measured the respondents' willingness to apply social distancing in order to reduce contagion: they found that both lower levels of political trust and lower generalized social trust are negatively associated with willingness to distance and that younger male respondents with the lowest levels of education and least political trust report lower willingness to distance. in a nationally representative survey of italian adults (n = ) conducted between the th and th of march , barari et al. ( ) observed high levels of understanding and self-reported compliance with containment measures, and noted that "even those who do not trust the government, or think the government has been untruthful about the crisis believe the messaging and claim to be acting in accordance." trust acts as a precious commodity both for institutions and for scientists, both of which are crucial actors in the public response to the covid- pandemic. in a large-scale background analysis of european social survey data on european countries (n = , ) focused on the covid- epidemic from january to april , , oksanen et al. ( ) found that institutional trust acts as a protective factor: countries with low levels of institutional trust prior to the outbreak (including italy) experienced significantly higher mortality rates during the crisis; moreover, their governments introduced restrictions against contagion later than countries with higher levels of institutional trust (calculated as the delta between the date when the restrictions came into effect and when the first confirmed covid- death was reported in that nation), which in turn contributed to the severity of their death toll. these results on the relevance of trust as a protective factor are in line with previous studies on other epidemics, e.g., ebola, showing how people with higher institutional trust are more likely to follow the advice and guidelines given by the health authorities (blair et al., ; vinck et al., ) , as well as investigating the interplay between scientific and non-scientific sources in modulating people's trust in healthcare information (falade and coultas, ) . as for trust in science, its role has been highlighted in a recent study by plohl and musil ( ) : using structural equation modeling (sem) on a sample of international, english-speaking respondents, the authors investigated whether and how risk perception and norm compliance for the covid- pandemic may be affected by several constructs, i.e., religious orthodoxy, conspiracy ideation, intellectual curiosity, and trust in science, all measured with validated scales. their results indicate that trust in science is by far the most important factor in producing appropriate risk assessment and high level of norm compliance. at the same time, trust in science, as opposed to the tendency to believe in alternative non-official sources, has been observed to be deeply affected by polarization and homophily (bessi et al., ) . looking at the specifics of the covid- pandemic, so far the most insidious threat posed by the virus has been the combination of the rapidity of its spread with the high number of patients requiring treatment in intensive care, resulting in unprecedented strain on the healthcare system of affected countries. this in turn has prompted an increasing number of national governments to adopt extreme measures to limit the spread of the virus, often imposing very demanding limitations on citizens' basic rights (e.g., social isolation, lockdown, and quarantine) and with dire socio-economic consequences (e.g., job insecurity, rising unemployment, loss of revenues, and inequalities). in such a unique scenario, the relevance of studying citizens' trust in public institutions is manifold: on the one hand, the effectiveness of these measures and the collective ability to overcome their costs is conditional on the compliance of the population, which in turn is affected by trust in institutions; for this same reason, institutions actively seek to promote citizens' trust, as a means to achieve their prevention goals; on the other hand, the very nature of the current crisis is likely to affect and shape how citizens conceptualize trust, and such socio-cognitive impact of the covid- pandemic needs to be understood. indeed, the current crisis acts as a magnifying glass in highlighting the essential role of trust in our societies (trust as "vinculum societatis, " the bond of society, to borrow john locke's famous expression), both for the psychological well-being of individuals and for the effective functioning of institutions. the study presented in this paper contributes to this fastgrowing body of knowledge on the interplay between trust in institutions and the covid- pandemic, by discussing the results of a large scale survey (n = ) conducted on italian citizens between march and march , . at that time, italy had the most active outbreak of the virus worldwide, and its death count was growing at alarming rates; at the same time, extreme prevention measures were still relatively recent and rapidly changing in nature, sometimes from day to day (e.g., on march new restrictions were introduced by the government, closing public places such as restaurants, pubs, and most shops). thus, our data offer insight into a time window in which the phenomenon was already in its acute phase in medical terms, yet still novel and unexpected for the population: this offers a privileged vantage point to observe how a pre-existing construct, trust in institutions, was affected by a sudden and profound change in the everyday functioning of the whole country, by a complete (albeit hopefully temporary) re-representation of one's role in society and in personal relationships, as well as in the relationship between citizens and institutions. the survey was theoretically inspired by the socio-cognitive model of trust developed by castelfranchi and falcone ( ) : we chose this theoretical framework because it provides a rich and nuanced description of various reasons for trust, thus allowing us to probe not only the degree by which italian citizens expressed trust toward the relevant public authorities engaged in the response to covid- but also on what grounds such attitude was based. our purpose, however, was not to look for direct validation of the theoretical model, but rather to collect as many detailed data as possible on the rapidly evolving italian response to the covid- emergency, from the standpoint of institutional trust: in this sense, this study was mostly intended as explorative. in particular, we wanted to compare our results with the welldocumented low levels of trust in institutions exhibited by italians before the onset of the crisis, which some have associated with tardiness in responding to the covid- emergency across various european countries (oksanen et al., ) : we intended to see whether such widespread distrust toward public institutions would be confirmed or subverted during the initial stages of the covid- outbreak in italy and to offer some insights and suggestions regarding the original and peculiar nature of any discontinuity in institutional trust that may be associated with the current pandemic. moreover, we intended to take a closer look at the cognitive and social factors responsible for trust toward public institutions in the face of pandemic threats: the survey was designed both to discriminate several potential predictors for trust, so that subsequent analysis would allow us to individuate the most relevant ones, and to facilitate comparison with the underlying theoretical model, thus emphasizing factors that, during this crisis, are exhibiting an anomalous impact on trust-either because they determine trust more intensely than usual (overcharged factors) or because their impact is minimal or non-existent (anesthetized factors). indeed, a key hypothesis that we wanted to test concerns the impact of covid- on the very nature of the institutional trust construct: not only the overall trust in public institutions is affected by the pandemic and how these institutions respond to it, but also the determinants of trust in institutions change and adapt to this crisis, in comparison with other situations. desperate times require desperate measures, and desperate measures induce a drastic reconfiguration of the cognitive underpinnings of trust in institutions. our survey was designed to collect data on such paradigm shift in how institutional trust was conceptualized by italian citizens during the early stages of the national response to the covid- pandemic. we used a snowball sampling method to determine the respondents: we collected a large sample (n = , % women, mean age = years, range = - years, sd = . ), relatively well-balanced in terms of geographical provenance ( % northern italy, % central italy, and % southern italy and main islands), with a significant portion of respondents ( %) residing in the regions most affected by covid- at that time (lombardy, veneto, emilia-romagna, marche, and piedmont) . the relatively uniform geographical distribution of the sample among the three macro-areas of italy, as well as the significant proportion of respondents from highly affected regions, allows interesting comparisons based on participants' residence. moreover, the introduction of more drastic restrictions by the italian government at the end of march , , invites considering also this temporal dimension in analyzing the data: in this respect, it is important that a fairly large set of participants (n = ) completed the survey after those new restrictions had been introduced. finally, it should be noted that the mean educational level of participants is very high: almost three quarters of respondents have a degree ( %) or postgraduate specialization ( %). the main characteristics of the sample are synthetized in table . data were collected with a -item questionnaire, using a fivepoint likert scale for most items: an english translation of the whole questionnaire is available in the supplementary materials. the questionnaire was based on the socio-cognitive model of trust developed by castelfranchi and falcone ( ) and explored the questionnaire was administered online using the google forms platform. the questionnaire fully complied with ethical guidelines for human subject research and participation was conditional on the preliminary approval of an informed consent by each subject; the compilation took an average time of min. data analysis was performed using the spss (version ) statistical software: the collected data were first analyzed through correlation analyses (given the asymmetric distribution of most variables, we considered spearman correlation values); secondly, given the high number of items in the questionnaire, we conducted a principal component analysis (pca) on each subsection of the questionnaire prior to running regression analyses on the aggregated data. full details on descriptive statistics for each item in the questionnaire are included in the supplementary materials, differentiating also based on geographical factors (northern, central, and southern italy; more affected vs. less affected regions) and temporal boundaries (before vs. after the march announcement of new restrictions by the national government). here, we report only the most relevant findings, prior to more in-depth analysis, and only in terms of aggregate data, since no significant differences emerged at this level between different areas and different dates (albeit some interesting patterns were detected via regression analysis, see section "regression analyses"). when asked to indicate which public authority is the most adequate to take decisions concerning the covid- emergency (item in the questionnaire), . % indicated the national government, . % indicated the civil protection, . % indicated the presidency of the republic, . % indicated the regional government, . % indicated the municipal authority, and . % indicated others. hence, the overwhelming majority ( . %) of respondents consider pandemics as a matter of national concern, which should be primarily addressed by national authorities. this should be taken into account while interpreting all other results, since most of the attitudes expressed by participants regarding features of public authorities (competence, intentionality, trust, etc.) should be understood with reference to national institutions, unless otherwise specified. moreover, it is remarkable that the presidency of the republic, which is mostly a moral authority, is seen as having a greater role than regional governments, in spite of their leading role in the healthcare system, which in italy is organized on a regional basis. equally significant is the fact that only . % of respondents (within the broader category "others") indicated any kind of international entity, including the european union, as having a primary role in facing a pandemic outbreak. in short, at this stage of the covid- emergency, italian citizens strongly believed that this pandemic was not to be prominently addressed by either regional or international authorities, but was rather mostly a matter of national concern. when asked to rank their overall trust in public authorities for the management of the covid- emergency (item in the questionnaire), % of respondents manifested either extreme ( . %) or high ( . %) levels of trust, . % were non-committal, and only . % expressed distrust (see figure , left panel). as we will see in the section "discussion and conclusions, " these numbers are in sharp contrast, to say the least, with the average institutional trust reported for italian citizens prior to the covid- crisis, especially considering that the main target of this newfound trust was national public authorities (see above). the competence of public authorities was assessed as their ability in planning both the right prescriptive measures (e.g., lockdown) and the appropriate behavioral guidelines (e.g., personal hygiene recommendations). on both counts, the majority of respondents expressed a positive belief in the public authorities' competence ( . % for measures, . % for guidelines), whereas only a relatively small minority was either undecided ( . % for measures, . % for guidelines) or skeptical ( . % for measures, . % for guidelines). moreover, correlational analysis indicates that competence scores for measures and guidelines are strongly and positively related (r = . , p < . ), suggesting that respondents did not really discriminate between prescriptive measures and behavioral guidelines, at least with respect to trust in public institutions: for this reason, in subsequent analyses, we collapsed these two items into a single competence value, calculated as the mean response for each subject to items (competence on measures) and (competence on guidelines) of the questionnaire (these are also the data reported in figure , central panel). other items in this section of the survey were designed to investigate the reasons behind participants' beliefs on the public authorities' competence: in summary, the overwhelming majority of the sample ( . %) believed that it was the public authorities' proper prerogative to take action and issue containment measures against the pandemic (item ), and most respondents ( . %) positively evaluated the use of experts' advice by the public authorities during the covid- crisis (item ); there was instead less confidence in the organizational capacity demonstrated by public authorities in the early stages of the emergency (item : . % expressed a positive evaluation, . % were undecided, and . % were critical), and the majority of the sample ( . %) agreed that institutional communication on the covid- presented some contradictions, either between different authorities or over time (item ). in spite of these partial concerns, a significant majority of the sample ( . %) did not express any skepticism on the competence of the public authorities in handling the emergency (item ). as for the competence, we inquired on the intentionality of public institutions separately for prescriptive measures and behavioral guidelines, asking participants whether they believed either type of intervention was both actively and honestly aimed at containing the covid- pandemic. again, respondents expressed an overwhelmingly positive belief in the good faith of public institutions, both in promulgating prescriptive measures ( . %) and in issuing behavioral guidelines ( . %): only a small minority was either undecided ( . % for measures, . % for guidelines) or skeptical ( . % for measures, . % for guidelines). correlational analysis reveals again that intentionality scores for measures and guidelines are strongly and positively related (r = . , p < . ), further confirming that respondents did not really discriminate between prescriptive measures and behavioral guidelines, when it comes to assessing the public authorities' trustworthiness in this emergency: hence, these two items on intentionality were collapsed into a unique intentionality value in subsequent analyses, using the mean response for each subject to items (intentionality on measures) and (intentionality on guidelines). other items in this section of the survey were designed to investigate the reasons behind participants' confidence, or lack thereof, in the nature of the public authorities' intentions: in summary, we found confirmation of the fact that most respondents ( . %) did not doubt that the intentions of the public authorities were consistent with their public statements (item ), whereas a smaller majority ( . %) considered the economic investment mobilized by the italian public authorities sufficient to fight the pandemic (item : notice that only . % considered it insufficient, with a significant portion of the sample, . %, remaining undecided). finally, asked whether other interests, e.g., political or economic, were at stake (item ), the larger part of the sample ( . %) answered in the negative, whereas . % acknowledged the presence of such ulterior motives and . % were unsure: as we will discuss further on, this question was probably easy to interpret in two markedly different senses-either negatively, as an accusation of having some hidden and problematic agenda, or positively, as the capacity to take into account all the key ramifications of the covid- crisis, including its political and socio-economic aftermath. overall, we registered strong confidence in the good faith of the intentions manifested by public institutions (figure , right panel): this parallels the belief in the public authorities' competence, and together, these attitudes support the high levels of institutional trust expressed by this sample. part of the survey was focused on the measures issued by public authorities as a response to the covid- pandemic, in order to estimate both their perceived usefulness and the goal attributed to these interventions by the participants. the vast majority of our sample ( %) perceived these measures as being either useful ( . %) or very useful ( . %) in fighting the pandemic, whereas only a tiny minority was skeptical ( . %), with the remaining . % being undecided (item ). when asked to assess the adequacy of the public authorities' intervention (item ), a more abstract notion involving a counterfactual comparison with alternative strategies, the majority rated current measures as adequate ( . %), . % were undecided, and only % considered them inadequate. in terms of the motivations associated with these measures, we asked participants to express agreement on three potential, non-mutually exclusive aims: reassuring the population (item ), curbing the spread of covid- (item ), and creating unmotivated alarm (item ). the vast majority ( %) agreed that the rationale of the public authorities' intervention is indeed to contain the pandemic, whereas only . % attributed to the public authorities the goal of reassuring citizens, and even fewer respondents ( %) regarded the proposed measures as a way of spreading unnecessary panic. when rating the personal burden of the proposed restrictions on their own lives (item ), % of participants expressed to feel a high level of impact, whereas . % indicated little discomfort for the current situation and the remaining . % reported medium levels of distress. however, regardless of the perceived impact on the public authorities' intervention, the overwhelming majority of respondents agreed that such sacrifices were crucially beneficial for themselves and their families (item , . % of agreement), for the society as a whole (item , . %), and for both (item , . %). moreover, when asked to assess the usefulness of one's personal contribution to these preventive measures, since they were intended for the whole population (item , a question aimed at implicitly measuring any "free-riding inclination" in our sample), as many as . % of the participants considered their personal role relevant for the collective effort. taken together, these data show that, albeit different people suffered more or less because of the containment measures, almost all agreed on their usefulness and on the necessity of personal sacrifice to deal with the pandemic: this suggests a mindset in which the shared goal of public safety trumps any individual concern, including personal discomfort, fear, and anxiety (an interpretation later confirmed by regression analysis, see section "regression analyses"). in terms of expectations on compliance with the sanitary restrictions by other fellow citizens (items - ), we observe a fairly varied pattern of response (see figure ): the most widespread belief ( . % of agreement) is that enough italian citizens, albeit not all, will comply with the regulations, thus making them effective (item ); in contrast, there is skepticism both on the most optimistic scenario, i.e., full compliance (item , . % of disagreement), and on the bleakest outcome, i.e., insufficient compliance (item , . % of disagreement), although it is worth noting that pessimism is rejected much more strongly than optimism. the possibility that only few people will comply, and yet their efforts will be useful (item ), is also rejected by the relative majority of the sample ( . % of disagreement), yet interpreting this result requires caution, since it could either express skepticism on figure | expectations on compliance by others. frontiers in psychology | www.frontiersin.org how many people will comply, or on the chances that limited compliance may indeed be useful. regarding the motivations useful to induce compliance, we asked participants to express agreement on four possible motivational triggers: the expectation that everybody else will follow the new regulations (item ), a personal concern for dangers (item ), a spirit of collaboration in the face of the emergency (item ), and trust in the fact that public authorities are doing everything in their power (item ). all four motivations engendered significant levels of agreement, with the highest being the feeling of a common cause against a shared threat ( %), followed by trust in maximum effort by the public authorities ( . %), concern for the associated risks ( . %), and expecting others to comply as well ( . %). it is interesting to note that a motivation tied to the collaborative dimension of trust in civil society, i.e., being united in pursuing a common goal, shows more than percentage points of distance from a motivation inspired instead by the sanctioning view of trust, i.e., being able to monitor compliance by others, possibly to punish free-riders, as well as from fear of personal harm: this suggests that emphasizing collaborative motives (a strategy employed quite consistently by the italian government in its public communications during the early stage of the covid- outbreak) may be more effective in promoting compliance than stressing individualistic goals. this section of the survey asked respondents to provide a metacognitive evaluation of the most relevant factors promoting their trust in how public institutions are handling the covid- crisis. of the eight factors explored, the type of measures adopted by the authorities was the most frequently cited as important (item , . %), followed by the information received on the crisis (item , . %), the capacity of public authorities to actually enforce protective measures (item , . %), the respondent's profession (item , . %) and his/her health condition (item , . %), the opinions expressed by social relations such as friends and relatives (item , . %) or colleagues (item , . %), and the political connotation of the relevant authorities (item , . %). later on, we will use regression analysis to investigate the extent by which these self-reported data correspond to the relative weight of the actual factors affecting participants' trust in institutions. for now, it appears that participants self-describe their theory of trust in fairly objective terms, giving priority to the factual nature of the proposed measures, the information they gathered (apparently with the exception of social channels; see below), and the extent by which public authority is able to enforce their recommendations; in contrast, relatively little weight is given to personal factors and social networks, and none at all to political partisanship. this last result suggests that the public response to the covid- crisis was initially perceived as a matter of shared concern of all political parties, which in turn prompted a temporary truce in the usual partisanship characteristic of italian politics; moreover, in their efforts to deal with the emergency, public authorities were regarded mostly for their institutional role, with little attention to their political affiliation (even when such authorities were the expression of certain political parties, as it was the case with the national government). this interpretation also helps to explain the extremely high level of trust in public institutions with respect to the covid- emergency during those few days, in a population well-known for its deep-seated distrust of politicians in general, and of political parties in particular: further analysis of this interesting anomaly will be presented in the section "discussion and conclusion." this section of the survey investigated both frequency of use (items - ) and perceived trustworthiness (items - and ) of various types of information sources in relation to the covid- pandemic, to get a better sense of what channels were most influential in affecting participants' opinions on this topic; in addition, we collected data on the trustworthiness directly assigned to public institutions as information sources (item ), which was high for . % of the sample, average for . %, and low only for . % of respondents. with respect to other information channels, the data summarized in table highlight four main findings: (i) official online channels, e.g., the website of the civil protection, and scientists are both frequently used and considered reliable as information sources; (ii) in contrast, traditional media, albeit often consulted, are regarded as reliable only by less than half of our sample; (iii) family physicians are in general considered trustworthy, yet they are rarely used as information sources; (iv), finally, both social relationships and unofficial online sources, e.g., social media, are neither frequently used, nor widely believed. the result on unofficial online channels is especially surprising: whereas the very low credibility associated to these sources is understandable and even commendable, the fact that only one respondent out of four admits to using them frequently is hard to swallow, especially at a time in which personal contact was severely limited in italy, thus making social media an even more attractive outlet for users. besides, recent national statistics on internet use in italy do not agree with the picture painted by these data: according to the global digital report , compiled annually by wearesocial and hootsuite, in , % of italian citizens were active social media users (with a growing trend with respect to ), and the average time spent on social media every day was a little less than h per person. besides social desirability effects (respondents may have been reluctant to admit gathering information via unofficial channels on such delicate topics), a possible explanation for this anomaly is in a common misperception of the role of social media as gatekeepers: someone who finds on facebook a link to an article on a traditional newspaper, or is made aware by a post on twitter of the latest press release on the official website of the civil protection, may be inclined to disregard the role of the social media in bringing these information to the user's attention. yet, this is how we use social media as information sources, often without even realizing it: we take advantage (or succumb, depending on the circumstances) of their agenda setting algorithms, which allow these platforms to act as powerful information brokers, rather than information producers. expectations on long-term impact on trust the final section of the survey intended to probe participants' expectations on the long-term impact of the covid- crisis on trust relationships between citizens and public institutions (item ), between citizens and the dominant economic model of development (item ), between citizens and the scientific community (item ), and among citizens as peers (item ). here, the big winner is expected to be science: . % of respondents believe that the current crisis will strengthen the trustworthiness of scientists as public figures. expectations on the impact of trust toward public institutions and among citizens are less triumphant, yet still positive: . % predict an increase in institutional trust after the covid- pandemic, whereas % make the same prediction with respect to social trust, i.e., trust among peers. finally, on future trust in the dominant model of economic development, our sample is evenly divided: % think that we will trust it more than before, . % are undecided, and . % expect an increase in distrust toward that model. as a preliminary step before running regression analyses, we used pca to identify strongly correlated items in the data set and simplify the variables' structure, in order to avoid multicollinearity issues in our regression models. since the survey was theoretically motivated by the socio-cognitive model of trust (castelfranchi and falcone, ) , we performed separate pca on subsets of items, to preserve relevant theory-based distinctions in the participants' responses. item , degree of trust toward public authorities in relation to the covid- pandemic, was not included in the pcas, since it was intended to act as the target of the regression models; we also excluded items (doubts on public authorities' competence) and (doubts on public authorities' intentions), since these were included in the survey merely as control questions for, respectively, items - and items - ; moreover, we kept separate from the pcas item (personal discomfort associated with public authorities' measures), item (usefulness of one's own personal contribution to the collective effort), and item (overall adequacy of public authorities' measures), since we wanted to test their role as individual predictors in the regression models; finally, item (expectation of very limited yet useful compliance by other citizens) was excluded for the pca and regression analysis, due to the ambiguity in its interpretation already mentioned in section "descriptive statistics." the remaining items led to the individuation of principal components, as summarized in table (full details on the pcas methods and results are provided in the supplementary materials). in order to be considered satisfactory, each pca had to explain at least % of the cumulative variance, and further components were added only if they improved by more than % the explained variance. in order to test our main hypotheses, we performed a multivariate regression model on raw data using ibm-spss software. the dependent variable to be predicted was the overall trust manifested by participants toward public authorities involved in the covid- response, i.e., item in the survey. after some explorative iterations and based on theoretical considerations, we decided to include independent variables in the final model: principal components identified via pcas (indicated with an asterisk in table ), individual items that were conceptually independent from the other sections of the survey (personal discomfort associated with public authorities' measures, usefulness of personal contribution to the collective effort, overall adequacy of public authorities' measures), and socio-demographic variables-age (coded as = - , = - , = - years of age), educational level (coded as = high school diploma or lower, = university degree or higher), region of residence (coded as = most affected regions, i.e., lombardy, emilia-romagna, veneto, marche, and piedmont, = all other regions), and time of data collection (coded as = before, = after the march new restrictions were announced). preliminary analyses indicated that the respondent's profession did not affect responses, so we excluded it from the model; as for gender, preliminary regressions showed no difference in the predictors of institutional trust between male and female respondents, so we excluded it from the final regression model and performed a separate set of analyses to assess its impact in our data (see section "gender effects"). we first run the regression analysis on the whole sample: the model had a good fit (r = . ) and explained % of the variance in the overall trust evaluation; out of independent variables were significantly correlated with trust (p < . ), and the most powerful predictors were positive indicators of competence of public authorities (β = . , p < . ), perceived adequacy of the adopted measures (β = . , p < . ), trustworthiness of official information sources (β = . , p < . ), public authorities' intention to ( ) the numbering used for items follows the order of presentation in the survey: the relevant items are from to , since item was the informed consent, whereas items - asked for demographic information. the asterisk (*) indicates principal components that were later used for regressions. contain the pandemic (β = . , p < . ), and perception that public authorities' efforts were focused on public safety, with no other agenda (β = . , p < . ). all other significant predictors had an absolute value of β equal to or lower than . . the non-significant predictors were personal discomfort due to the adopted measures, perceived usefulness of personal sacrifice, expectation of sufficient compliance (but notice that expectation of universal compliance was positively correlated with trust, whereas expectation of insufficient compliance was negatively correlated with it, both p < . , suggesting an "all or nothing" attitude toward compliance), individualistic reasons for compliance (while collectivist reasons for compliance were strongly and positively associated with trust, p < . ), educational level, time of data collection, and age (the last one showed a marginally significant negative correlation, β = − . , p = . ). we also applied the same regression model to subsets of participants, distinguishing first geographically (most afflicted regions vs. all other regions), then temporally (before and after the announcement of new restrictions by the italian government on march ), in order to detect differences in how trust was processed depending on the severity of the sanitary emergency in various areas, and the strictness of the measures implemented by public authorities while the pandemic was still progressing. we already knew from descriptive statistics that no overall change in trust toward public authorities was observed across these contexts, yet we wanted to probe for more subtle differences, e.g., different predictors of trust, or different contribution of the same predictors, depending on region of residence and time of data submission. all β and p-values for the various multiple regressions are reported in table ; in what follows, we will focus only on the most relevant results. applying the model only to participants from the most affected regions in italy at that time (lombardy, emilia-romagna, veneto, marche, and piedmont) revealed again a good fit (r = . ), explaining . % of variance in trust assessment; the same model also had a good fit when applied only to participants from all other italian regions (r = . , . % of explained variance). in both cases, the strongest predictors remained the same as in the whole sample, and also their order of importance was identical across regions, regardless of current outbreak severity (p < . for all the following predictors): positive indicators of competence (most affected: β = . ; other regions: β = . ), adequacy of the adopted measures (most affected: β = . ; other regions: β = . ), trustworthiness of official information sources (most affected: β = . ; other regions: β = . ), pa's intention to contain the pandemic (most affected: β = . ; other regions: β = . ), and perception that public authorities' efforts are focused on public safety, with no other agenda (most affected: β = . ; other regions: β = . ). in spite of the substantial similarity in how trust in public authorities was attributed by respondents in different areas of the country, some fine-grained distinctions emerge looking at those factors that were significant in one context but not in the other-and also exercising due caution, since a difference in significance does not necessarily imply a significant difference. in the most affected regions, we observed eight non-significant predictors, whereas there were only six in the other regions: four of these factors were irrelevant across both contexts (personal discomfort, perceived usefulness of the sacrifices, individualistic reasons for compliance, and time of data collection), whereas negative factors affecting competence of public authorities, intention to downplay the emergency, impact of personal effort, and trustworthiness of unofficial information sources were immaterial for respondents from the most affected areas, whereas they acted as significant predictors (albeit weak ones) for participants from other regions of italy; in contrast, an expectation of sufficient compliance from other people had a significant negative correlation with trust in the most affected regions (β = − . , p = . ), whereas it had a marginally significant positive correlation with it elsewhere (β = . , p = . ). taken together, these results suggest that participants living in areas that were currently experiencing very severe outbreaks of covid- had a more focused mindset when deciding whether to trust public authorities to deal with the emergency: less factors were considered relevant, and in particular, it was probably taken for granted that some inconsistency in public communication and intervention may occur, without necessarily jeopardizing trust (negative factors on competence), and that unofficial sources were not to be taken seriously when deciding whom to trust; at the same time, expecting that only a sufficient number of people would comply with the emergency measures had a negative impact on trust in public authorities, probably highlighting the fact that, in those regions, people believed that "enough is not enough"that is, either everybody cooperates in facing the crisis (universal compliance) or we will not be successful in overcoming it. this extreme mindset is confirmed by the fact that the relevance of one's own personal contribution did not affect trust attribution to public authorities in the most affected regions, whereas it did in other areas: this indicates again that collective compliance, not personal efforts, are perceived as the key to success by people currently facing the worst of the covid- pandemic. looking instead for short-term shifts in trust assessment over time, in relation to relevant public events (i.e., the introduction of new measures by the italian government on march ), we divided our sample based on time of data submission: before or after the public press release when the prime minister giuseppe conte announced the new restrictions to be implemented nationwide, to contain the covid- outbreak. the model performed well across both time windows (before: r = . , . % explained variance; after: r = . , . % explained variance) and the strongest predictors remained the same, as well as their relative order of importance (p < . for all the following predictors): positive factors affecting competence of public authorities (before: β = . ; after: β = . ), perceived adequacy of the adopted measures (before: β = . ; after: β = . ), trustworthiness of official information sources (before: β = . ; after: β = . ), attributing to public authorities the intention to contain the pandemic (before: β = . ; after: β = . ), and the perception that their efforts were focused on public safety, with no other agenda (before: β = . ; after: β = . ). again, we observed substantial stability over time in how trust in public authorities was attributed, with minor differences emerging only by comparing the significance and direction of some secondary variables. in general, the introduction of more severe restrictions had the effect of simplifying the metrics used to assess trust toward public authorities: before the march announcement, only four variables failed to correlate significantly with trust, whereas after it, the number of irrelevant predictors increased to , indicating a more narrowly focused mindset in assessing the trustworthiness of the institutions in charge of dealing with the emergency. in particular, intention to downplay the emergency, personal discomfort associated with the proposed measures, and trustworthiness of unofficial information sources became irrelevant for trust in public authorities; unfortunately, the expectation of universal compliance also became equally irrelevant (before: β = . , p = . ; after: β = − . , p = . ), while the negative correlation between expectation of insufficient compliance and trust was much stronger after the march announcement (before: β = − . , p = . ; after: β = − . , p = . ). this suggests a turn for the worst in people's expectations: before the new restrictions, trust was positively supported by expectation of universal compliance (the more i believe all others will behave responsibly, the more i trust the authorities), whereas after them, the influence of pessimistic fear became dominant (the more i doubt enough people will comply, the less i trust the authorities). as a possible reaction to this shift, it is worth noting that the positive correlation between impact of personal efforts in the covid- response and trust in public authorities became significant only after march (before: β = . , p = . ; after: β = . , p = . ), suggesting that the new measures strengthened in italian citizens a sense of personal responsibility for the collective reaction to the virus. finally, region of residence was a significant (albeit weak) predictor of trust before, but not after, the announcement of new restrictions by the italian government (before: β = − . , p < . ; after: β = − . , p = . ): this shows a stronger tendency to trust public authorities in the most affected regions before march , whereas this was no longer true after that date. since overall trust in public authorities did not decrease after march in the whole sample, this indicates a leveling in trust attribution across the country after the introduction of new measures, which in turn could be interpreted as a shift in the perception of the emergency: whereas in early march, a significant part of the italian population still believed the outbreak to be somehow contained to specific regions, and thus a local problem unlikely to affect everybody in the same way, the nationwide interventions announced on march made it crystal clear to all that covid- was indeed a national concern. overall, these regression analyses show that, in italy, trust in the capacity of public authorities to deal with the covid- emergency was attributed in a fairly consistent manner during the time window of this survey (march - , ) across different areas of the country, giving central prominence to positive indicators of competence in public institutions, assessing the adequacy of the proposed measures, verifying that proper intentions supported their application, and paying attention mostly to official information sources. all considered, this suggests a fairly reasonable and well-balanced judgment-making process for trust attribution, while the true anomaly remains the high levels of trust in public authorities recorded during the early stages of this emergency (see section "descriptive statistics"), which are in sharp contrast with both long-term trends and recent surveys on institutional trust in italy, prior to the covid- pandemic. at a more fine-grained level, region of residence and time of data completion did reveal some interesting shifts in trust assessment, yet these insights should be interpreted carefully, since they concern relatively minor changes in the significance of secondary predictors, within a regression model with a high number of independent variables. comparing male and female respondents, a χ test revealed a small but significant difference (p = . ) in institutional trust in relation to the covid- emergency: in particular, men were more likely to express high levels of trust toward public authorities involved in contrasting the outbreak ( . % men vs. . % women), whereas women were more often neutral ( . % women vs. . % men). running the regression model described in section "regression analyses" separately on male and female respondents showed that, although the main predictors remained the same (positive indicators of competence, adequacy of the measures, trustworthiness of official information sources, public intention to contain the pandemic, and institutional focus on public safety), age and region of residence were significant predictors only for women and not for men (age: women β = − . , p = . , men β = − . , p = . ; region: women β = − . , p < . , men β = . , p = . ). to further investigate this interaction between gender and other socio-demographic factors influencing institutional trust during the covid- emergency, we run a trivariate analysis on, respectively, gender × age × trust and gender × region × trust. the first analysis revealed that gender effects on institutional trust are significant (p = . ) only in the age range - years, which is also the most vulnerable to the virus: among respondents in this age range, the majority of those that expressed low levels of institutional trust were male ( %), whereas most of those neutral or highly trustful were female ( . and . %, respectively). it is also worth noting that, after performing a bivariate analysis on the impact of age on trust, we found a highly significant effect (p < . ), with . % of elderly respondents ( - years old) expressing high trust in public authorities, whereas this percentage drops to . % for participants in between and years of age: this further confirms the role of vulnerability to the covid- virus in eliciting higher attributions of trust, and it is consistent with previous findings on a negative correlation between age and willingness to comply with social distancing measures during the covid- pandemic (wirz et al., ) . the second analysis showed that the relationship between gender and institutional trust is significant (p = . ) only in those regions that were most affected by the covid- outbreak: in these areas, most of the respondents that manifested distrust in public authorities were men ( . %), while the majority of the neutral and trustful participants were women ( . and . %, respectively). taken together, these results suggest that, whenever the situation was most critical (i.e., for the most vulnerable age range and in the most affected regions), men were overrepresented in the (small) group of people expressing distrust toward public authorities, whereas women were overrepresented among those neutral or trustful. although this may suggest an interesting gender effect on resilience under extreme stress (women seem more likely than men to suspend judgment or look on the bright side, precisely when the situation is the most dire), it is worth noting that, regardless of gender, only a small minority of respondents were expressing distrust toward public authorities, even in the most affected age range (men . %, female . %) and in the most affected regions (men . %, female . %). thus, these gender effects invite further investigation, but on their own, they do not justify any hasty conclusion on how different genders may react against health emergencies. the most striking result of this survey is the very high level of institutional trust expressed by respondents: % of them trust italian public authorities to be able to deal with the covid- emergency. this is in sharp contrast with the relatively low levels of institutional trust characteristic of italy, both historically and in recent surveys: according to the demos & pi nd annual report on "the italians and the state" , based on a large representative sample (n = ) of italian citizens over years of age interviewed in december , only % respondents trusted the state, whereas both regional governments ( %), european union ( %), and municipal authorities ( %) fared better, while political parties were in the worst shape, with only italians out of willing to trust them; in fact, of the main national institutions, the only one with decent levels of trust was the presidency of the republic ( %, still in sharp decline with comparison to years before, in , when it was as high as %). also international estimates indicated relatively low levels of institutional trust: according to the eurofound report on eurofound ( ) , italians' trust in the national government has been declining in the last few decades and is now below %, while the more recent data of the eurispes report-italy , presented in february , indicated trust in institutions at . % ( . points lower than in ). institutional trust in italy in recent years is extremely weak not only in absolute terms but also in relation to other european countries: in their comparison of eu states, based on data from the european social survey, oksanen et al. ( ) reported very low levels of institutional trust in italy, measured by respondents' trust in five institutions (parliament, politicians, political parties, the police, and the legal system); in fact, only cyprus, poland, slovakia, and bulgaria expressed stronger institutional distrust than italy. moreover, this trend toward widespread distrust of public institutions is not a particularly recent feature of italian politics: while in recent decades, it developed mostly against the backdrop of increasing tensions between populist movements and traditional political parties (urbinati, ) , massive erosion of public confidence in political figures was already ongoing in italy well before the recent resurgence of populism worldwidein the last decade of the th century, following the corruption scandals of tangentopoli and its media resonance (giglioli, ; vannucci, ) , and with the largely failed shift toward bipolarism during the berlusconi age (viroli, ) . even before that, a longitudinal analysis reveals that the confidence gap between electors and political institutions, characteristic of many post-wwii democracies, appeared in italy much earlier than in other countries (segatti, )-so much so, that already in the s lapalombara ( ), a highly influential political scientist, described italians' attitudes toward politics with three emblematic words: alienation, fragmentation, and isolation. such a deeply rooted tradition of distrust in public institutions underscores the importance of the opposite trend registered in our survey, i.e., a sudden boost in institutional trust prompted by the covid- crisis-a significant result that is also supported by other data collected in this survey, as seen in the "results" section. moreover, insofar as this newfound trust is grounded on trust in the expertise of the scientific authorities involved, it is also at odds with the widespread anti-scientific sentiment considered to be on the rise at the global level, variously stigmatized as "the death of expertise" (nichols, ) and the crisis of epistemic deference (marconi, ) . surprising as it may be, there are several reasons to consider this finding on trust as reliable: (i) internal consistency: as discussed in section "results, " all other responses to the survey are consistent with a high attribution of trust to public authorities and indeed provide justification for such attribution. (ii) external validation: just a few days after data collection for this study was concluded, a survey on a representative sample of italian citizens (n = , - march ) was conducted by the independent research center demos & pi , providing substantial support to our main results: e.g., % trust both the italian government and the current prime minister, with % approval of the adopted measures, strong endorsement for the sanitary system ( %), the civil protection ( %), and the national government ( %), coupled with lower levels of confidence in political parties (none of them above % of approval) and a rising skepticism toward the european union ( % of respondents believe the italian response to the covid- emergency to be better than that of other eu countries, and only % consider the role played by the eu as positive in this crisis). (iii) low chances of social desirability effects: as demonstrated by the very low levels of institutional trust recorded in previous surveys, including recent ones, italians have no qualms expressing public distrust toward public authorities-quite the opposite, in fact. thus, there is no reason to assume that the current data on trust are inflated by social desirability effects. thus, there is a genuine phenomenon to be explained here: a veridical "trust boom" during the early stages of the covid- crisis in italy. the socio-cognitive theory of trust (castelfranchi and falcone, ) that inspired our survey provides the tools needed to craft a tentative interpretation of this remarkable fact, although the questionnaire itself was designed to record such a phenomenon, rather than explain it. thus, the speculative nature of our interpretation cannot be stressed enough: our study revealed a highly significant and surprising phenomenon, for which now we look for an explanation. the interpretation we favor is the one that, to the best of our knowledge, appears more adequate to account for the pattern of results obtained in this survey; later on, we will contrast it with other alternative explanations and argue in favor of its superiority. nonetheless, such interpretation remains tentative, and it is intended as a springboard and an inspiration for further studies that may either confirm or falsify it, rather than as something set in stone. with this in mind, let us focus on the fact that trust, at its cognitive core, entails the decision to delegate to someone else (the trustee) the realization of a goal that is important to the agent who is expressing trust (the trustor). as a result, being able to choose not to trust someone requires either having alternative means to achieve the desired goal (e.g., "i will do it myself " or "i will delegate it to someone else") or being ready to forsake that goal. however, neither of these options are available in the face of a pandemic: the relevant goal is personal and public safety, which is non-negotiable, i.e., it is not something we can decide to forget about, and the only course of action that offers reasonable chances of achieving it is to put our collective trust in public authorities, since there are no other available agencies we might appeal to (indeed, the only choice we have concerns the level of public authority we should confide in, and our sample clearly indicated the national level as the most pertinent one). in other words, a pandemic like covid- creates the preconditions for a collective case of necessary trust in public authorities, or institutional trust by force majeure: not in the sense that we are being manipulated by some hidden power, as some conspiracy theorists may be prone to believe, but because the very nature of the health crisis leaves us with no other option than to put our trust in public authorities (that is why we emphasize a need, a necessity for trust). it is worth noting that these pressures toward trust between citizens and public authorities in times of sanitary crisis are symmetrical: citizens have no alternatives to reliance in the relevant public institutions, yet these institutions themselves cannot help but trust in civic compliance to the proposed regulations, on pain of failure in containing the contagion, due to the limits of enforcement already emphasized in previous studies (siegrist and zingg, ; lewnard and lo, ; olsen and hjorth, ; van bavel et al., ) . necessary trust is a two-way street in health emergencies, for both citizens and public authorities. moreover, this two-way street is often cyclically traveled: in fact, the citizens themselves become fully aware (perceive the request and expectation) of the need for public authorities to receive the right degree of trust from citizens as a tool for achieving the common goal, and this awareness becomes one of the reasons for citizens to trust public authorities themselves. in other words, in the best-case scenario, this becomes a trust-based "alliance" toward a supreme common purpose. this civic alliance, or social pact, is grounded in a specific dynamic of trust: the trustor deliberately bestows trust on the trustee, even if partially skeptical of the trustee's qualities, in an attempt of motivating the trustee to "rise to the occasion" and become trustworthy. this is the sense in which trust breeds trust, as noted both by trust theorists (e.g., falcone and castelfranchi, b) and by political economists (e.g., feld and frey, ) . in the context of the early stages of the covid- pandemic in italy, we suggest that italian citizens put their trust in public authorities in charge of facing the crisis as a way of opening up a "trust credit line" and thus putting pressure on such authorities to prove themselves worthy of that credit. similarly, public authorities frequently manifested full trust in citizens' compliance with regulations (a topos often belabored on public occasions by all institutional actors, including the prime minister, the president of the republic, and representatives of the civil protection), precisely for the same reason: by declaring their trust in the common sense and civic responsibility of italian citizens, they were putting pressure on citizens to actually demonstrate such qualities. clearly, the objective need for trust created by a pandemic does not automatically evolve in greater trust toward public institutions. that need may find different outlets, so that other, bleaker outcomes may be equally possible: for instance, an already vulnerable trust relationship between citizens and public authorities may be shattered completely by a sudden crisis, especially if such crisis (or its poor management) are blamed on those authorities, possibly leading to a severe governmental crisis, and maybe even a takeover by authoritarian forces, or, in another scenario, public trust toward central authorities may dissolve, with citizens taking a turn toward tribalism and trying to face the crisis at the local level. while these options are certainly viable in general, our results suggest that neither of these paths was being seriously considered by most italian citizens in early march : our survey revealed a sudden increase of trust toward public institutions, rather than its collapse or further erosion, and that trust was directed toward national authorities, not toward specific charismatic leaders or local powers. according to our findings, faced with an unexpected need for public trust, the italian people in early march opted for putting their trust (at long last) in their elected representatives at the national level, rather than turning to authoritarian figures or local authorities for solutions. beyond the evidence of our data, how the management of the pandemic unfolded over those weeks provides further support to this interpretation. the italian government consistently acted as a mediator between all the social forces affected by the crisis, repeatedly demonstrating high reliance on the indications of the experts in crafting every containment measure: in short, the national authorities acted as the very antithesis of an authoritarian leader. at the same time, local authorities at all levels were relying on the guidance of the national government for facing the pandemic and, in some cases, were actively asking for its direct intervention to solve a crisis that they were not equipped to deal with; more generally, there was widespread consensus, both in political debate and in the media, on the need for a national response to the covid- emergency (a need well understood by our participants, as seen in the section "results"). again, an attitude that stands in sharp contrast with any shift toward tribalism. thus, assuming that the need for public trust prompted the high levels of institutional trust manifested by participants, we propose to interpret their other responses within the broad framework of motivated reasoning (kunda, ) and cognitive dissonance theory (festinger, ) : as the chosen path to pursue the paramount goal of personal and public safety, trusting public authorities became in turn a necessary instrumental goal, thus coloring all other attitudes expressed by the respondents; more precisely, it prompted them to actively look for reasons to justify their (unavoidable) trust in public authorities, in order to minimize cognitive dissonance. indeed, the need for trust experienced by italian citizens during the covid- emergency was at odds with their widespread attitude of distrust toward the very same public authorities they now needed to rely upon in the face of the outbreak: this, we argue, produced a massive and sudden shift in their perception of those public authorities, to better accommodate the new reality they had to deal with. in this interpretation, the trust boom observed in the survey was not produced by any collective epiphany on the actual qualities of the public institutions involved, but rather by a cognitive realignment of individual attributions to the current needs citizens were experiencing. all of a sudden, italian citizens found themselves pressured to rely on some key public authorities in ways and to a degree never experienced before, at least since the worst days of world war ii. regardless of how well these authorities behaved in the first stages of the covid- crisis, italians opted to re-frame their attributional states in a way that made this novel institutional trust justified, thus flipping the usual causal connection involved in acts of trust: it is not a case of detecting the appropriate qualities in public authorities and therefore deciding to trust them, but rather an instance of having first the need to trust those authorities and then justify such trust by assuming that these authorities would manifest the qualities required to warrant that trust. this is also justified and supported by the implicit pact with which public authorities communicated the need for this responsible and trusted attitude toward them as decisive for the achievement of the common purpose. it is worth noting that our reliance on motivated reasoning to explain some of these survey data is very different from the most common use of this notion in recent studies on public opinion: although originally conceived in much broader terms (kunda, ) , motivated reasoning in recent decades has become more and more associated with political ideology, with several studies investigating how partisan affiliations affect and filter our beliefs on matters of public interest (e.g., redlawsk, ; slothuus and de vreese, ; kahan, ; bolsen et al., ) . in fact, the same approach has been applied, with mixed results, to the public reaction to the covid- pandemic, e.g., looking at how political partisanship affected people's ability to discriminate between reliable information and fake news (pennycook et al., ) , timeliness in the adoption of restriction measures (rosenfeld, ) , health behaviors (kushner gadarian et al., ) , and compliance with social distancing guidelines (rothgerber et al., ) and stay-athome regulations (goldstein and wiedemann, ) . while the relevance of politically grounded motivated reasoning provides an interesting perspective on public opinion dynamics, other predictors have been found to be more relevant in explaining some of the target phenomena (e.g., fake news vulnerability, see pennycook and rand, ) ; more to the point, this is not the type of motivated reasoning we are discussing here. on the contrary, our data show no effect of political partisanship on trust attributions toward italian public authorities in charge of coordinating the covid- response, including those that did have a clear political connotation, e.g., the national government. instead, we appeal to the notion of motivated reasoning in relation to a manifestly non-partisan goal, i.e., public safety, and the related need to trust public authorities to be able to ensure such goal: this is the kind of motivated reasoning we argue influenced responses in our sample, independently from the political affiliation of either the survey participants or the relevant public authorities. alongside the preservation of consistency in citizens' beliefs toward public authorities, there is also another, more emotional path through which a need for trust may generate broader shifts in public perception. as noted by many trust theorists (luhmann, ; gambetta, ; batson, ; hardin, ) and also described in the socio-cognitive model adopted here (falcone and castelfranchi, a; castelfranchi and falcone, ) , a fundamental function of trust is to allow both individuals and groups to face uncertainty, to moderate it and deal with it. trusting someone or something immediately reduces the perception of risk; in this sense, trust offers the advantage of a subjective sense of safety, before and without being able to reach that safety objectively. it allows us to face the risk and take it, partially by giving us control over part of that risk, since trusting implies actively choosing to expose ourselves to a risk, i.e., the risk of having our trust betrayed (mayer et al., ) . this is why koller ( ) individuated risk as a key determinant of trust, in the sense that a risky situation may bias people toward trustworthiness when assessing potential allies in facing such risk: "to the degree that the individual fears the occurrence of an event of negative valence (...) he exaggerates the subjective probability of an event of positive valence, which implies that he expects the interaction partner to behave promotively" (koller, , p. ) . this is very much in line with the higher levels of trust we observed in the most vulnerable age groups and in the italian regions most affected by the covid- outbreak (see section "gender effects"). in the context of a health emergency such as the covid- pandemic, this subjective dimension of trust becomes particularly apparent: consider how physicians and nurses in italy turned overnight from marginalized workers in a distrusted field to the most revered national heroes. the individual and collective gain of this sudden change of perception is obvious: faced with the danger of contracting a deadly virus, the belief that your life will be in the hands of trusted professionals is incredibly valuable, not only for the unlucky few that will actually have to rely on those professionals, but for everybody, since it greatly helps in calming down their fear and anxiety. in this perspective, the trust boom recorded in our survey should be considered not only as a merely intellectualistic attitude but also as a response with deep emotional undertones: this is the type of trust that is not only cognitively justified, but also felt, insofar as it provides us with the calmness needed to remain productive under the extreme stress of a pandemic. it is worth noting that emphasizing the motivated nature of institutional trust during a pandemic is not the same as treating this newfound trust in italian public authorities as a fiction, just a desperate figment of the imagination of a population looking for solace from a terrible crisis. nothing could be farther from the truth: precisely because this institutional trust was experienced as a matter of necessity by the italian people, it is also genuinely (and dramatically) authentic. italian citizens, during those terrible days in early march , truly believed that public authorities would prove themselves worthy of their trust-possibly for the first time after many decades of increasing institutional distrust. yet, it is a very fragile belief, because it is massively based on assumptions: should the public authorities subsequently fail to prove themselves equal to the task at hand, this huge "trust credit" would come due, producing an even bigger backlash in terms of the gap between citizens and institutions. this would indicate the clear failure of an "alliance" in which citizens have invested their trust in public authorities. on the other hand, an actual demonstration of trustworthiness by the public authorities during the covid- emergency may engender a more durable and long overdue step change in institutional trust in italy. as the nobel prize joseph stiglitz put it in a recent interview to the italian newspaper la repubblica ( april ), we should "not waste this crisis, " since it opens up genuinely new opportunities for rethinking the fabric of our societies. what is more, respondents in our sample were fairly optimistic on the future of trust relationships with their institutions, with scientists, and among themselves, while expressing reservations on the adequacy of the current economic model (see section "descriptive statistics"). however, optimism is, by its very nature, a delicate thing, so the danger of experiencing a "trust crack" right after the initial trust boom is as real as ever. indeed, other ongoing research on the relationship between institutional trust and public response to the covid- emergency may invite a bleaker outlook on how things will unfold: in their comparison of data from european countries, oksanen et al. ( ) highlighted a negative correlation between institutional trust prior to the crisis and the delay in introducing restrictions to curtail contagion-the less trust was manifested in public authorities before the covid- outbreak, the more time passed after the first confirmed virus-related death and the introduction of containment measures. while we do not dispute the role of institutional trust as a protective factor against virus outbreaks (already well documented with ebola, see blair et al., ; vinck et al., ) , we are skeptical of the particular correlation observed by oksanen et al. ( ) , since it does not take into account the fact that different european countries were affected by the covid- outbreak at different times: in particular, italy, france, and spain [all "late intervention countries, " according to oksanen et al. ( ) ] were among the first countries to record severe outbreaks, and much of the measures later adopted by other countries were largely based on the evidence coming in from these first, unwilling testbeds for the public response to the virus. this is confirmed by the same data used by oksanen et al. ( ) : in terms of absolute dates, italy was among the first countries to endorse all the five types of interventions considered in their study, much earlier than many others that are instead regarded as "early adopters." moreover, the alleged correlation considers only the adoption of some form of interventions, without discriminating between countries that adopted all of them (like italy) or just a few, sometimes even only one (as in the case of sweden). this is probably why subsequent data do not seem to support the proposed correlation: for instance, sweden, one of the countries with one of the highest levels of institutional trust before covid- , as of may , has a very high ratio to the number of deaths per million inhabitants (among the top six nations in the world); similarly, belgium, where containment measures were adopted much more promptly than in italy according to oksanen et al. ( ) , in early may had the world's highest number of covid- confirmed deaths per million inhabitants. for all these reasons, we are not persuaded that prior institutional trust was the main factor determining timely adoption of containment measures by public authorities: while early intervention remains critical in facing virus outbreak, in the case of covid- , we believe that this was determined mostly by other factors, e.g., where the outbreak manifested sooner in europe. looking at the main predictors of trust highlighted by our regression analyses, respondents exhibited a matter-of-fact, evidence-based attributional strategy toward public authorities: consistently with the socio-cognitive model of castelfranchi and falcone ( ) , competence, intentionality, trustworthiness as information sources, and the perceived adequacy of the proposed interventions were the most relevant factors in justifying trust in public authorities. the relevance given to the role of public authorities as information sources is also consistent with the significant weight that information has in shaping participants' institutional trust, based both on their own self-report and on regression analysis (see sections "descriptive statistics" and "regression analyses"): this highlights the importance of feedback and control for trust. even when trust on public authorities is perceived as a necessity by citizens, they try to retain a measure of control over it, by monitoring the quality of institutional information channels. equally suggestive are some of the factors that failed to impact institutional trust in our sample: most notably, the amount of personal sacrifice imposed upon participants by the restrictions introduced by the government. significantly, this dimension did not affect citizens' trust in public authorities, contrary to what would be reasonable to expect under different circumstances: this, in turn, provides further support to our interpretation of the observed trust boom as a matter of necessity-insofar as public safety is the paramount goal, the severity of the necessary costs are immaterial in modulating institutional trust. this provides a nice illustration of the complex and context-dependent nature of feedback mechanisms on trust attributions: whether or not a certain observable feature of the situation (in this case, personal costs) will affect trust depends on its role within a broader attributional process, which cannot be oversimplified as a single feedback loop (for discussion, see falcone and castelfranchi, ) . finally, it is worth stressing that the main predictors of trust remained stable both geographically and temporally: nonetheless, controlling for region of residence allowed us to notice a more focused mindset for trust attribution in the most affected regions, whereas comparing responses before and after the new restrictions introduced in italy on march highlighted a leveling effect of these measures, which made us realize the national character of the covid- crisis to everybody, including citizens living in areas with only minor outbreaks. this last point underscores a common pattern to many of our main results: a shift from the particular to the general in how institutional trust is granted and justified by citizens, apparently caused by the unique circumstances of the covid- pandemic. as we discussed in section "results, " the responsibility of dealing with this emergency was clearly assigned to the national government, whereas regional and local authorities were perceived as marginal; moreover, high confidence was granted to public institutions, largely ignoring their political affiliation, unlike what happened in other countries, e.g., the united states (goldstein and wiedemann, ; kushner gadarian et al., ) , and without concern for any further agenda they might serve (in fact, trust in public authorities was paralleled by distrust in the various political parties, including those currently in power); consistently with this mindset, collectivistic reasons for institutional trust trumped individualistic concerns, and the perception of a common effort toward shared goals overshadowed any personal sacrifice that may be required to individuals and groups (this also relates to the fact that personal health itself obliges to look and reflect primarily on collective health, on which the former strictly depends); finally, confidence in each other's compliance with general rules was high, and the future outlook on trust was positive for public institutions, science, and civic society, not so much for the overall model of development. in short, participants responded to this survey not as individuals calculating trust based on likelihood of personal gains or losses (the standard economic view of trust), but rather as members of a collective subject, jointly engaged in facing a problematic situation. this tendency to make common cause against a shared concern is one of the most valuable assets any society can leverage to fight a public crisis, so in this sense, our data paint a positive picture of how italian citizens responded to the covid- emergency, as far as trust in public authorities is concerned. however, as repeatedly stressed above, this asset is also incredibly delicate, especially in a country with a complex and thorny history of institutional distrust, like italy. hence, a crucial research priority for future research, both in the short run and in the long term, is to keep monitoring how trust dynamics between citizens and public authorities will be affected by the next stages of the covid- pandemic: in fact, while our data suggest a generally positive reaction in the early phases of the emergency, they provide no guarantee of the fact that such trend will continue in the same direction. on the contrary, as mentioned, things could either turn for the best, as our respondents chose to believe, or turn for the worst, should public authorities fail to live up to their citizens' high expectations. all datasets generated for this study are included in the article/supplementary material. this study complied with all the ethical guidelines and standards for online surveys with human participants, in accordance with the local legislation and institutional requirements. the participants provided their written informed consent to participate in this study and were free to quit the survey at any time. rf led the design of the survey. ec, as, and sf performed data analysis. rf and fp wrote most of the manuscript. all authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication. evaluating covid- public health messaging in italy: self-reported compliance and growing mental health concerns the altruism question: towards a social social-psychological answer homophily and polarization in the age of misinformation public health and public trust: survey evidence from the ebola virus disease epidemic in liberia the influence of partisan motivated reasoning on public opinion trust theory: a socio-cognitive and computational model the mechanics of motivated reasoning societal change and trust in institutions scientific and non-scientific information in the uptake of health information: the case of ebola. south afr social trust: a cognitive approach the socio-cognitive dynamics of trust: does trust create trust? trust dynamics: how trust is influenced by direct experiences and by trust itself trust breeds trust: how taxpayers are treated a theory of cognitive dissonance trust: making and breaking cooperative relations political corruption and the media: the tangentopoli affair who do you trust? the consequences of partisanship and trust in government for public responsiveness to covid- trust and trustworthiness ideology, motivated reasoning, and cognitive reflection risk as a determinant of trust the case for motivated reasoning partisanship, health behavior, and policy attitudes in the early stages of the covid- italy: fragmentation, isolation and alienation scientific and ethical basis for socialdistancing interventions against covid- covid- : how to be careful with trust and expertise on social media trust: a mechanism for the reduction of social complexity fake news, the crisis of deference, and epistemic democracy an integrative model of organizational trust the death of expertise: the campaign against established knowledge and why it matters regulation and trust: a social science perspective on covid- predictors of attitudes and misperceptions about covid- in canada, the lazy, not biased: susceptibility to partisan fake news is better explained by lack of reasoning than by motivated reasoning modeling compliance with covid- prevention guidelines: the critical role of trust in science hot cognition or cool consideration? testing the effects of motivated reasoning on political decision making political ideology and the outbreak of covid- in the united states politicizing the covid- pandemic: ideological differences in adherence to social distancing italy, forty years of political disaffection: a longitudinal exploration the role of public trust during pandemics political parties, motivated reasoning, and issue framing effects me the people. how populism transforms democracy using social and behavioural science to support covid- pandemic response the controversial legacy of 'mani pulite': a critical analysis of italian corruption and anti-corruption policies institutional trust and misinformation in the response to the - ebola outbreak in north kivu. dr congo: a population-based survey the liberty of servants: berlusconi's italy self-reported compliance and attitudes about social distancing during the covid- outbreak the supplementary material for this article can be found online at: https://www.frontiersin.org/articles/ . /fpsyg. . /full#supplementary-material key: cord- -lt t shi authors: gilad, vered; masoero, giovanni title: treatment of covid- patients in italy: a physician’s experience and insights date: - - journal: rambam maimonides med j doi: . /rmmj. sha: doc_id: cord_uid: lt t shi the outbreak of coronavirus disease (covid- ) in italy, the first western country hit by the pandemic, seriously impacted the italian healthcare system and social and economic environment. this perspective piece focuses on the main challenges faced by italian hospital managements: hospital overcrowding; the need for urgent reorganization of the country’s healthcare systems; the lack of data regarding covid- diagnostics, clinical course, and effective treatment; individual and collective consequences of the crisis; and the importance of disease containment measures and early treatment strategies. february , marked the beginning of the worst healthcare system crisis of the last years in italy. when the first case of "communityacquired" coronavirus disease (covid- ) was diagnosed, italy-the first western country facing the new virus outbreak-was caught unprepared and did not entirely and properly foresee the subsequent events. the lack of understanding of the asymptomatic or pre-symptomatic carrier state of the virus, , together with an underestimation of its virulence, an exaggerated concern related to public opinion, and the potential for economic damage, led to the loss of precious time and widespread dissemination of the infection. written by two physicians working within an italian hospital, this paper provides an important perspective based on first-hand experience, regarding the challenges faced by italy's healthcare system during the covid- crisis. shortly after the first patient was diagnosed with covid- , italy witnessed a rapidly increasing number of emergency department referrals; many of the patients were in a critical condition, needing intensive treatment and mechanical ventilation. recognizing that the number of patients requiring hospitalization was much higher than hospital capacities led to a rapid reorganization of activity. all nonurgent and elective procedures were immediately suspended and routine outpatient clinic activity cancelled, leaving many chronic patients without adequate medical assistance. the number of dedicated covid- hospitalization departments grew rapidly, with some hospitals becoming fully dedicated to treatment of only new pandemic victims. within a few days, the number of hospital intensive care beds was at least doubled, operating rooms became small intensive care units (icus), and emergency departments were transformed into sub-icus, enabling non-invasive and mechanical ventilation of increasing numbers of patients. due to a shortage of healthcare workers (hcw) in many hospitals, specialists from different departments and resident doctors were transferred to covid- departments, and external specialists were hired to help with the growing workload. despite the eventual adoption of the above measures, in areas with early spread and high contagion rate of the virus, treatment was frequently inadequate due to extreme overcrowding. many hospitals had an insufficient number of icu beds due to the rising demand, forcing doctors to make difficult ethical decisions, and an unknown number of patients remained undertreated. many patients were sent home despite displaying evident signs of pneumonia, and many others died before reaching the hospitals. treatment delay led to advanced pulmonary and multi-organ damage and was often fatal, contributing to the high number of deaths in italy- , patients as of july , corresponding to about % of diagnosed cases. virus spread was highly variable among the regions in italy, resulting in significant divergence in covid- and overall mortality rates. for example, bergamo, a city where the virus spread widely in the early stage, had a prevalence of . % with a . % covid- mortality rate, and , total deaths between february and march , , corresponding to a mortality rate that increased by % compared to the same time period in previous years ( - ). , conversely, during the same time, rome, as a result of early lockdown, had a prevalence of . %, and significantly lower mortality rates, with covid- deaths ( . %) and an overall decrease in mortality (- %) compared to the same time period in - . , however, the tragedy extended far beyond numbers; owing to local precautionary isolation measures, patients often faced the disease completely alone. contact between cities and regions was limited, and many families remained separated for many weeks. throughout italy, during lockdown relatives were not allowed to enter the hospitals, and information about clinical status and communication with patients were possible only via telephone and electronic technology. families lost their loved ones without any possibility to say farewell, nor to even hold funerals, which were suspended until the beginning of may (and for which attendance was then strictly limited to only a small number of relatives, no more than people). the incidence of anxiety and stress disorders was high among hcw; a recent cross-sectional study reported a high prevalence of post-traumatic stress symptoms ( . %), depression ( . %), anxiety ( . %), and insomnia ( . %) among , italian hcw during the outbreak. incidences of psychological distress were higher among frontline hcw, similar to data published from china. the main factors associated with stress disorders included supporting the challengingly high number of critically ill patients, scarcity of intensive care beds, unfamiliar treatment strategies, feelings of inadequacy, uncertainty about pandemic duration, the risk of infection and deficiency of personal protective equipment, high-workload shifts, and physical distress related to the need to use heavy protective uniforms. [ ] [ ] [ ] contagion was high among hcw, especially during the first weeks of the outbreak, due to scarcity and non-optimal use of personal protective equipment. to date, more than , hcw have been infected and doctors have died, many of them working on the frontlines. moreover, in light of concerns regarding infection transmission to family members, many health workers decided to stay away from their homes, further complicating the challenges related to the lockdown, such as closed schools and separation from grandparents and other relatives. while the number of covid- patients was rapidly increasing, italy saw a drastic decrease in hospital admissions for other pathologies, since patients were afraid of contracting the infection at the hospital. recent italian reports document an almost % decrease in myocardial infarction admissions for acute coronary syndromes compared to equivalent periods in previous years, with a subsequent increase in severe cardiac complications and a -fold higher cardiovascular mortality rate. facing a novel virus, the primary challenges were related to the lack of data regarding diagnostic criteria, disease management, and treatment options. diagnostic challenges due to the limited availability of tests, slow processing of samples, and low reliability of some diagnostic kits created anxiety among patients and hcw. to avoid nosocomial spread of the virus, "grey" departments were established for patients with high clinical suspicion; nevertheless, it was almost impossible to prevent inhospital spread. given the limited clinical experience and lack of evidence-based data, new channels for relaying updates emerged; doctors who first started seeing covid- patients primarily shared their experience and findings through internet pathways. we followed daily updates by attending videoconferences and open webinars to discuss emerging findings, including new clinical signs, typical radiologic patterns, laboratory findings, common complications, and possible treatment strategies. additionally, open groups on the social networks and dedi-cated forums were created, reaching up to , doctors who used these platforms for consultation and exchange of essential real-time information. as clinical experience advanced, knowledge accumulated about the clinical course of covid- and the essential role played by cytokine storm and coagulopathy for disease progression and complications. [ ] [ ] [ ] given the novelty and the rapid spread of the virus in italy, most pharmacological treatments were initially based on compassionate and "offlabel" use, with wide heterogeneity between centers. as clinical evidence solidified, protocols became more uniform and included a combination of antimicrobial drugs (antiviral, antibacterial, and antimalarial), anti-inflammatory and immunomodulatory drugs, anticoagulants, and, later, convalescent plasma. a comprehensive management algorithm was recently published by galluccio et al., summarizing the timing indication for each treatment, based on clinical features and laboratory and imaging findings. although no targeted treatment is yet available, following the first weeks of outbreak it became clear that recognition of early "warning signs," and thus early treatment, could drastically reduce disease progression, complications, and mortality. in fact, many critically ill patients developed initial symptoms several days prior to hospital admission and exhibited tardive dyskinesia and sudden clinical deterioration associated with a hyper-inflammatory state. , furthermore, many patients presenting with only mild symptoms were found to have a significantly reduced oxygen saturation level and extensive pulmonary damage. in view of the importance of identifying patients in the early stages of disease, an open letter signed by more than , italian doctors was sent to the italian ministry of health, asking for a strengthening of community assistance and the telemedicine infrastructure to allow home surveillance and care of covid- patients. from the end of march , many italian regions started building dedicated medical teams (called "usca"-"unità speciale di continuità assistenziale"), assigned to monitor covid- patients at home through daily calls or home visits. they evaluated clinical parameters (e.g. symptoms, fever, hemodynamics, and neurologic state), and in selected patients laboratory exams and lung ultrasound, which led to hospital admissions earlier in the course of the disease, resulting in a significant decrease in the number of critically ill patients. in view of the concrete risk for a second wave, we believe that there should be more emphasis placed on community assistance, focusing on clinical surveillance and early pre-hospitalization treatment, in order to prevent repeated overcrowding of hospitals and healthcare system collapse. a schematic representation of the covid- management approach-initially, in the midst of the pandemic, and looking ahead-is shown in figure . the covid- outbreak led to an unprecedented health crisis in italy, as for the rest of the world, with severe consequences on personal and public health, substantial psychological, social, and economic repercussions, and considerable challenges for governments and hcw. although errors were undoubtedly made at the beginning of the outbreak that significantly contributed to extended illness and the high number of deaths, the italian government, the national healthcare system, and italian citizens have reacted with purpose and made important decisions that allowed italy to gradually recover from the severe crisis. the decision to impose extensive and national lockdown measures, despite the psychological, social, and economic impact, has permitted significant containment of the virus spread, preventing additional pressure on the healthcare system (which had already exceeded its maximum capacity in many places), and limiting further high morbidity rates and numerous deaths. moreover, primary importance is attributed to the cooperation of citizens in following the rules, fast reorganization of hospitals in response to the extraordinary workload, broad and real-time sharing of medical findings and updates, strong cooperation between doctors, and the community's sense of solidarity. in italy, there is still much to be done to prevent recurrences, and focus should be on preventing new outbreaks and reinforcing early treatment. social distancing measures, extensive testing, fast and comprehensive tracing of new cases, and isolation of covid- -positive patients in dedicated recovery centers can prevent further widespread diffusion of the virus and the need for more stringent actions. resources should be invested to strengthen the national healthcare system with an emphasis on community assistance, and digital and telemedicine infrastructures. we believe that it is essential to extend and reinforce outpatient care to prevent clin- * executed by dedicated medical teams, "usca-unità speciale di continuità assistenziale," and independently organized by the regional local health public body. ical deterioration of many patients, reduce hospital admissions, and avoid hospital overcrowding as well as the related consequences on the public healthcare system, non-covid- patients, and hcw. the covid- outbreak in italy demonstrated to the world the destructive effects that a highly contagious virus can have on unprepared healthcare and social systems. the high number of patients needing hospital treatment can lead to the overcrowding of hospitals and a very real consequent risk of inadequate treatment of both covid- and non-covid- patients. containment measures are fundamental to prevent disease dissemination, while monitoring and early treatment are essential to prevent complications and reduce mortality. italy introduced serious lockdown measures and is now gradually recovering, but the deep wounds incurred by the pandemic have also provided important lessons and insights. we hope that the lessons learned will contribute to the prevention of similar tragedies in the future. a novel coronavirus outbreak of global health concern presumed asymptomatic carrier transmission of covid- presymptomatic transmission of sars-cov- -singapore siaarti recommendations for the allocation of intensive care treatments in exceptional, resource-limited circumstances covid- integrated surveillance data in italy impact of the covid- epidemic on total mortality of the resident population first quarter covid- : daily update mental health outcomes among frontline and second-line health care workers during the coronavirus disease (covid- ) pandemic in italy mental health survey of medical staff in a tertiary infectious disease hospital for covid- the emotional impact of covid- : from medical staff to common people personal protective equipment and intensive care unit healthcare worker safety in the covid- era (ppe-safe): an international survey online ahead of print. crossref . editorial staff. list of doctors who died during the covid- epidemic reduced rate of hospital admissions for acs during covid- outbreak in northern italy reduction of hospitalizations for myocardial infarction in italy in the covid- era sars-cov- and covid- group for doctors only. private facebook group covid- illness in native and immunosuppressed states: a clinical-therapeutic staging proposal prevalence of venous thromboembolism in patients with severe novel coronavirus pneumonia isth interim guidance on recognition and management of coagulopathy in covid- a brief note on randomized controlled trials and compassionate/off-label use of drugs in the early phases of the covid- pandemic letter to the italian health ministry from , doctors key: cord- -c y rtb authors: chiara, berardi; marcello, antonini; mesfin g, genie; giovanni, cotugno; alessandro, lanteri; adrian, melia; francesco, paolucci title: the covid- pandemic in italy: policy and technology impact on health and non-health outcomes date: - - journal: health policy technol doi: . /j.hlpt. . . sha: doc_id: cord_uid: c y rtb italy was the first western country to experience a major coronavirus outbreak and consequently faced large-scale health and socio-economic challenges. the italian government enforced a wide set of homogeneous interventions nationally, despite the differing incidences of the virus throughout the country. objective: the paper aims to analyse the policies implemented by the government and their impact on health and non-health outcomes considering both scaling-up and scaling-down interventions. methods: to categorise the policy interventions, we rely on the comparative and conceptual framework developed by moy et al. ( ). we investigate the impact of policies on the daily reported number of deaths, case fatality rate, confirmation rate, intensive care unit saturation, and financial and job market indicators across the three major geographical areas of italy (north, centre, and south). qualitative and quantitative data are gathered from mixed sources: italian national and regional institutions, national health research and international organisations. our analysis contributes to the literature on the covid- pandemic by comparing policy interventions and their outcomes. results: our findings suggest that the strictness and timing of containment and prevention measures played a prominent role in tackling the pandemic, both from a health and economic perspective. technological interventions played a marginal role due to the inadequacy of protocols and the delay of their implementation. conclusions: future government interventions should be informed by evidence-based decision making to balance, the benefits arising from the timing and stringency of the interventions against the adverse social and economic cost, both in the short and long term. the novel coronavirus (covid- ) has been declared a global pandemic by the who, with countries registering coronavirus outbreaks. governments have put in place various interventions to respond to the rapid growth of infection and death ( ) which have had considerable negative impacts on society. with no covid- vaccine available, countries have been relying on non-pharmaceutical public health interventions ( ) . mitigation and containment strategies were targeted to flatten the contagion curve and reduce the rate of transmission, and to ensure the sustainability of health care systems in dealing with limited icu capacity and equipment. the international experience suggests that technologies such as contact tracing, drones and robots have played a crucial role in the fight against the virus in some countries, however the experience is mixed and their overall effectiveness is still uncertain. ( ) . italy was the first western country to experience the covid- emergency with a spiral of infections and deaths placing the country at the top of the international rankings, overtaking china on th march . the covid- burden has challenged the cost and sustainability of regional healthcare systems and the concomitant safety of healthcare professionals, requiring a + . % gdp increase in public health expenditure compared to the previous year for hospital reorganisation, community infrastructure, health personnel recruiting and equipment supply ( ). the incidence of the virus has been particularly severe in northern regions, moderate in central regions and mild in the southern regions of italy ( ) . the italian government implemented a wide range of measures to balance the complex trade-offs between ethical, public health, legal and economic problems. in the early phase of the epidemic, the italian government applied targeted measures to the most affected areas. as of th march , the policy interventions were extended homogeneously to all the regions despite the varied severity of the spread. the national exit strategy plan announced at the end of april began on th may with the gradual relaxation of containment measures carried out in three different phases. this paper aims to investigate and assess policy interventions implemented in italy and the impact on health and non-health outcomes. the literature offers different measures providing a systematic cross-country tracking of covid- policies ( , ( ) ( ) ( ) . our analysis considers a set of interventions with targeted objectives in the escalation and de-escalation phases across italian regions from nd january to nd august . the remainder of this paper is organised as follows. in section , we present an overview of the health profile of the population and health care system of italy. section presents the analysis of covid- epidemiological trends at national and regional levels. in section , we describe and analyse scaling-up and scaling-down policies implemented in italy based on a comparative conceptual framework ( ) . this section considers various interventions such as measures to contain the spread of the virus, policies for prevention and cure, interventions for economic stimulus, and the introduction of new health technology. section gives an overview of the response of the health care system. in section , we discuss the long-term challenges and spillover effects arising from the pandemic and associated government interventions. the final section presents the implications of our results for policy and draws conclusions. it is important to understand italy's demographic and epidemiological features to recognise the factors associated with covid- . with a population of over million and a surface area of over , km , italy is one of the largest and most populous countries in europe. it is a highly developed country ( ) , with the eighth largest economy in the world. the government is a parliamentary republic, with a multi-level governance system across twenty administrative regions and provinces and metropolitan cities. the italian constitution recognises health as a fundamental individual and collective right and stipulates that care should be guaranteed to disadvantaged people ( , ) . this is reflected in the italian healthcare system (servizio sanitario nazionale, or ssn), established in to provide universal coverage to all citizens, eu nationals, and legal residents. further, emergency and basic services are provided for undocumented immigrants. since , health policies and constitutional reform have driven a decentralisation of the ssn ( ) . the decentralisation is reflected in the financing, provision, and governance of the twenty regional health systems (see in table in the appendix). regional models range from integrated model to a quasi-market in lombardy ( ) . italian healthcare expenditure amounts to . % of gdp, on par with oecd average ( ) . the general budget is pooled nationally and distributed to the regions. in , government compulsory healthcare expenditure per capita was usd , (ppp), below the oecd average of usd , ( ) . across regions the healthcare expenditure per capita is heterogeneous, ranging from usd , in campania to usd , in friuli venezia giulia ( ). national government financing accounted for . % of total health spending in , while out-of-pocket payments for . % and voluntary schemes for the remaining . % ( ). arguably, having one of the best healthcare systems worldwide ( , ) , italy has the second highest life expectancy at birth ( . years) among european countries, and the eighth highest in the world ( ) ( ). as a result, its population (median age . years, % over years) is the oldest in europe and the second oldest in the world ( , ). italy's longevity is associated with high morbidity rates, with % of the total population having a chronic condition, and nearly % being affected by multi-chronic conditions ( ) ( ) . behavioural risk factors such as diet, tobacco smoking, high body mass index, alcohol consumption and low physical activity levels contribute to the italian population burden of disease ( ). empirical data confirms that age and morbidity are factors associated with covid- mortality ( ) (see table in the appendix). among the deaths recorded in the sample period, .% are older than years, and % had at least one underlying comorbidity/condition ( ) (see figure in the appendix). in the early stages of the pandemic, the incidence was higher amongst men, however, in april, the distribution evened out (see figure in the appendix), limiting the role sex plays in the incidence of mortality arising from the diseases. as of st july , % of confirmed cases were female (see figure in the appendix in italy, covid- data is made available by different institutions at national and regional levels. the inconsistency of data between different administrative levels has been a major issue (see table in the appendix). the italian government started to publish data on th february , with a reasonable degree of transparency, but only a moderate level of accessibility. important data such as icu survival rates, hospitalised patients' outcomes, number and occupation of new beds introduced since the emergency are still missing (see table in the appendix). this section describes the epidemiological trend of covid- throughout the country. covid- appeared in italy in late january , when two chinese tourists tested positive. one month later, patient was detected in lombardy. in the following days, lombardy and veneto became the two initial clusters of infection, experiencing a rapid escalation of cases. increased surveillance, through contact tracing and testing of both symptomatic and asymptomatic persons exposed to positive cases, revealed that the virus had already been spreading in many municipalities of southern lombardy since january ( ). the contagious nature of covid- caused cases to rapidly spread throughout the country ( ). nationally, the peak of contagion to compare the epidemiological figures at the regional level, we used the day each region's th case was confirmed as the start of that region's outbreak. similarly, for the numbers of deaths, we used the day the th fatality was recorded. we arbitrary set these starting points to reduce potential bias of the testing strategy and deaths recording at the beginning of the outbreak. figure in the appendix shows the distribution of covid- cases and fatalities in italy. lombardy was the most affected region followed by veneto, emilia romagna and piemonte. to compensate for the likely underestimation of cases due to the classification method and the testing strategy, the italian bureau of statistics (istat) compared the excess deaths recorded in the first four months of with the average number of deaths across all causes in the first four months of - ( ). this empirical analysis covers a sample of italian municipalities ( % in march, % in april and . % in may). as a result, the integrated surveillance indicates that % of excess mortality registered in march, % in april and . % in may can be attributed to covid- . the unexplained number of deaths might be attributed to three main causes: i) the higher mortality associated with the cases that were not tested; ii) indirect mortality in untested patients who died from organ dysfunctions possibly caused by covid- ; and iii) indirect mortality due to strains on the healthcare system in the most affected areas ( ). the italian government declared a state of emergency on st january . as the first western country to experience a major outbreak of covid- , italy was faced with escalating crisis in a period of extreme uncertainty. in the absence of a covid- vaccine, the only measures to contain the spread of the virus are case isolation, contact tracing and lockdown measures. the decentralised nature of the italian health care system combined with the heterogeneous epidemiological incidence at the regional level created the need for a diverse set of policies responsive to emerging patterns, rather than a one-size fits all approach. the policy interventions ( ) are categorised as follows: . policy interventions to contain the spread of the virus (behaviour, containment, mitigation) (see roadmap and table in the appendix); policy interventions for prevention and cure (treatments, health monitoring) (see roadmap and table in the appendix); technological interventions for testing, tracing and treating (see roadmap and table in the appendix); each policy categorisation has its own spectrum of escalating and de-escalating measures. the escalation implies the implementation of stricter or more invasive policy intervention, while the de-escalation implies the opposite. scaling-up and scaling-down interventions are ranked on an ordinal scale gradient that ranges from to , where policies are classified as none ( ); minimum ( ); medium ( ); significant ( ); very significant ( ) based on their significance and invasiveness. therefore, the upper extreme of the gradient gathers the implementation of all the other measures that belong to the lower levels of the spectrum. for instance, referring to the policy interventions to contain the spread of the virus, the significance gradient ranges from no restriction to enforced lockdown ( table in the appendix). looking at the technology interventions, the gradient measures the invasiveness ranging from no interventions to centralised gps contact tracing ( table in the appendix). the comparative and conceptual framework ( ) uses a systematic approach to categorise policies targeted to specific objectives. it is a tool to assess the impact of policies on different sets of health and non-health related outcomes. we focus our evaluation on the following outcomes: daily reported number of deaths; case fatality rate; confirmation rate; icu saturation; ftse mib index value; and unemployment rate. despite death data might be biased as mentioned above, daily reported number of deaths and case fatality rate are selected to represent epidemiological outcomes. such outcomes are preferred to the confirmed cases indicators as the number of fatalities is less likely to be underestimated due to the country testing strategy ( ) . furthermore, the case fatality rate provides an estimate of the daily severity of the disease over time. icu saturation reflects the response capacity of the healthcare system. the confirmation rate shows the testing strategy variation over time, ceteris paribus . the financial and economic indicators reflect the investors' expectations of the italian economy. the unemployment rate provides a measure on how the containment measures impact the job market and the productivity nationally. we break down the analysis at regional level for all of the outcomes except for the ftse mib index value and unemployment rate, which are at national level. the analysis does not include other relevant health and non-health outcomes due to data availability or inconsistency during the period considered (i.e. length of stay in hospital, readmission rate, icu death rate and survival rate, public and private health care expenditure, public and private bed and personnel repurposing, public and private service provision, domestic violence, mental health demand). as a robustness check, the outcomes considered are compared with the oxford covid- government response tracker that includes the response, stringency, containment and health and economic support indices ( ). we report an overview of the policy categorisation considering their major impact on some outcomes of interest. among the four policy categories mentioned above, categories and are combined as there is an overlap between the preventive measures and the technology development and utilisation. in the absence of a covid- vaccine, the introduction of health resources and technology plays a fundamental role in preventing the spread of the virus. roadmap reports the principal interventions adopted by the italian government relative to the triple ts strategy: testing, tracing and treatment. technological solutions using geolocation tools have been used with success to control the spread of the virus in china, singapore and south korea ( ) . the effectiveness of such technologies relies on wide adoption, however, one possible barrier to this is the perceived invasiveness and potential breaches of privacy ( , ) . during the outbreak, the ministry of health issued national guidelines for testing. the testing criteria were updated at later stages (see table in the appendix), adopting who and european commission recommendations. the strictness of the criteria reflected the necessity to ration the supply of swabs, reagents and laboratory capacity. despite the national guidelines, a homogeneous testing strategy has not been consistently applied over time, across regions, neither in terms of number nor modality ( ) (see figure in the appendix). amongst the three most affected regions, lombardy ran fewer tests than veneto and emilia despite the huge expansion of the digital health sector in italy (with % increase, i.e. . billion euro), digital health strategies are decentralised, resulting in inconsistent utilisation across different regions ( ) . the covid- pandemic and the related lockdown measures have led to unprecedented economic costs around the world. the pandemic is a global shock that has affected the international economy, from financial markets where asset prices have decreased and volatility has increased characterising both the impact and future uncertainty involved with the pandemic ( ), to the impacts on the supply-chain ( ) . decision-making necessary to prevent an economic collapse in such a context involves a trade-off between public health and economic prosperity ( , ) . using some micro and macro indicators, this section shows that italy has suffered roadmap describes the principal economic interventions implemented by the substantial economic losses. italian government and the european central bank. to prevent the economic collapse of the country, italy has implemented several fiscal policies. the two most significant policies were the "cura italia" decree implemented on th march and the "decreto liquidità" implemented on the th april . the "cura italia" brought an immediate tax boost of billion euro to help most affected sectors, strengthen the healthcare system and provide unemployment benefits. the "cura italia" decree was reinforced when the council of ministers approved the "decreto liquidità" (decree-law of th april , n. ), allocating a total of billion euro to assist businesses by offering loan guarantees and a certain targeted tax relief. for instance, it provided billion euro as credit for small and medium enterprises, and injecting liquidity into the banking system. furthermore, the decree earmarks billion euro to support exporting enterprises located in italy to access liquidity. the state and the export credit agency cover respectively % and % of the guarantee to support enterprises financial obligations ( ). despite the substantial fiscal stimulus, the country has experienced the biggest quarterly economic contraction since the financial crisis. according to the latest data provided by the italian bureau of statistics (istat) ( ), the gdp in the first quarter of decreased by . % in comparison to the first quarter of . the overall gdp contraction estimated for is a contraction of . % ( ). the crisis also impacted international trade flows. figure in the appendix shows the monthly index of import and export in millions of euro. in the quarter march-may , despite the growth in may, the economic trend is conditioned by a sharp downturn of the previous months and is largely negative for both exports and imports (respectively - . % and - . % compared to the previous quarter december -february ) ( ). according to the latest data provided by the italian bureau of statistics (istat) ( ), in may , exports record a marked decline on an annual basis (- . %), but with improvements compared to april (- . %), for both the non-eu area (- . % ) and the eu (- . %). compared to exports, the contraction in imports (- . %) is wider and summarizes the drops in purchases from both markets (- . % from non-eu countries, - . % from the eu area). in may , the trade balance is estimated to increase by million euro (from + , million in may to + , million in may ). net of energy products, the balance is + , million euro (it was + , million in may ). considering the domestic market, retail sales recorded a collapse for non-alimentary goods, partly offset by a marked increase in e-commerce (see figure in the appendix). among the non-alimentary goods, the large negative variations correspond to the clothing and fur sector, followed by goods such as games, footwear and travel items. pharmaceutical products also recorded a negative variation ( ).the negative variation recorded in these sectors is likely to impact the economic fabric of the country, mainly composed by small and medium enterprises with limited investments in digitalisation directed toward to the online market. the perceived trade-off between public health benefits and the economic impact seemed to cover a central role in the exit phase as well. until the th april , the government imposed a homogenous exit strategy. as the number of cases decreased, regional governors put pressure on central government to relax some restrictions on economic activities. after th may , the government's policy changed, leaving the exit strategy to be decided by each region. this choice implied a heterogenous re-opening of economic activities, which helped small and medium scale companies to re-start their businesses. after the lockdown, with the de-escalation measures, the government faced a crucial phase in terms of economic recovery. to invert the negative economic trend, the italian government announced a massive fiscal and monetary stimulus on th may . the decree "rilancio" allocated around billion euro in five main areas with the aim of reorganising the hospital network to deal with covid- emergency. additionally, it guaranteed liquidity and support for italian companies, aiding their stability during the emergency period and encouraging their revival at the time of recovery. this section describes the policy implemented by the government to cope with the limited capacity of the health care system and the challenges of the covid- pandemic. the central government is responsible for public health interventions; however, the decentralisation of the italian healthcare system hindered the implementation of a homogeneous strategy. regional health care systems differ widely in terms of hospital organisation (public versus private), equipment (number of beds etc.) and medical workforce ( ). following a decree implemented on st february , the government facilitated the urgent increase of hospital beds in all regions by % in icu and % in pulmonology and infectious disease wards. the measure entailed the immediate redistribution of hospitalised patients to accredited private structures to ease the pressure on the public system. the national health system is composed of % public and % private beds, with substantial regional variations, ranging from . % of public beds in lombardy to . % in basilicata ( ). increasing the number of icu beds appears to have largely prevented saturation, except for lombardy, which experienced an overloading of the system from st april (see figure in the appendix). the available data does not give further information on patient outcomes. patients' length of stay, discharge, re-admission and mortality rate data are necessary to fully evaluate the healthcare system performance and the health policies implemented by the government ( ). the overall standard national health budget increase for amounts to . billion euro with a decree of th march ( ). as part of this budget increase, the government spent million euro to implement the "aid distribution system", distributing disposable and durable medical materials to each region ( ). the most common disposable materials distributed were masks ( %), gloves ( %), and diagnostic kits ( %). durables materials included glasses ( %) and thermometers ( %). veneto received the highest amount of materials and molise the least. between th march and th april , the government also distributed , ventilators, of which % went to lombardy and % to emilia romagna. following a decree implemented on th march , the government committed million euro to hire , medical personnel on six-month contracts. regions autonomously managed this hiring process, making it hard to access the relevant data ( ). on th may, "decreto rilancio" allocates , million euro to national emergency fund and , million euro to strengthen emergency departments and community care ( ). this section has two aims. firstly, we test whether the escalation and de-escalation stringency measures to contain the spread of the virus were justified by the underlying epidemiological trend for all the regions, using the t-test analysis. our goal is to test whether differences in means of scaling-up and scaling-down policies are statistically significant. the daily death trend is chosen as the indication of the epidemiological trend. secondly, we describe through a graphical analysis how the health and non-health outcomes were impacted by the policies presented in the conceptual framework. the graphical analysis aims at evaluating if differences in the levels of the policies gradient have had a detectable impact on given outcomes across different areas of the country and may be assessed over different lags of time . in the absence of a counterfactual scenario, we run a t-test analysis on the mean of the daily number of deaths for each region throughout the period of each single policy (see table ). the analysis defines whether the difference in the daily number of deaths between each containment policies implemented in the escalation and de-escalation phase is statistically significant to justify the implementation of a more or less stringent policy. despite the death trend might be influenced by other policy interventions (such as increase of the icu capacity and more effective preventive method), it still is more reliable compared to other epidemiological measures. the analysis covers the period th february to th august . a value between (no intervention) and (very significant intervention) is assigned to each policy to represent its strictness (see table in the appendix). the policy classifications of lombardy, emilia romagna and veneto are displayed in three separate columns since targeted lockdown measurers were implemented before the national lockdown (see table in the appendix). overall, the escalation measures were found to be justified by the underlying death trend. considering the very that the enhanced testing capacity corresponds to the flattening of the case fatality rate and to a reduced confirmation rate (see figure in the appendix). in early march , significant containment interventions were required to ensure the sustainability of the italian healthcare system, especially in northern regions. the lockdown implemented on the th of march , and the closure of business activities of the nd march coincides with a decreasing trend in daily mortality, especially in northern regions in late march (see figure in the appendix). faster policies escalation in the epicentre of the pandemic might have resulted in a lower peak of deaths, flattening the contagion curve (see figure in the appendix). despite the substantial distribution of equipment throughout the regions, the icu wards were close to full capacity in the northern regions. although northern and central regions faced a similar increase in the saturation rate until th march , the lockdown timing seemed to be effective in the central and southern regions where the severity of the contagion was mitigated, starting to flattener before than in northern regions (see figure in the appendix). the case fatality rate stabilisation coincides with the government's announcement of the exit strategy at the end of april (see figure in the appendix). during the period considered in the analysis, the government did not invest resources for the development of tracing technology, which was instead developed for free by a private company. on th may , at the start of the exit phase, the government launched a seroprevalence study on a sample of , individuals. however, significant technological interventions seemed to be far from having any impact on the outcomes considered (daily number of reported deaths and icu saturation) due to delayed implementation (see figure in the appendix). the case fatality rate flattened, and the confirmation rate decreased even though minimal technological interventions were in place (see figure in the appendix). the impact of significant technological interventions could be better assessed if a second wave of covid- (or similar diseases) were to occur in the future. the stringency of the measures is negatively correlated with socio-economic factors. figure in the appendix shows an inverse relationship between the stringency of the containment measures and the stock market index value. figure in the appendix shows the daily performance of ftse mib and the response to major fiscal stimulus packages in italy. the period from february to mid-march saw some of the most significant daily drops in the performance of ftse mib index. following the two major decrees, "cura italia" and "decreto liquidità", it recorded an increase. in particular, the week beginning th march showed an increase in ftse mib (see figure in the appendix). since th may , the ftse mib index has steadily increased, in response to the stimulus and improving expectations surrounding the recovery effort associated with covid- . the lockdown and the subsequent closure of most activities also affected the job market although the unemployment rate did not entirely reflect the lockdown effect due to the reduction in the labour force that decreased by % in april compared to january decreasing ( ) (see figure in the appendix). the results displayed in this analysis are consistent with the oxford covid- government response tracker indices (see figures , , and in the appendix) ( ). covid- also had a huge impact on patients' access to health care, essential services, and education facilities. the high saturation rate in icu due to a large number of severe covid- cases caused a . % decline in organ donation. as of th march , schools and universities closed their facilities and began offering online classes. although online schooling may represent an effective means of education provision, access is dependent on the availability of internet connection and electronic equipment (i.e. computer, laptop, tablets). with schools remaining closed during the exit strategy, and concerns for a potential second wave in autumn, the inequality in access may persist, with potential long-term consequences. the outbreak of covid- significantly affected italy with severe health, social and economic consequences. the production of future government policy; with the transparency and ready availability of data essential. evidence-based interventions is relevant for reducing uncertainty around the interventions, thereby maximising the resource and investment allocations. a detailed appraisal of the data management system between regions and central government is missing and represents a limitation for further studies. the threat of future pandemics should drive the government's investments and resources to prevent and promote public health, strengthening community and territorial services, which demonstrate to be particularly successful in some regions to respond to health services organisation and delivery challenges. as far as the sustainability of the healthcare system is concerned, policymakers should focus on the elaboration of the promotion, prevention and early intervention framework to prevent suicide and lower the long-term impact on people's mental health due to isolation, social distancing and high stress levels. mental health programs should be targeted for different population groups, prioritising those at higher risk. moving forward governments need to identify and implement plans to mitigate the negative effects of a pandemic on vulnerable groups across society which includes elderly in the home care facilities, students, families with children and the impacted workforce. ischemic heart disease ( %) atrial fibrillation ( . %) chronic renal failure ( %) chronic obstructive pulmonary disease ( . %) dementia ( . %) active cancer in the past years ( . %) hearth failure ( . %) source: data provided by iss ( ) ( ) prime minister signed "phase " decree-starting from the th may. it has three phases . from the th may: ) parks reopening; ) free movement in the same region; ) free movement in different regions has to be justified by heath, work reasons; ) relatives visiting with personal protections; ) sport activities is allowed at meters social distancing form others; ) athletes training will be allowed for individual sports; ) funeral ceremonies open air: people maximum; ) bar and restaurants take away; ). restart of activities: manufacturing, building companies, transportations respecting security and hygienic -new security guidelines . from the th may: ) reopening of commercial activities, museums, libraries; ) team sport activities allowed. from the  programs of the biomedical and telemedicine sector,  strengthening of the national system of production of medical devices and  services aimed at the prevention of health emergencies / / significant ( ) decree "rilancio"  billion financing workers, firms, healthcare system, touristic sector  billion to finance the firms' debt ( ) variation in government responses to covid- . version blavatnik school of government working paper impact of non-pharmaceutical interventions (npis) to reduce covid mortality and healthcare demand the covid- pandemic two waves of technological responses in the european union. hague centre for strategic studies; . . camera dei deputati. misure sanitarie per fronteggiare l'emergenza coronavirus covid- in italy: actual infected population, testing strategy and imperfect compliance. . . international monetary fund. policy responce to covid- -policy tracker key country policy tracker categorising policy & technology interventions for a pandemic: a comparative and conceptual framework united nations development programme italy: health system review the italian healthcare system. thomson s et al international profiles of healthcare systems la politica sanitaria in italia: dalla riforma legislativa alla riforma costituzionale. institute of public policy and public choice-polis quasi-market and cost-containment in beveridge systems: the lombardy model of italy. health policy world health organization. the wold health report : health systems: improving performance most efficient healthcare world bank -world development indicators. life expectancy at birth, total (years) population structure and ageing patologie croniche in costante aumento in italia con incremento della spesa sanitaria. la cronicità non colpisce tutti allo stesso modo: si confermano le diseguaglianze di genere, territoriali, culturali e socio economiche geographical tracking and mapping of coronavirus disease covid- /severe acute respiratory syndrome coronavirus (sars-cov- ) epidemic and associated events around the world: how st century gis technologies are supporting the global fight against outbreaks and epidemics quantifying sars-cov- transmission suggests epidemic control with digital contact tracing on the responsible use of digital data to tackle the covid- pandemic lessons from italy's response to coronavirus: harvard business review coronavirus, si estende in emilia-romagna il test drive-through: tamponi direttamente dall'auto covid- italia -monitoraggio della situazione survey nazionale sul contagio covid- nelle strutture residenziali e sociosanitarie the unprecedented stock market impact of covid- (no. w ) the covid- shock to supply chains. the university of melbourne pandemics depress the economy gianluca public health interventions and economic growth: revisiting the spanish flu evidence. ssrn comparison of the icu beds saturation rate with the capacity before and after the covid- note: the red line coincides with the total saturation of the icu capacity ( %) in the region considered. the x-axis reports the saturation rate = %; = %; = %. source: personal elaboration of data provided by minsitero della salute and protezione civile ( ) source: data provided by ministry of health technology intervention gradient and case fatality rate (panel a), confirmation rate (panel b), daily number of reported deaths (panel c) . patients become asymptomatic due to symptoms resolution . patient is negative to sars-cov- test clearance . rna and sars-cov- absence in body fluids for patient that presented symptoms and for those who did not present any symptoms . for asymptomatic patients that resulted positive, the test should not be repeated before days after the first positive diagnosis . two molecular tests need to be performed in hours. both of them have to be negative http://www.salute.gov.it/portale/nuovocoronaviru s/dettaglionotizienuovocoronavirus.jsp?lingua=i taliano&menu=notizie&p=dalministero&id= key: cord- -f yglaz authors: forte, giuseppe; favieri, francesca; tambelli, renata; casagrande, maria title: the enemy which sealed the world: effects of covid- diffusion on the psychological state of the italian population date: - - journal: j clin med doi: . /jcm sha: doc_id: cord_uid: f yglaz background: starting from the first months of , worldwide population has been facing the covid- pandemic. many nations, including italy, took extreme actions to reduce the diffusion of the virus, profoundly changing lifestyles. the italians have been faced with both the fear of contracting the infection and the consequences of enforcing social distancing. this study was aimed to understand the psychological impact of the covid- outbreak and the psychopathological outcomes related to the first phase of this emergency. methods: the study included respondents. an online survey collected information on socio-demographic variables, history of direct or indirect contact with covid- , and additional information concerning the covid- emergency. moreover, psychopathological symptoms such as anxiety, mood alterations and post-traumatic symptomatology were assessed. results: the results revealed that respectively . %, . % and . % of respondents reported levels of general psychopathological symptomatology, anxiety, and ptsd symptoms over the cut-off scores. furthermore, a significant worsening of mood has emerged. being a female or under the age of years, having had direct contact with people infected by the covid- , and experiencing uncertainty about the risk of contagion represent risk factors for psychological distress. conclusions: our findings indicate that the first weeks of the covid- pandemic appear to impact not only on physical health but also on psychological well-being. although these results need to be considered with caution being based on self-reported data collected at the beginning of this emergency, they should be used as a starting point for further studies aimed to develop interventions to minimize both the brief and long-term psychological consequences of the covid- pandemic. in december , an outbreak of pneumonia associated with a new coronavirus (i.e., severe acute respiratory syndrome due to coronavirus (sars-cov- )) was reported in wuhan, china. in the following weeks, the infection attracted worldwide attention for its rapid and exponential diffusion across different countries around the world. on february , who named it coronavirus disease (covid- ) [ ] . at the beginning of april , covid- has infected more than one and a half million people, causing over , deaths in countries [ ] . this viral infection spread quickly, becoming unstoppable, and forcing the who to declare it a pandemic [ ] . although the containment measures a web-based cross-sectional survey, implemented using the kobo toolbox platform and broadcasted through mainstream social-media (such as facebook, twitter, instagram, telegram), was used to collect data among the italian speaking population. in our opinion, this procedure represents the best data collection strategy in the present phase of forced social distancing, and it leads to reaching the largest number of people. the survey was carried out from march to march . a brief presentation informed the participants about the aims of the study, and electronic informed consent was requested from each participant before starting the investigation. the survey took approximately min to complete. when the participants' responses to the survey lasted less than min or more than min, data were excluded to ensure a standard quality of questionnaires. participation was entirely voluntary and free of charge. to guarantee anonymity, no personal data, which could allow the identification of participants, was collected. for the current research, being at least years old was the only inclusion criterion employed. after a short demographic questionnaire, the participants answered questions that assessed knowledge and perceptions related to the spread of covid- and the government measures adopted to contain it. finally, italian versions of standardized questionnaires were administered to assess psychological dimensions. this study was conducted in accordance with the declaration of helsinki and was approved by the ethics committee of the department of dynamic and clinical psychology of the "sapienza" university of rome (protocol number: ). participants could withdraw from the study at any time without providing any justification, and the data were not saved. only the questionnaire data that had a complete set of answers were considered. ninety-eight per cent of the total respondents ( out of people) who started the questionnaires completed the entire survey, and the related data were considered for statistical analyses. the main demographic characteristics of the sample are shown in table . the authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the helsinki declaration of , as revised in . the first session of this questionnaire required information about gender, age, education and occupation, city, and region of origin. the second section aimed to evaluate personal knowledge about covid- diffusion, individual perception of the situation, and lifestyle changes related to government restrictions. the scl- [ ] (italian version: ) is a -items questionnaire aimed to assess psychological distress and symptomatology. the items are rated on a five-point likert scale, ranging from 'not at all' ( ) to 'extremely' ( ). ten primary symptom dimensions are measured: somatization, obsessive-compulsive, interpersonal sensitivity, depression, anxiety, anger-hostility, phobic anxiety, paranoid ideation, psychoticism, and sleep disturbance. a global severity index provides measures of overall psychological distress. higher scores in each dimension indicate greater distress and psychopathological symptomatology. a cut-off score of . was selected to define higher psychopathological symptomatology, in line with previous studies on the general italian population [ , ] . the internal consistency in the participants of the present study was α = . . the stai measures state and trait anxiety [ ] (italian version: ). the questionnaire includes items. twenty items refer to state anxiety (stai-s) and evaluate how participants feel about anxiety "right now, at this moment"; items refer to trait anxiety (stai-t) and assess how people "generally feel" about anxiety. the items are rated on a four-point likert scale, ranging from (not at all) to (very much so). in both the state and trait anxiety scales, higher scores indicate greater anxiety levels. a cut-off point of was used to define higher state anxiety, according to kvaal et al. [ ] . although this study was interested in assessing state anxiety, trait anxiety was also measured to check whether the anxious state could be explained by a high anxious trait of the italian population. the internal consistency of stai in the sample of this study was adequate (α = . ). fifteen mood aspects (insecurity, helplessness, sadness, fear, anger, frustration, stress, anxiety, depression, boredom, serenity, happiness, preoccupation, tranquility, energy) both positive and negative were assessed to examine the emotional impact of the current situation. in these evaluations, the participant was required to refer to two different periods. the first was december, preceding the outbreak of the contagion (december ); the second period referred to the last week. the mood scales required a response on a -point likert scale [ ] , from (not at all) to (very much). the use of mood scales has mainly been adopted to analyse the self-reported conditions of individual mood [ ] [ ] [ ] . the items on the mood scales presented high internal consistency (α = . ). the ies-r is a self-report measure designed to assess ptsd symptomatology according to the diagnostic and statistical manual of mental disorders-fourth version (dsm-iv) criteria for ptsd. the questionnaire requires the indication of the magnitude of distress on specific dimensions (e.g., recurring dreams, feelings of anger and irritability) related to specific life events (i.e., the current covid- emergency) referring to the last seven days [ ] (italian version: ). the three subscales measure avoidance (the tendency to avoid thoughts or reminders about the incident), intrusion (difficulty in staying asleep, dissociative experiences similar to flashbacks), and hyperarousal (irritated feeling, angry, difficulty in sleep onset). the ies-r requires a response on a -point likert-scale, from (not at all) to (extremely). the score on an ies-r subscale is the mean of the scores of the items of that cluster. the ies-r also gives an overall score (ies-r total that is the sum of the scores of the three subscales). the cut-off of was adopted to indicate a high risk of ptsd symptomatology [ , ] . in the present sample, the ies-r presented high internal consistency (α = . ). descriptive analyses were conducted to describe demographic characteristics, and covid- related aspects in the italian population, considering the different italian territorial areas. student's t-test was performed to compare our data on anxiety, general psychological symptomatology, and ptsd symptomatology with data from the general italian population, reported by previous studies. specifically, our data on anxiety were compared with those reported by corno et al. [ ] , scl- outcomes were compared with the data given by holi et al. [ ] , and ptsd indices were compared with the results of ashbaugh et al. [ ] . analyses of variance (anovas) were performed to explore the potential difference in the impact of covid- in the italian territorial areas. the differences between north italy, central italy, and south italy were reported for state and trait anxiety, psychopathological symptomatology (somatization, obsessive-compulsive, interpersonal sensitivity, depression, anxiety, anger-hostility, phobic anxiety, paranoid ideation, psychoticism, and sleep disturbance), and ptsd symptomatology (ies-r). furthermore, within-subjects anova designs were adopted to compare the respondents' self-reporting mood before and during the covid- emergency. logistic regressions were performed to explore the influence of demographic factors and experiences which were covid- related in determining risk for state anxiety (stai), psychopathological symptoms (scl- ), and ptsd symptomatology (ies-r). all data were analyzed using statistical package for social sciences (spss) version . and statistica . (statsoft.inc., tulsa, ok, usa). p-values of less than . were considered statistically significant. to better control the results for the multiple comparison analyses, the bonferroni correction was adopted; in these cases, an adjusted p-value of less than . was considered statistically significant. the characteristics of the respondents are shown in table . two thousand two hundred ninety-one individuals completed the questionnaires, ( . %) were males, and ( . %) were females; the mean age of the participants was . years (sd: . years; age range: - ). the most represented age range was - years ( . %). most of the participants ( ; . %) received a high school education and were students ( ; . %) or employees ( ; . %). the respondents' current locations were sorted considering territorial area: north ( . %), central ( . %), and south ( . %) of italy. most of the participants live in urban areas ( ; . %) with a number of inhabitants between , and , . among all respondents, only ( . %) were infected by the covid- , and ( . %) were sure that they had had close contacts with individuals suspected of covid- infection (see table ). of the overall sample, respondents ( . %) and ( . %) respectively knew people dead and patients in intensive care units (icu) because of covid- infection. comparisons of state and trait anxiety, psychopathological symptomatology, and post-traumatic symptomatology during the covid- epidemic were made with data from the general population. the comparisons of psychological outcomes during the covid- epidemic in the italian population with data from the general population are presented in table . considering scl- indices, depression (t = . ; p < . ), anxiety (t = . ; p < . ), anger-hostility (t = . ; p < . ), phobic anxiety (t = . ; p < . ), psychoticism (t = . ; p < . ), and global severity index (t = . ; p < . ) significantly differ from holy's data [ ] , indicating greater psychopathological symptomatology in our sample. considering stai indices, state anxiety appears to be higher in our sample compared to data reported by corno et al. [ ] in an italian sample that considered the levels of anxiety separately in both males and females (males: t = . ; p < . ; females: t = . ; p < . ), while no significant differences were present considering trait anxiety. finally, ptsd related symptomatology assessed by the ies-r resulted higher in our sample compared to the data reported by ashbaugh et al. [ ] (t = . ; p < . ) (see table ). table reports the differences in psychological outcomes, considering the three territorial areas of italy. considering psychopathological symptomatology assessed by the scl- , significant differences were reported only in the sleep disturbance subscale (f , = . ; p < . ; pη = . ). people from north italy reported higher sleep disturbances compared to people from south italy (p < . ). however, no other significant differences were observed (see table ). anovas on stai subscales did not highlight significant differences between individuals from north, central, and south italy. finally, considering ptsd, no significant differences were reported in ies-r subscales (see table ). the results on the difference in subjective mood before and during the covid- epidemic are shown in table and figure . the analyses confirmed for all dimensions a perceived worsening of mood by the respondents. figure shows the prevalence of psychopathological symptomatology, state of anxiety, and ptsd, stratified by gender, age, territorial areas, knowledge of people affected by covid- , and loneliness in social distancing experience. table ). sudden outbreak events always pose huge challenges to the countries where they occur, impacting not only on physical health but also on social and mental well-being. from this perspective, the covid- pandemic will have long-term consequences, influencing international and national public health policies. this study is part of a series of works aimed at investigating the characteristics and the psychological effects of the covid- pandemic and the restrictive measures adopted by the italian government during the early and more severe stages of the covid- outbreak [ , ] . since the outbreak of the covid- epidemic, the italian government imposed a lockdown in north italy, expanding it nationwide following the exponential diffusion of the pandemic from the northern territorial areas to both the central and south areas. these severe limitations included the request for both people infected by the virus and healthy citizens to isolate themselves at home, prohibiting all other than indispensable activities, and making it mandatory to wear surgical masks to enter public places. our data were collected near the infection peak (between the end of march and the beginning of april ) [ ] , and they provide an accurate snapshot of italians' perception of this emergency. this study delivers further information to add to the findings reported on the chinese population that was the first to be severely affected by ] , indicating that the effects of this pandemic on the psychopathological conditions are similar in the italian and chinese populations. in both countries younger age, student status, female gender and direct contact with covid- infection are associated with a greater psychological impact of the emergency, involving many psychopathological dimensions (e.g., anxiety, distress, sleep disturbance) [ ] [ ] [ ] [ ] [ ] ] . one of the aims of the study was to analyse the psychological impact of the covid- outbreak in the different italian territorial areas. north italy was the first area in italy infected by the covid- and in which social distancing was imposed. it continues to have the highest prevalence of contagion and deaths, with a heavy burden on the public health system. accordingly, we expected an impact of these conditions on the psychological well-being and mental health of its inhabitants. however, although respondents from north italy reported more sleep disturbances and a relatively higher state of anxiety compared to those from central and south italy, no other differences were observed in psychopathological symptoms and ptsd risk [ ] . these results would seem to underline that psychological status is not only influenced by the direct effects of a justifiable fear of contagion but also by the indirect consequences of the covid- outbreak such as the restrictive measures, that equally influenced people of all the italian regions, generating a similar psychological pattern. this assumption would be confirmed by the comparison of our results with data from the general italian population. the differences in the selection of the sample do not allow a generalizability of these results. most of the psychological symptoms assessed by the scl- subscales are significantly higher in our sample compared to data from the general population. only somatization and paranoid ideation resulted in being not significantly different from data on the general population. these last findings do not agree with recent data on the chinese population [ ] , and they could appear incongruous because medical emergencies might induce higher somatization and intrusive and threatening thoughts. however, these results concord with those found during the sars epidemic [ ] . the high prevalence of anxiety evidenced in our sample highlights that the covid- pandemic has increased alert levels and generated a high level of state anxiety in the population, confirming results of previous studies on sars, influenza a virus subtype h n [ ] [ ] [ ] , and covid- [ ] [ ] [ ] . in our sample, . % of the respondents presented ptsd symptomatology, and risk of ptsd higher than that reported in the general population, at least as regards the symptoms evaluated with the ies-r questionnaire [ ] . this result should be interpreted with caution because it referred to the first weeks of the emergency when people could perceive the rapid spread of the virus and the extraordinary measures adopted by the government as sudden stressors, and it is known that sudden stressors affect the daily lives of individuals drastically. on the other hand, this first italian perception of the current situation would seem to give a photograph of the real impact of the covid- outbreak on mental health. another interesting result concerns the impact of the pandemic on mood. respondents perceived a significant change in their mood, with a sensitive decrease of positive mood (e.g., happiness, serenity) and a high increase of negative mood (e.g., sadness, preoccupation, boredom) after the covid- spread and the consequent social distancing measures. from a clinical point of view, this result could suggest a possible risk of mood disorders, such as depression, as long-term consequences of a pandemic [ ] . however, it must be underlined that these data are not obtained prospectively, and the causal relationship cannot be confirmed. self-reported moods are subject to memory distortions and bias, and they should be taken with caution. overall, the results highlighted high levels of anxiety, psychopathological symptoms and ptsd symptoms in italian respondents during the first critical phase of the spread of the covid- pandemic and of the government measures taken to contain it. however, the results of the present study also suggested which people are most vulnerable to the psychological consequences of the covid- outbreak. this unexpected situation seems to have had a higher impact on females and people under years. moreover, to have had direct contact with people infected by the virus, and to know people more or less severely infected by the covid- (i.e., people hospitalized in an intensive care unit or people dying as consequences of covid- infection) emerged as other relevant risk factors for psychological well-being. all these characteristics would make people more vulnerable to developing anxiety, psychopathological symptoms, and ptsd-related symptoms, confirming results observed in previous studies [ , ] . these risk factors may depend on different aspects of the covid- pandemic. the high psychopathological risk related to direct experience with the covid- infection could depend on the fear of contagion, while being younger could be a risk factor due to the sense of constraint caused by social distancing and the other measures taken by the italian government [ ]. our study reports that covid- infected . % of the sample. this result is higher than the data on the general italian population ( . %), updated on the march [ ] , but it indicates the high rate of healthy individuals in the sample. both this consideration and the data on risk factors would confirm that, even without real exposure to the covid- and an actual infection, fighting against an invisible enemy could affect mental health. uncertainty, fear about infection and social consequences of a pandemic could be triggers for psychopathological symptoms, and they should be considered in further studies. although some psychological characteristics are linked to medical conditions [ ] [ ] [ ] [ ] , psychological consequences of at-risk people are often overlooked during an epidemic emergency as reported for sars and h n [ , , ] . once again, the importance of not disregarding mental health and intervening during and after the pandemic emergency in the most affected psychological dimensions appear relevant in a long-term perspective. this study gives a picture of the psychological well-being of the italian population at the beginning of the covid- emergency. however, some limitations must be considered. despite the large sample size, it is not possible to overcome the limitation of a cross-sectional study, which does not allow us to determine a causal relationship between the variables. also, the use of an online survey presents other limitations. selection bias of participant recruitment is a consequence of this methodological choice. this bias is expressed by some characteristics of our sample, such as the higher number of respondents younger than years, and the high number of females and people from south italy. another limit related to the online survey can be associated with convenience sampling that may have induced the collection of responses primarily from people who feel strongly about the considered issue. these limitations reduce the representativeness of our findings and may have influenced the results of the study. therefore, they must be considered. however, the adoption of an online survey was the best solution in this emergency in which social distancing measures limit data collection. in conclusion, a global response is desperately needed to prepare health systems to face the new challenge of the covid- outbreak. despite the underlined limitations, these preliminary findings, in line with the results of previous studies, evidenced that the diffusion of this pandemic can be related to anxiety, changes in mood, high psychopathological symptomatology, and could be associated with the development of ptsd. moreover, similarly to the results of other studies on the covid- pandemic, these findings should be considered preliminary, but they can be useful to predispose interventions aimed at improving the psychological conditions of the population. generally, there is still a lack of relevant research on psychological aspects during the covid- epidemic. it would be essential to analyse further psychological dimensions related to the covid- outcomes, such as lifestyle changes, fear, and perception of the emergency, to assess their role in influencing the psychological status of the italian population. we hope that these preliminary data can be useful to other researchers in analysing the impact of the infection and social isolation due to covid- diffusion. it is our desire that covid- be defeated but also that the research on this topic grows so that we can start thinking about the mental health of those involved in this severe emergency. coronavirus disease (covid- ) situation reports the psychological impact of quarantine and how to reduce it: rapid review of the evidence mental health outcomes of quarantine and isolation for infection prevention: a systematic umbrella review of the global evidence the psychological impact of the covid- epidemic on college students in china prevalence and risk factors of acute posttraumatic stress symptoms during the covid- outbreak in wuhan generalized anxiety disorder, depressive symptoms and sleep quality during covid- epidemic in china: a web-based cross-sectional survey factors associated with mental health outcomes among health care workers exposed to coronavirus disease confirmation of the dimensional structure of the scl- : a study in construct validation comparison of the ghq- , the ghq- and the scl- as psychiatric screening instruments in the finnish population manual for the state-trait anxiety scale the spielberger state-trait anxiety inventory (stai): the state scale in detecting mental disorders in geriatric patients reliability and validity of survey scales sleepiness: evaluating and quantifying methods validation and findings comparing vas vs. likert scales for psychosocial measurements the impact of event scale-revised the impact of event scale: validation of an italian version psychometric properties of the impact of event scale-revised assessing positive and negative experiences: validation of a new measure of well-being in an italian population psychometric validation of the english and french versions of the posttraumatic stress disorder checklist for dsm- (pcl- ) the italians in the time of coronavirus: psychosocial aspects of unexpected covid- pandemic the enemy who sealed the world: effects quarantine due to the covid- on sleep quality, anxiety, and psychological distress in the italian population immediate psychological responses and associated factors during the initial stage of the coronavirus disease (covid- ) epidemic among the general population in china psychological symptoms of ordinary chinese citizens based on scl- during the level i emergency response to covid- psychological distress of nurses in taiwan who worked during the outbreak of sars kikwit ebola outbreak: lessons hospitals and physicians can apply to future viral epidemics the impact of communications about swine flu (influenza a h n v) on public responses to the outbreak: results from national telephone surveys in the uk psychosocial effects of an ebola outbreak at individual, community and international levels depressive symptoms among survivors of ebola virus disease in conakry (guinea): preliminary results of the postebogui cohort prevalence of psychiatric morbidity and psychological adaptation of the nurses in a structured sars caring unit during outbreak: a prospective and periodic assessment study in taiwan alexithymia: a facet of uncontrolled hypertension coping styles in individuals with hypertension of varying severity emotion and overeating behavior: effects of alexithymia and emotional regulation on overweight and obesity the night effect of anger: relationship with nocturnal blood pressure dipping funding: this research received no specific grant from any funding agency, commercial or not-for-profit sectors. there is no funding support for this survey. we would like to thank gianluca pistore and all the people who helped in the data collection by sharing our survey on various social media. the authors declare no conflict of interest. key: cord- -hlkwwaqe authors: grippo, antonello; assenza, giovanni; scarpino, maenia; broglia, lidia; cilea, rosalia; galimberti, carlo andrea; lanzo, giovanni; michelucci, roberto; tassi, laura; vergari, maurizio; di lazzaro, vincenzo; mecarelli, oriano title: electroencephalography during sars-cov- outbreak: practical recommendations from the task force of the italian society of neurophysiology (sinc), the italian league against epilepsy (lice), and the italian association of neurophysiology technologists (aitn) date: - - journal: neurol sci doi: . /s - - - sha: doc_id: cord_uid: hlkwwaqe background: during covid- lockdown, non-urgent medical procedures were suspended. grade of urgency of electroencephalography (eeg) may vary according to the clinical indication, setting, and status of infection of sars-cov- virus. “italian society of clinical neurophysiology” (sinc), “italian league against epilepsy” (lice), and the “italian association of neurophysiology technologists” (aitn) aimed to provide clinical and technical recommendation for eeg indications and recording standards in this pandemic era. methods: presidents of sinc, lice, and aitn endorsed three members per each society to formulate recommendations: classification of the degree of urgency of eeg clinical indications, management and behavior of physicians and neurophysiology technologists, hygiene and personal protection standards, and use of technical equipment. results: scientific societies endorsed a paper conveying the recommendation for eeg execution in accordance with clinical urgency, setting (inpatients/outpatients), status of sars-cov- virus infection (positive, negative and uncertain), and phase of governmental restrictions (phase and ). briefly, in phase , eeg was recommended only for those acute/subacute neurological symptoms where eeg is necessary for diagnosis, prognosis, or therapy. outpatient examinations should be avoided in phase , while they should be recommended in urgent cases in phase when they could prevent an emergency room access. reduction of staff contacts must be encouraged through rescheduling job shifts. the use of disposable electrodes and dedicated eeg devices for covid- -positive patients are recommended. conclusions: during the different phases of covid- pandemic, the eeg should be reserved for patients really benefiting from its execution in terms of diagnosis, treatment, prognosis, and avoidance of emergency room access. the coronavirus pandemic (covid- ) required a substantial reorganization of healthcare services. since the beginning of the emergency period, international [ , ] and italian governmental agencies [ , ] recommended the suspension of all elective and "non-urgent" procedures, including surgical procedures radiologic and neurophysiological tests. therefore, routine outpatient visits were suspended and individual attempts of remote assistance (by telephone, e-mail or video-audio-conferencing consultations) tried to limit the harmful consequences for patients [ , ] . however, the definition of "elective" or "non-urgent" is largely variable in relation to the type of specialty of the diagnostic procedures and clinical setting; therefore, classification of the clinical request is frequently left to the individual medical judgment. postponing non-essential health interventions is crucial to prevent the overload of health system services and its employees, in order to maintain stocks of personal protective equipment (ppe) and care supplies, and to ensure the safety of patients, the community, and healthcare workers. however, despite the covid- pandemic and its related risks, urgent care should be provided in a timely and safe manner to avoid delays in diagnoses and treatments. clinical decisions on the priorities of examinations are not always easy to balance. in acute life-threatening conditions, there is no doubt about acting. in other situations, as in the case of neurophysiological procedures such as electroencephalogram (eeg), the choice can be questionable. eeg is a commonly used neurophysiological diagnostic procedure and is particularly useful for neurological conditions with acute/subacute onset. in such circumstances, eeg cannot be indefinitely postponed. in this scenario, there is an urgent need to carefully identify clinical indications for eeg according to criteria of appropriateness and priority [ ] . the most representative scientific societies of health professionals involved in the eeg execution and reporting, "italian society of clinical neurophysiology" (sinc), "italian league against epilepsy" (lice), and the "italian association of neurophysiology technologists (aitn), received several requests for indications on how to perform eeg examinations during this health emergency from their members. other professionals (emergency doctors, intensivists, etc.) also expressed the need for a specialistic regulation of the patient selection for eeg. for these reasons, sinc, lice, and aitn decided to draw up the following recommendations for the selection of appropriate clinical indications in different contexts (inpatient or outpatient) and of the procedures to follow during the execution of the eeg. representative members of sinc, lice, and aitn were mandated by the presidents of their respective scientific societies to provide a joint paper on recommendations about the clinical indications and technical precautions to perform eegs during the covid- pandemic. the document aims to provide guidance on: . classification of clinical indications for the execution of eegs according to the degree of urgency . specific methods of management and behavior of medical personnel . specific methods of management and behavior of neurophysiology technologist (npt) personnel . hygiene and personal protection standards, use of ppe . use maintenance and disinfection of technical equipment recommendations are proposed for the different phases of covid- pandemic: -phase (restriction of outpatient services to emergencies) -phase (intended as the initial outpatient reopening phase, with partial restrictions on services) the clinical setting that could apply the recommendations have been identified as: -inpatients (emergency room, intensive care, ordinary ward, epilepsy monitoring unit) -outpatients the different types of patients were classified, based on the results of the sars-covid- virus swab, in: -covid+ (positive) -covid-uncertain (suspected positive, pending swab outcome) -covid− (negative) the document was approved by the respective boards of directors on may , , and published on the web sites of the respective scientific societies (www.sinc.it; www.lice.it; www.aitn.it). the clinical conditions in which the eeg request is considered undeferrable or not are listed below. -acute alterations in the level of consciousness (when clinical data and basic investigations do not identify the etiology; differential diagnosis with non-convulsive status epilepticus (ncse)) -monitoring of pharmacological treatment of convulsive and non-convulsive status epilepticus -suspicion of acute encephalitis -neonatal asphyxia and related treatment with therapeutic hypothermia -prognostic evaluation of post-anoxic coma -brain death assessment urgencies that can be postponed: -first epileptic seizure -occurrence of unprovoked seizures in known epilepsy -prolonged neurological deficits after epileptic seizures -suspicion of psychogenic non-epileptic seizures phase of the covid- pandemic eeg examinations should only be carried out in emergency situations and clinically cannot be delayed, both for inpatients and outpatients. the scientific societies promoting the present document consider that, with the exception of the "undeferrable" category, the neurological specialist physician responsible for the eeg examination should assess a potentially useful diagnostic indication on a case-by-case analysis. however, the physician should always reserve the execution of the eeg only for critical patients in whom the findings of the examination could significantly change the diagnostic, prognostic, and therapeutic management. it is advisable that: the neurological specialist assesses the clinical need and establishes the appropriate timing of all eeg studies in covid+/covid-uncertain patients, in collaboration with the team of clinicians involved in direct care; if the npt staff is limited, the neurological specialist assesses the appropriateness of all eeg requests, regardless of covid- status; the possibility of performing a prolonged eeg recording rather than repeated standard eeg examinations ( - min) is examined beforehand, depending on the request, in order to reduce overall npt efforts and the contact time of covid+/covid-uncertain patients; the usefulness/necessity of standard activation procedures, particularly hyperventilation, is carefully assessed in the individual case. hyperventilation in covid+/ covid-uncertain patients is not recommended. the execution of hyperventilation with the mask does not produce the physiological changes presumed to sustain an epileptiform eeg activation, thus it should never be performed. -at the present, there are no reasons to use different rules and precautions (see below) for inpatients of intensive care units respect to those in non-intensive care units. actually. the real or presumed covid status (positive, negative, uncertain) is the only variable that should modify the medical and technical approach to the eeg. currently, following the italian ministry of health's disposal [ , ] , non-urgent and elective outpatient activities are suspended (i.e., they can be postponed without compromising the subject's state of health). exceptions are all services related to transplanted patients or for oncological pathologies or perinatal neurological problems requiring serious controls. some limitations may be prolonged and possibly extended to later stages of the pandemic. it is therefore considered appropriate to provide some generic indications, useful both in phase and , until new conditions arise. when booking the examination, however, the patient should always be informed that the examination will be postponed if he/she presents, in the days preceding the date of the appointment, one or more symptoms which described in the first stages of sars-cov- infection: headache, myalgia, cough, fever (> . °), dyspnea, generalized asthenia, gastrointestinal symptoms, ageusia, anosmia, etc. (for a facsimile triage card, see attachment ). at the time of the appointment, it is advisable to ask the patient if he/she has performed serological tests or swabs and if the symptoms listed above have occurred among the cohabitants. for a correct and safe organization of the booking and execution of the examinations, it is advisable to observe some precautions: -increase the pause time between consecutive examinations so as not to prolong the patients' stay in the waiting room allowing the sanitation of the outpatients' clinic and equipment. -with the exception of minors and/or disabled people who need their presence, try to reduce as much as possible the presence of accompanying persons both in the waiting areas and in the laboratory. -to distance the seats in the waiting room by at least m from each other. -in the laboratory, the patient must always wear a surgical mask. -all surfaces and furnishings in the laboratory must be sanitized with antiseptic products and the room must be frequently ventilated. no change from phase is expected with regard to eeg testing in hospitalized patients. for outpatients' activities, the risks associated with the potential exposure of the patient and healthcare professionals to infection and the benefits of performing the examination should be balanced. however, it is sometimes necessary to perform an urgent, non-delayable outpatient eeg, also in order to avoid improper access to the emergency department. the followings are some decision-making criteria for the different eeg investigations: in phase of the emergency, the specialist doctor should conduct a risk/benefit assessment that takes into account the following questions in particular: -can the standard eeg provide useful diagnostic information to prevent a possible and imminent emergency condition, thus avoiding access to the emergency room (er, e.g., for a seizure or epileptic status), with an increased risk of exposure to infection and other comorbidities? -can the eeg be useful for changes to the treatment plan? in addition to emergencies, in phase , the recording of a standard eeg in the outpatient clinic should be reserved for those with a first epileptic seizure, or cases in which there is a diagnostic doubt that may influence the treatment, or for those who have already received a diagnosis of epilepsy (especially if it is a severe and drug-resistant form) and who need accurate monitoring that cannot be postponed indefinitely (seizure recurrence; control effectiveness, changes or withdrawal of antiseizure treatment, etc.). even more than standard eeg, this more laborious procedure should include a risk/benefit balance with regard to the impact that the examination results could have on therapeutic procedures and on the reduction and prevention of accesses to the er. some specific measures are recommended: -if possible, the npt personnel unit itself should take care of the application and dismantling of the equipment. -the possibility to perform the connection/removal at the outpatient laboratory and not in the hospital should be considered in order to limit the patient's exposure and access to the hospital environment. this type of examination is at greater risk especially for npt personnel who must remain in close contact with the patient during preparation, even for prolonged periods of time, and perform maneuvers on the face and near the upper respiratory tract (application of electrodes for electro-oculogram and electro-myogram of the mylohyoid muscle, oronasal/cannula sensor or positioning of masks for cpap). it should be noted that during positive pressure positive airway ventilation (cpap) healthcare workers are particularly exposed to aerosolized particles. for this type of examination, the importance of cleaning and sanitizing all devices between consecutive recordings is an option to be considered. long-term monitoring (ltm) is defined as a polygraphic recording associated with video. the purpose is not only diagnostic, but it allows individualized therapeutic approaches (based on the correct classification of seizures and eeg alterations), such as modification or withdrawal of the pharmacological treatment in progress in case of a diagnosis of a different syndrome. two are the modalities of execution of the ltm: -prolonged (variable length in terms of days) hospitalization for the recording of habitual seizures, for pre-surgical purposes daily recordings, lasting more than h, often associated with sleep recording in phase of the pandemic, in line with national directives, which almost banned elective hospitalizations in order to optimize the availability of beds for potential covid- patients, from march , most facilities had to cancel elective hospitalizations for ltm, both for inpatients and outpatients. in this phase, the ltm study with diagnostic purposes was considered almost always deferrable, unless special clinical conditions established the real unavoidable urgency. in the daily management of patients at home, therefore, the acquisition by home-video witnesses of the reported critical manifestations, the recording of an ambulatory eeg (possibly associated with home-video), or the ex-adjuvantibus treatment with drugs can be considered useful alternatives. considering that most of the new digital recording systems have a video program, it is possible to consider, for patients who present in the er with frequent seizures, the possibility of performing an urgent hospitalization for video-eeg monitoring, which allows a quick diagnosis and may avoid further accesses to the er, or an outpatient dynamic eeg if readily and quickly available. the ltm study for pre-surgical purposes has no urgency at this stage. in phase of the pandemic, in view of the reopening of ltm in the inpatient settings, the following points should be considered: finally, it should be stressed that the mask worn by the patient, if maintained throughout the registration period (especially in persons with disabilities and children), will prevent the acquisition of clinical elements that could be fundamental and hardly tolerated. during the pandemic, it is desirable that specialist medical staff (neurologists, neurophysiopathologists, child neurologists) can work remotely whenever possible (examination reports, surveillance of eeg monitoring, video-consultations, etc.) making maximum use of telemedicine resources (see national guidelines -clinical teleneurophysiology, sinc intercompany gds -digital sit [ ] ). it is advisable to consult in advance with the hospital administrations to develop a plan to provide the necessary neurophysiological services while supporting the needs of the hospital itself [ ] . the institutional recommendations concerning the participation in the care activity of medical specialists in training, contract workers, doctoral students, and students of the degree courses in npt should be followed. in addition, due to the reduction in the volume of eeg examinations, it should be considered that medical staff may be able to work on alternate days or rotating shifts to limit exposure to the infection, or may be moved to other areas depending on the needs of the hospital. the current guidelines of the italian higher institute of health [ ] recommend all staff to always wear surgical masks while in clinical care areas or unable to maintain a safe social distance in hospital facilities where there are possible covid+ cases. in general, these masks should be used for the entire shift and replaced during the shift in case of deterioration. the guidelines also recommend that all people in the hospital's environment (patients, family members, etc.) should wear a surgical mask. institutional policies should regulate inpatient visits and only one family member/caregiver should accompany patients for outpatient procedures and monitor patients and staff on a daily basis for symptoms related to possible infection by completing questionnaires and measuring body temperature. information should be collected on any covid+ or covid-uncertain status prior to performing the diagnostic procedure. in case of covid-uncertain inpatients it is necessary to consult the clinical team about the possibility to delay the procedure till the knowledge of serological results; if the eeg investigation is undeferrable, the patient should be considered as covid+. since hospital policies and specific procedures are variable between institutions and operating units and could change over time, it is necessary to keep the npt staff constantly updated. during the pandemic period, many hospitals complained of a shortage of npt staff for various contingent reasons (sick leave, parental leave, home quarantine, etc.). in case of persistence of such critical human resources issues, it may be necessary to reschedule the work plans (afternoon and holiday shifts, on call availability) of each facility. in case of continuous/prolonged eeg monitoring, it would be desirable to limit the number of npt entering the rooms of covid+/covid-uncertain patients. considerations should be given to quick mounting of electrodes (e.g., pre-wired headcap, possibly disposable, although this material is currently only available in a few hospitals). the following general provisions are recalled: -to follow scrupulously who recommendations regarding hand hygiene [ ] -to maintain physical distance and to limit the number of people in the laboratory as much as possible -to use ppe as recommended by institutional guidelines for services distributed in different buildings, it is advisable to dedicate specific personnel to each service avoiding continuous exchanges of personnel with a consequent increase in the number of contacts. the staff going to the er or to departments dedicated to covid+ patients should go alone, avoiding assemblage with other colleagues or other professional figures. before the contact with the patient, the eeg npt must wear the following ppe in order to implement contact, droplet, and airborne isolation measures in addition to standard precautions and respiratory hygiene: a) gloves, protective clothing (long-sleeved waterproof non-woven shirt, over-shoes, headgear) to protect the body b) mask/eyewear for face protection c) ffp mask for respiratory protection all ppe must be removed and disposed of in designated areas. it is essential that ppe is dressed and undressed in accordance with procedures established by the ministry of health (http://www.salute.gov.it/portale/news/p _ _ _ _ .jsp? menu=multimedia&id= ) and who [ ] . use, maintenance, and sanitation of eeg equipment & all surfaces of eeg equipment that have entered a covid+/covid-uncertain patient environment must be sanitized using antiseptic wipes (for equipment disinfection protocol see [ ] ) & eeg equipment to be used in environments with covid+/covid-uncertain patients can be protected with a "super-stick" plastic film that can be removed after the examination (uneven surfaces that are difficult to sanitize with wet cleaning agents such as pc keyboards and mice should be particularly protected -seventy percent ethyl alcohol for non-disposable material, such as cup or head-cap (immersion for no more than min in a cellophane bag). -water, common detergent and % sodium hypochlorite or other disinfectant for cleaning rooms and surfaces. & in eeg laboratories, it is advisable to limit the amount of eeg equipment and furniture exposed to possible contamination. & if instrumental equipment permits, it is preferable to dedicate individual eeg equipment to covid+ sectors. & both the environment and the eeg equipment should be sanitized in strict compliance with institutional policies. the need to sanitize the eeg equipment, accessories, furnishings, and laboratory environment may however limit the number of examinations that can be carried out in a shift. this must be considered in the activities planning of the neurophysiology service. making a professional contribution to limiting the spread of sars-cov- virus infection, neurological specialists need to take in charge the complex assessment of the actual urgency/ not deferrable of the different eeg exams, despite limited resources. the physician should examine the risks and benefits of each eeg examination and ensure appropriate precautions are properly applied. delays or cancelations of an eeg examination can be problematic for both patients and physicians. it may be helpful to share unique messages to patients, based on ministerial recommendations, to reduce the risk of infection, while at the same time providing empathy with patients' concerns. involving the clinicians who are in charge of the patient in the decision-making process is important because the reasons for requesting the urgent eeg may not have been clearly expressed. a direct interview with the referring physician can therefore better explain the clinical manifestations and their course and refine the assessment of the various clinical scenarios, contributing to a rational decision. the patient's referring physician is therefore an excellent resource to evaluate the urgency of the eeg's request in a timely manner. finally, in selected cases, a teleconsultation (telephone or video) with the outpatients can help the decision-making process on the execution of the eeg by reassuring the patient. one of the limitations of this document is that not all clinical diagnostic scenarios can be reasonably described. however, our aim is to provide general recommendations to guide physicians and npt staff so that patients requiring eeg studies continue to receive them according to safe and appropriate procedures, minimizing the risk of infection for healthcare professionals. the development of the covid- pandemic is still under observation and increasing infection rates may require maintaining current restrictions over a long period of time. in the interest of patients and healthcare professionals, neurologists should engage as much as possible in planning eeg investigations to ensure patient and healthcare worker safety. the evolution of the pandemic and related health measures could open scenarios needing a revision of these recommendations in the next future. author contributions ag, ga, ms, rc, cag, gl, rm, and mv wrote the manuscript and revised it according to the recommendations and finalities of each scientific societies. lb, lt, vdl, and om conceptualized and revised the manuscript. conflict of interest the authors declare that they have no conflict of interest. ethical approval not applicable. consent to participate all authors gave their explicit consent to participate to the present study. consent for publication all authors gave their explicit consent for publication in the present form. non-emergent, elective medical services, and treatment recommendations interim guidance for healthcare facilities: preparing for community transmission of covid- in the united states ministero della salute italiano -circolare ministeriale n. del . . : linee di indirizzo per la rimodulazione dell'attività programmata differibile in corso di emergenza da ministero della salute italiano -circolare ministeriale n. del . . : aggiornamento delle linee di indirizzo organizzative dei servizi ospedalieri e territoriali in corso di emergenza covid- ″ covid- is catalyzing the adoption of teleneurology neurologists respond to covid- with telemedicine: the challenges and opportunities eeg during the covid- pandemic: what remains the same and what is different linee di indirizzo nazionali -teleneurofisiologia clinica gds intersocietario sinc -digital sit practical considerations when performing neurodiagnostic studies on patients with covid- and other highly contagious diseases indicazioni ad interim per un utilizzo razionale delle protezioni per infezione da sars-cov- nelle attività sanitarie e sociosanitarie (assistenza a soggetti affetti da covid- ) nell'attuale scenario emergenziale sars-cov- . gruppo di lavoro iss prevenzione e controllo delle infezioni who library cataloguing-in-publication data ( ) who guidelines on hand hygiene in health care license: cc by-nc-sa . igo . ppe and equipment disinfection procedures for patients on respiratory isolation precautions publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations key: cord- -ls q g v authors: balsamo, michela; carlucci, leonardo title: italians on the age of covid- : the self-reported depressive symptoms through web-based survey date: - - journal: front psychol doi: . /fpsyg. . sha: doc_id: cord_uid: ls q g v the pandemic of coronavirus disease (covid- ) has affected the italian community. the widespread use of quarantine had the desired impact of controlling the epidemic, although it caused many psychological consequences. to date, compliance of the italian public with voluntary home quarantine has been very high, but little is known about the impact of psychological health on sociodemographic categories during the quarantine. the purpose of this study was to assess the prevalence of depressive symptoms in specific sociodemographic categories during the covid- quarantine lockdown and the potential factors that contribute to, or mitigate, these effects. in the very early stage of the nationwide lockdown, , quarantined italian adult residents ( % females, ranging from to years) participated in a web-based cross-sectional survey, including measures of depressive symptoms, which were measured by the teate depression inventory, and state anxiety levels. the overall prevalence was . % for moderate and . % for severe levels of depressive symptoms. a generalized logistic model was used to identify the factors associated with mental health problems. among these factors, sociodemographic variables (e.g., sex, age, employment status) and adherence to quarantine guidelines were analyzed. females, younger people, students, singles, residents in northern italy, people who were reluctant to adhere to quarantine guidelines, and people less worried about being infected with covid- were at high risk of developing depressive symptoms during the covid- epidemic, also after controlling for state anxiety. these findings showed that public levels of depressive symptoms did not increase the greater likelihood of being infected. our study suggested that the monitoring of psychological outcomes for outbreaks could identify groups at higher risk of psychological morbidities due to the current pandemic in order to target future psychological interventions for implementation. the pandemic of coronavirus disease , caused by the severe acute respiratory syndrome coronavirus (sars-cov- , previously known as -ncov) has affected the italian community since late january. according to the imperial college covid- response team (ferguson et al., ) , cumulatively, . ( . - . ) million people had been infected as of march , giving an infection rate of . % ( . - %) of the italian population. to contain the rapid spread of this pandemic, the italian government ordered nationwide lockdown by march : all public places were closed (included educational, religious, and public/cultural institutions, such as schools, universities, museums, and law courts), all public events and any form of congregation were banned, and a distance of at least m had to be maintained (government of italy, ). all italian people were in quarantine at home (#iorestoacasa) until may : people had to stay at home apart from essential tasks. the slowing growth in daily reported deaths in italy was consistent with the significant impact of these restrictions. the effective reproduction number, rt, dropped to close to around the start of the lockdown, with , ( , - , ) deaths averted. the widespread use of quarantine had the desired effect of controlling the epidemic, this was also due to the fact that the compliance of the italian public with quarantine guidelines had been very high (carlucci et al., ) . yet, the pandemic created a breeding ground for direct psychological consequences, suddenly throwing many individuals into daily lives filled with health threats, existential depression, and generalized stress (holmes et al., ) . a recent review of the psychological impact of quarantine, due to earlier outbreaks, suggested that there were high rates of negative psychological effects among the public, including post-traumatic stress symptoms, persistent depression, substantial anger, panic attacks, and suicidality (liu et al., ; maunder et al., ; brooks et al., ) . social distancing and isolation exacerbate the burden of stress, and often cause effects on immune, cardiovascular, and mental health because these measures frustrate the deep-seated human instinct to connect with others. on this point, social connection helps people to regulate negative emotions, remain resilient during difficult times, and cope with stress (rimé, ; hawkley and cacioppo, ; haslam et al., ; doré et al., ; jetten et al., ) . remarkably, the mental impact of quarantine can depend largely on the characteristics of participants and the quarantine variables selected. as documented by reynolds et al. ( ) , taylor et al. ( ) , and brooks et al. ( ) , while compliance with quarantine guidelines requirements are significant factors behind a higher level of post-traumatic stress disorder (ptsd), healthcare workers were more likely to be affected than the public. in italy, in the early phase of quarantine ( - days after the decree of nationwide lockdown), deleterious consequences on the population's psychological health were analyzed in a nationally representative survey of , participants (barari et al., ) . different demographic groups were struggling with different aspects of quarantine. older adults expressed worry or anxiety, while those who were likely working parents ( - years) cited consistent economic distress and struggles with home-schooling and smart-working, compared to other groups. younger people were struggling with increased boredom, perceived immobility, and conflicts within family, while vulnerable groups, like the elderly and health-compromised people, cited consistent loneliness relative to others. overall, the average level of anxiety surrounding the crisis in the italian population was high: no respondents reported being completely without anxiety. according to barari et al. ( ) , the negative psychological consequences of the quarantine were beginning to wear on people and seemed likely to become more serious over time. further findings derived from an online survey showed the prevalence of psychiatric symptoms in , participants drawn from the general population from march to . female gender, negative affect, and detachment were associated with higher levels of stress, anxiety, and depression. having an acquaintance infected was associated with increased levels of both depression and stress, whereas a history of stressful situations and medical problems was associated with higher levels of anxiety and depression. finally, those with a family member infected and young people who had to work outside their domicile presented higher levels of anxiety and stress, respectively. thus, it is important that the potential advantages of home quarantine are weighed against the possible mental costs (rubin and wessely, ; torales et al., ) . quarantine as an efficacious public health measure also needs to lower the psychological strain associated with it. research evidence aims of this study were to explore ( ) the likely effects of quarantine on mental health (anxiety and depressive symptoms), immediately after the nationwide lockdown issued by the italian government, and ( ) the factors that contribute to, or mitigate, these consequences. among these factors, sociodemographic variables (gender, age, employment status, marital status, education, geographic area, and income per year), worry about being affected by covid- , and adherence with quarantine guidelines were analyzed. depression was the principal outcome, while anxiety was used as a covariate, given its close association with depression (clark and watson, ; barlow and campbell, ) . respondents were italian quarantined adults aged and older with access to a networked computer. an online cross-sectional study was conducted from march to , immediately after the nationwide lockdown issued by the italian government on march (#iorestoacasa). a virtual snowball sample via social media was used within a wider web-based study including other psychological measures (carlucci et al., ) . this study has been approved by the department of psychological sciences, health and territory, university of chieti, italy review board. written informed consent was obtained from all the participants included in the study. of psychiatric illnesses and medical problems (e.g., hospitalizations) were collected. the -item teate depression inventory (tdi; balsamo et al., , b , developed via rasch analysis (rasch, ) , was employed to evaluate depressive symptoms in participants in the past weeks. respondent answers were measured on a point likert scale ranging from "never" to "always." cronbach's α coefficient in our study was . . the -item state scale of the state-trait inventory for cognitive and somatic anxiety (sticsa) (ree et al., ; balsamo et al., ; carlucci et al., ) was administered to evaluate cognitive (e.g., "i have trouble remembering things") and somatic (e.g., "my muscles are tense") symptoms of state anxiety. individuals rated how often a statement was true in the past weeks, from "not at all" to "very much so." cronbach's α coefficient was . . adherence to quarantine guidelines in response to the covid- outbreak was measured by a global index composed of items classified into preventive (i.e., handwashing with soapy water/alcohol-based solution) and avoidant (i.e., avoidance of gatherings in public or open to public places, handshaking) disease behaviors (carlucci et al., ) . respondents were asked about the frequency of which they had carried out quarantine restrictions on a -point likert scale from "never" to "always." cronbach's α coefficient was . . worry about being infected with covid- was assessed by a single item drawn from a multidimensional questionnaire of risk perception for the covid- infectious disease outbreak. responses were classified according to three levels of worry severity: none ("not worried at all"), moderate ("slightly worried"), and quite a lot ("really worried"). descriptive statistics were computed for sociodemographic characteristics, physical symptoms and health service utilization variables, knowledge and concern-related variables, precautionary measure variables, and additional health information variables. prevalence of depressive and anxiety symptoms during the covid- outbreak in the italian population were also computed for sex and age. in line with similar studies (e.g., giallonardo et al., ; wang et al., ) , the tdi outcome score was categorized into "minimal, " "mild, " "moderate, " and "severe" depression levels . a sticsa-s score of points or greater was indicated as the cut-off point for the presence of anxiety symptoms (van dam et al., ) . sensitivity analysis was conducted using the hmisc r package (harrell and dupont, ) in order to assess the power and sample size of ordinal outcomes under the proportional odds ordinal logistic model. next, generalized linear regression (glms) was applied to the explanatory model to analyze whether the severity of depression during the covid- quarantine could be predicted by high levels of adherence to quarantine guidelines and worry about being infected with covid- , and by sociodemographic variables (model ), that resulted as significant in our previous study (carlucci et al., ) . since depression was measured in terms of severity levels, we specified multinomial (ordinal) as the distribution and cumulative logit as the link function. predictors were selected according to a two-step process. firstly, a potential set of considered variables were correlated with the outcome variable. all the potential variables that correlated significantly with the outcome were selected as predictors in the glms model. hence, based on the test of model effects (wald chi-square statistic and p-values), the predictors were compared (guisan et al., ) . only the resulting significant predictors (p < . ) were retained in the model. a wald test (and its % confidence interval) based on robust estimates of the coefficients and covariance matrix were used to assess the models, and residual deviance as a goodnessof-fit statistic was applied to evaluate model overdispersion (mccullagh, ) . the model with the deviance/df ratio closest to the unit was retained as the most parsimonious model (mccullagh, ) . in addition, due to the high comorbidity between depression and anxiety symptomatology (clark and watson, ; barlow and campbell, ) , the model was re-estimated controlling for anxiety as the covariate (model ), in order to increase the ability to detect differences on dependent variables (depression severity levels) by an independent variable inserted as the covariate. differences between the two models were interpreted in terms of unique contribution of any independent variable on depression severity symptoms. the data were statistically analyzed with spss for windows . (ibm corp, ). statistical significance was set by p-values of less than . . the characteristics of the participants are shown in table . of the , respondents, , ( . %) were male and , ( . %) were female. the mean (±sd) age of the participants was . ± . years. a total of . % of them were located in the south of italy (n = ). among these, , ( . %) respondents held a high school diploma, while , ( . %) held a higher education qualification (bachelor/master/doctorate). in terms of occupational status and income per year, , ( . %) participants were students, ( . %) were healthcare workers, and , ( . %) were employed. concerning marital status, , ( . %) participants were unmarried/single, ( . %) were married, ( . %) were divorced/separated, ( %) were cohabiting, and ( . %) were widowed. most of the subjects had a high level of health: , ( . %) were found to show no physical disease, while ( . %) were detected as "fragile, " having more than three diseases, and with a long history of chronic medical illness. a total of ( . %) participants had previously carried out psychotherapeutic treatment. among these, . % of respondents had carried out at least one psychotherapy treatment (individual, family/couple, and/or group treatment), . % had undergone psycho-pharmacological treatment, and . % had participated in other psychological treatments. most ( . %) of the participants spent their quarantine period with family members. a total of , ( . %) were found to be highly adherent to quarantine guidelines, and , ( . %) reported that they were worried about being infected with covid- . a sample size of , was used for the statistical power analyses, and a : odds ratio was used as a baseline. the alpha level used for this analysis was p < . . the post-hoc analyses showed that the statistical power for this study was . . thus, there was an adequate power at the moderate to large effect size level. an n of approximately , would be needed to obtain statistical power at the recommended . level (cohen, ) . the overall prevalence was . % for severe, . % for moderate, . % for mild, and . % for minimal depressive symptoms. the overall prevalence of anxiety symptoms was . %, using the cut-off of > . taking together, the prevalence of depressive symptoms and state anxiety severity was significantly higher in female participants, and those younger than years compared to participants aged years or older (p < . , as shown in tables , ). in addition, those who received psychotherapeutic treatment in the past reported higher severity levels of depressive [χ = . ( ), p < . ] and anxiety [χ = . ( ), p < . ] symptoms relative to the general public. preliminarily, nonparametric correlations (spearman's rho) were performed in order to select independent variables as predictors in the glms. as expected, all the sociodemographic variables, as well as worry were found to correlate negatively with the outcome variable (depressive symptom severity levels) ranging from rho = . (age, p < . ) to rho = − . (education, p < . ), except for sex and worry (rho = . , p < . and . , p < . , respectively). parameter estimates of the generalized linear model and the exponentiated values of the coefficients [the "exp(b)" column] are displayed in table . the first model resulted in an underestimation of the data with no statistical association for education [wχ (df ) = . ( ), p = . ], and income per year [wχ (df ) = . ( ), p = . ] as predictors of depressive symptoms. to improve model fit, we discharged them and reestimated the model. the omnibus test [χ (df ) = . ( ); p < . ], and residual deviance/df ratio (deviance/df = . ) of the re-estimated models suggested that the refined model fit significantly better than the proposed model (mccullagh, as expected, a preliminary analysis showed that depression and anxiety symptoms shared approximately % of the common variance, as derived by the spearman rho coefficient (rho = . , p < . ). thus, state anxiety, as measured by the state sticsa, as the covariate was inserted in our model (model ). compared to the previous model, no statistical differences were found in sex, age, and adherence level to quarantine guidelines groups when predicting depression symptom severity, when controlling for anxiety (see appendix a). statistically significant differences were maintained in depression symptom severity for marital status, geographic area, and occupational status groups, after controlling for anxiety. in detail, divorced/separated participants were less likely to experience depressive symptoms compared to single people, with a significant decreased risk of depressive symptoms (odds ratio: . [ % ci, . - . ]). on the other hand, no statistical differences on depressive symptom severity was found in cohabiting participants compared to unmarried participants in the second model. concerning geographic area, the participants living in central italy were found less likely to experience severe depressive symptoms (odds ratio: . [ % ci, . - . ]) compared to those living in north-west italy, after removing anxiety effects. far from the previous model, the healthcare professionals' group were not found to differ from other occupational groups in predicting high levels of depressive symptoms compared to unemployed participants. interestingly, participants "moderately" worried about being infected with covid- were more prone to experience high levels of depressive symptoms (odds ratio: . [ % ci, . - . ]) compared to participants who were "quite a lot" worried, after controlling for the state anxiety effect. we fitted both the glms models (without and with anxiety as the covariate) in the subsample of participants with healthcare professionals (β = − . ; se = . ; p < . ) were less depressed compared to unemployed participants, with a decreased risk of depressive symptom severity (odds ratio: . [ % ci, . - . ]). next, those who reported to adhere to the quarantine guidelines had significantly lower levels of depressive symptoms compared to those who were not adherent (β = − . ; se = . ; p < . ), with a decreased risk of depressive symptoms (odds ratio: . [ % ci, . - . ]). no statistical association was found for worry about being infected with covid- and geographic area. surprisingly, in model , no statistically significant differences were maintained in depression symptoms for the sex, marital status, and occupational status groups, after controlling for anxiety. participants aged from to years reported significantly lower levels of depression scores compared to the younger participants (β = − . . ; se = . ; p < . ), with a decreased risk of depressive symptoms (odds ratio range: . [ % ci, . - . ]). concerning geographic area, the participants living in south italy were less likely to experience severe depressive symptoms (odds ratio: . [ % ci, . - . ]) compared to those living in north-west italy, after removing anxiety effects. as expected, participants "moderately" and "none" worried about being infected with covid- were more prone to experience high levels of depressive symptoms (odds ratio: . / . [ % ci, . - . / . - . ]) compared to participants who were "quite a lot" worried, after controlling for the state anxiety effect. quarantine has been used extensively in all countries of the world to lower the spread of the covid- infection and to protect individuals' health, at different times (sohrabi et al., ) . quarantine includes the separation and restriction of movement of people who have potentially been exposed to a contagious disease to ascertain if they become unwell, so reducing the risk of them infecting others. it is an unpleasant experience for those affected (hiremath et al., ) . imposed isolation and separation from loved ones, loss of mental health needs (freedom, social contacts, stimulation), uncertainty over disease status, family conflict, and boredom can, on occasion, contribute to the onset of psychological disorders (brooks et al., ) . due to the fact that the psychological impact of quarantine depends largely on the characteristics of participants and the quarantine variables selected, several sociodemographic characteristics have been selected here, with depressive symptoms measured by the tdi as the outcome. in the second model, anxiety, as measured by the sticsa state scale, was inserted as the covariate, given the close relationship with depression (brooks et al., ) . about gender, depressive symptoms were more likely to occur in female participants, with a risk of developing depressive symptoms higher . compared to male participants in our sample. this finding was in accordance with studies by qiu et al. ( ) and wang et al. ( ) among the chinese general population in the first weeks following the outbreak, as well as broche-pérez et al. ( ) among the cuban population. also, among the italian general population higher levels of psychological distress were reported in the female gender compared to their male counterparts rossi et al., ) . sex differences in depression were not caused by a higher prevalence of covid- infection in women because mortality and vulnerability to the covid- disease indicated that more men are dying from covid- (lancet, ) . thus, these differences seem to be caused by the fact that women carry a different kind of burden from this epidemiological emergency. gender inequities disproportionately affect the wellbeing and economic resilience during lockdown. households are under strain, but children and elderly care, as well as housework generally fall on women (cluver et al., ) . by increasing caregiving needs, covid- has intensified the pressure on women to uphold prescriptive feminine norms. women have to bear more of the burdens of providing additional support for children's distance learning, and alleviating children's emotional tedium, isolation, and anxiety of shelter-in-place (rosenfeld et al., ) . in addition, increased intimate partner violence has grown during the quarantine due to covid- because women are required to stay uninterruptedly with their partners and away from those people who can give help or at least validate their experiences and, particularly if these women live in small houses (bradbury-jones and isham, ; van gelder et al., ) . indeed, some studies suggest that sudden forced proximity with their immediate household members is a risk factor for domestic violence, and aggression (taylor et al., ; brooks et al., ) . in italy, since the beginning of the covid- quarantine, three domestic homicides and murder-suicides have been registered to date. furthermore, while covid- has coincided with greater rises in unemployment for women than men, the rise in unemployment for men remains substantial (bureau of labor statistics, ). about the age groups, depressive symptoms were most likely to occur in younger people (aged - years). with increasing age, depressive symptoms were less prevalent during the italian lockdown due to the covid- outbreak. our results were similar to those from previous studies, such as a study during the sars outbreak in taiwan (su et al., ) , a study of horse owners quarantined because of equine influenza (taylor et al., ) , and one recent study during the covid- epidemic in china (huang and zhao, ) . as well, like gender, for this sociodemographic variable, the prevalence of the depressive symptoms in different age groups and the probability of risk of developing depressive symptoms depending on age do not relate to the greater likelihood of being infected. being elderly has been reported to correlate with adverse clinical outcomes, including hospitalization and mortality (applegate and ouslander, ; zhou et al., ) . indeed, in italy the mean age of covid- patients who died was years (remuzzi and remuzzi, ) and the case fatality rate was % from to years, . % from to years, and % for years and older (livingston and bucher, ) . despite this, respondents older than years had the lowest risk for developing depressive symptoms compared to the younger age groups. in a population where loneliness and isolation have already been described as an epidemic (luo et al., ) , the impact of even short-term social distancing measures and the resulting distress did not influence the vulnerability to mental health issues (jeste et al., ; vahia et al., ) . this finding is in accordance with part of the literature. although mixed results derive across current and previous studies on the association between participants' age and depression as a psychological outcome of health-related emergency (hawryluck et al., ; qiu et al., ) , some authors reported that only young age was found to be associated with increased distress as a psychological outcome of the covid- quarantine (barari et al., ; and of the sars quarantine (hawryluck et al., ) . the higher psychological distress reported by the younger population could be due to their greater and uncontrolled access to the amount of information ("infodemic") through social media, which can easily trigger distress (cheng et al., ) . as regards marital status, unmarried/single people were the most depressed group with quarantine policies in the event of this outbreak. it is plausible that single people had greater difficulty in relying on or obtaining the assistance of others during the italian lockdown, thus are at risk of depressive symptoms and lower self-confidence more than cohabiting and married participants. this datum is in line with part of previous literature reporting that being married was protective for depression or associated with a lower risk of depressive symptoms (inaba et al., ; yan et al., ; bulloch et al., ) , although other studies conducted during the sars outbreak suggested that demographic factors such as marital status, as well as living with other adults, and having children were not associated with psychological outcomes (hawryluck et al., ; mihashi et al., ) . as to geographic area of residence, people living in the south of italy showed the lowest risk of developing depressive symptoms among all the groups, followed by participants from regions of central italy compared to residents from northern regions and the islands. as expected, residents in the most severely affected regions are at the highest risk of developing depressive symptoms. southern and central regions recorded a smaller number of deaths and diagnosed cases ( , and , deaths, respectively), compared to the north-east and north-west regions ( , and , deaths, respectively), where the disease spread first on a large scale. to explain this datum, it should be considered that the authorities introduced control measures in the northern regions (the "red zone"), before any other region and carried out extraordinary efforts to restrict the movement of people (carlucci et al., ) . in addition, residents from northern italy were found less adherent to restrictive measures compared to the those from the south of italy. people who have shown more adherence were found less at risk of depressive symptoms compared to people with less adherence (see under). as for occupational status, this study highlighted students as suffering from the highest level of psychological distress among all the other groups, including the unemployed group. also, in this case, the public's level of depressive symptoms did not increase with an increased probability of contracting the disease. since the physical spaces of universities were closed, students' mental well-being was affected by the sudden interruption of social interactions. however, the possibility of having online lessons and maintaining social contacts through social networks would not explain the onset of the depressive symptoms compared to other groups, for example employees who had been laid off or were retired. the reason for students' greater risk of depressive symptoms, reported also by wang et al. ( ) among the chinese population, could lie in a sense of uncertainty toward the future that this emergency, not only in health, but also in economic, social, and political areas, is eliciting all over the world (chong et al., ; wenzel et al., ; tan and enderwick, ) . compared with other professions and the general population, healthcare workers were associated with a lower risk of psychological outcomes compared to the unemployed in our sample. "learned helplessness" (seligman, ) could explain why health professionals were the least depressed group. after being exposed to inescapable difficult events, people become passive and stop trying after being exposed to events such as uncontrollable bursts of noise (alloy et al., ) and as a result show greater levels of anxiety and depression. on the contrary, health professionals, considered the real heroes of this emergency, were associated with a lower risk of psychological outcomes compared to unemployed participants. through the practice of their profession, they felt more useful to society, despite their increased risk for infection and transmission (al-rabiaah et al., ) . after students, the unemployed were at a higher risk of depressive symptoms compared to the other groups (stuckler et al., ; reeves et al., ) . this datum is inserted within the context of the covid- -related risk unemployment and economic losses and insecurity with the closure of community services and the collapse of industries negatively impacting the national economy. it should be a critical public health priority to prevent suicide. indeed, during the most recent economic recession, a % rise in unemployment was correlated with a rise in the suicide rate of . % in the united states ( % ci: . - . , p < . ) (reeves et al., ) . similarly, each percentage point increase in unemployment was accompanied by . % rise in suicide ( % ci: . - . , p = . ) in europeans aged years or less. as to adherence, people with low adherence were more likely to exhibited depressive symptoms relative to people with a great level of adherence to covid- preventive measures. as expected, adherence has been found to be a protective factor against mental health problems (hawryluck et al., ; koenig and schultz, ; brooks et al., ) . adopting the preventive behaviors contribute to lower the uncertainty of the epidemic progression which would cause higher psychological pressure on the public. as to worry about being infected with covid- , people with more worry were more depressed than people with a low level of worry. as for the whole sample, also in the subsample of participants with psychotherapeutic treatment history, the same sociodemographic factors and behaviors that contribute to, or mitigate mental effects of the quarantine in terms of depressive symptoms were reported. however, when anxiety symptoms or concomitant stressful events were present in comorbidity, these participants were found to experience higher levels of worry associated with increased depression symptomatology, compared to the whole sample. current evidence showed similar results. a history of stressful situations and medical problems was associated with a greater degree of depression and anxiety during the covid- quarantine in the italian population (hao et al., ) . again, psychiatric patients were significantly more likely to experience a higher degree of the negative mental impact of the outbreak, including stress, anxiety, and depression, compared to the general public (hao et al., ) . during the covid- quarantine outbreak in italy, female participants, younger people, single people, students, people living in northern regions, and who were less compliant with quarantine guidelines and less worried about being infected with covid- were at a high risk of displaying psychological issues. these findings suggest public levels of depressive symptoms did not increase with the greater likelihood of being infected. for example, although female and younger people reported a lower risk of covid- infection, they experienced higher levels of depressive symptoms during the covid- quarantine in italy. therefore, ongoing monitoring of the psychological strain associated with outbreaks of epidemic-potential, life-threatening diseases should become routine as part of preparedness efforts worldwide by establishing early targeted mental health interventions. in other words, more vulnerable groups, likes those cited above, should benefit from personalized "moraleboosting" interventions. or, intervention research could be valuable to combat amplifications of gender inequalities, particularly to address the added challenges women are likely to face. this research can provide timely insights for government agencies toward improving and safeguarding the psychological well-being of women, younger people, and categories of subjects at a higher risk of suffering from psychological distress on the occasion of subsequent waves of the spread of covid- or other epidemic diseases. this study has several limitations. firstly, the analyses presented here were derived from a crosssectional design, thus it is difficult to make causal inferences. secondly, given that the research was conducted in close temporal proximity to the period of the covid- quarantine, a webbased survey method was necessary to recruit a convenience sample by avoiding possible infections. this limited sampling in our study. as a consequence, a self-selection effect may have occurred and should be considered with those people who were experiencing the greatest or least levels of distress responding to the survey (saggino et al., ) . in addition, participants were required to access the internet and to be familiar with online devices to respond, which suggests that they might be more educated, younger or/and have a higher socioeconomic status than the overall surveyed quarantined population. thirdly, due to the uncontrolled occurrence of this health-related emergency, an accurate picture of the individual's psychological conditions before the covid- outbreak was not conducted. although it would been interesting to conduct pre-post analyses, these data could provide a baseline for future research on the psychological consequences of quarantine in the italian population throughout the rest of the current covid- pandemic. fourthly, depressive and state anxiety symptoms were measured by means self-report inventories that are notably biased by response set, such as social desirability (innamorati et al., ; carlucci et al., ; balsamo et al., a,b) . thus, future research should include methods, such as observational methods and psychophysiological or behavioral assessment, in order to objectively record the levels of these mood states (campbell and fiske, ) . fifthly, the assessment of state anxiety could be completed or replaced by adding the specific fear of covid- scale (ahorsu et al., ; broche-pérez et al., ; soraci et al., ) . in addition, it should be acknowledged that the study was carried out was not sufficiently heterogeneous for sex, with marked female preponderance, and age sample, with a prevalence of juveniles. hence, these findings may not translate accurately to the public at large. finally, it could not evaluate whether the outcomes considered in this study will be long-lasting after the covid- outbreak. however, follow-up with these participants will continue in order to facilitate our understanding about how long these outcomes will last. a deeper understanding of how the epidemic affects italians' psychological health by identifying which groups were at a high risk of psychological morbidities due to the current pandemic can help to guide and target future psychological intervention implementations. the raw data supporting the conclusions of this article will be made available by the authors, without undue reservation, to any qualified researcher. the studies involving human participants were reviewed and approved by the department of psychological sciences, health and territory, university of chieti, italy, review board. the patients/participants provided their written informed consent to participate in this study. both authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication. we thank maria pia pugliese, ines d'ambrosio, and gianluca balsamo for their assistance and support in data collection and writing assistance. the supplementary material for this article can be found online at: https://www.frontiersin.org/articles/ . /fpsyg. . /full#supplementary-material the fear of covid- scale: development and initial validation attributional style and the generality of learned helplessness middle east respiratory syndrome-corona virus (mers-cov) associated stress among medical students at a university teaching hospital in saudi arabia covid- presents high risk to older persons a new measure for trait and state anxiety: the state trait inventory of cognitive and somatic anxiety (sticsa). standardization in an italian population assessment of anxiety in older adults: a review of self-report measures assessment of late-life depression via self-report measures: a review building a new raschbased self-report inventory of depression determining a diagnostic cut-off on the teate depression inventory evaluating covid- public health messaging in italy: self-reported compliance and growing mental health concerns. medrxiv mixed anxiety-depression and its implications for models of mood and anxiety disorders the pandemic paradox: the consequences of covid- on domestic violence gender and fear of covid- in a cuban population sample the psychological impact of quarantine and how to reduce it: rapid review of the evidence the depression and marital status relationship is modified by both age and gender the employment situation convergent and discriminant validation by the multitrait-multimethod matrix demographic and attitudinal factors of adherence to quarantine guidelines during covid- : the italian model religious fundamentalism and psychological well-being: an italian study dimensions of anxiety, age, and gender: assessing dimensionality and measurement invariance of the state-trait for cognitive and somatic anxiety (sticsa) in an italian sample psychological health diathesis assessment system: a nationwide survey of resilient trait scale for chinese adults psychological impact of severe acute respiratory syndrome on health workers in a tertiary hospital tripartite model of anxiety and depression: psychometric evidence and taxonomic implications parenting in a time of covid- statistical power analysis for the behavioral sciences released . ibm spss statistics for windows helping others regulate emotion predicts increased regulation of one's own emotions and decreased symptoms of depression impact of non-pharmaceutical interventions (npis) to reduce covid- mortality and healthcare demand the impact of quarantine and physical distancing following covid- on mental health: study protocol of a multicentric italian population trial decree of the president of the council of ministers generalized linear and generalized additive models in studies of species distributions: setting the scene do psychiatric patients experience more psychiatric symptoms during covid- pandemic and lockdown? a case-control study with service and research implications for immunopsychiatry the hmisc package the social cure: identity, health and well-being loneliness matters: a theoretical and empirical review of consequences and mechanisms. anna sars control and psychological effects of quarantine covid : impact of lock-down on mental health and tips to overcome mental health strategies to combat the psychological impact of covid- beyond paranoia and panic multidisciplinary research priorities for the covid- pandemic: a call for action for mental health science generalized anxiety disorder, depressive symptoms and sleep quality during covid- outbreak in china: a web-based cross-sectional survey depression in the united states and japan: gender, marital status, and ses patterns factor validity of the beck hopelessness scale in italian medical patients battling the modern behavioral epidemic of loneliness: suggestions for research and interventions advancing the social identity approach to health and well-being: progressing the social cure research agenda koenig and schultz's disaster medicine: comprehensive principles and practices the gendered dimensions of covid- recommendations on diagnostic criteria and prevention of sars-related mental disorders coronavirus disease (covid- ) in italy loneliness, health, and mortality in old age: a national longitudinal study the immediate psychological and occupational impact of the sars outbreak in a teaching hospital a nationwide survey of psychological distress among italian people during the covid- pandemic: immediate psychological responses and associated factors danger in danger: interpersonal violence during covid- quarantine generalized linear models predictive factors of psychological disorder development during recovery following sars outbreak a nationwide survey of psychological distress among chinese people in the covid- epidemic: implications and policy recommendations probabilistic models for some intelligence and attainment tests (reprint, with foreword and afterword by distinguishing cognitive and somatic dimensions of state and trait anxiety: development and validation of the state-trait inventory for cognitive and somatic anxiety (sticsa) increase in state suicide rates in the usa during economic recession covid- and italy: what next? understanding, compliance and psychological impact of the sars quarantine experience emotion elicits the social sharing of emotion: theory and empirical review conducting social psychological research in the wake of covid- covid- pandemic and lockdown measures impact on mental health among the general population in italy the psychological effects of quarantining a city a validation study of the psychometric properties of the other as shamer scale- learned helplessness world health organization declares global emergency: a review of the novel coronavirus (covid- ) validation and psychometric evaluation of the italian version of the fear of covid- scale the public health effect of economic crises and alternative policy responses in europe: an empirical analysis prevalence of psychiatric morbidity and psychological adaptation of the nurses in a structured sars caring unit during outbreak: a prospective and periodic assessment study in taiwan managing threats in the global era: the impact and response to sars. thunder. int factors influencing psychological distress during a disease epidemic: data from australia's first outbreak of equine influenza the outbreak of covid- coronavirus and its impact on global mental health covid- , mental health and aging: a need for new knowledge to establishing a trait anxiety threshold that signals likelihood of anxiety disorders covid- : reducing the risk of infection might increase the risk of intimate partner violence immediate psychological responses and associated factors during the initial stage of the coronavirus disease (covid- ) epidemic among the general population in china lessons from severe acute respiratory syndrome (sars): implications for infection control marital status and risk for late life depression: a meta-analysis of the published literature clinical course and risk factors for mortality of adult inpatients with covid- in wuhan, china: a retrospective cohort study the authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.copyright © balsamo and carlucci. this is an open-access article distributed under the terms of the creative commons attribution license (cc by). the use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. no use, distribution or reproduction is permitted which does not comply with these terms. key: cord- -vl zhxyh authors: giallonardo, vincenzo; sampogna, gaia; del vecchio, valeria; luciano, mario; albert, umberto; carmassi, claudia; carrà, giuseppe; cirulli, francesca; dell’osso, bernardo; nanni, maria giulia; pompili, maurizio; sani, gabriele; tortorella, alfonso; volpe, umberto; fiorillo, andrea title: the impact of quarantine and physical distancing following covid- on mental health: study protocol of a multicentric italian population trial date: - - journal: front psychiatry doi: . /fpsyt. . sha: doc_id: cord_uid: vl zhxyh the covid- pandemic and its related containment measures—mainly physical distancing and isolation—are having detrimental consequences on the mental health of the general population worldwide. in particular, frustration, loneliness, and worries about the future are common reactions and represent well-known risk factors for several mental disorders, including anxiety, affective, and post-traumatic stress disorders. the vast majority of available studies have been conducted in china, where the pandemic started. italy has been severely hit by the pandemic, and the socio-cultural context is completely different from eastern countries. therefore, there is the need for methodologically rigorous studies aiming to evaluate the impact of covid- and quarantine measures on the mental health of the italian population. in fact, our results will help us to develop appropriate interventions for managing the psychosocial consequences of pandemic. the “covid-it-mental health trial” is a no-profit, not-funded, national, multicentric, cross-sectional population-based trial which has the following aims: a) to evaluate the impact of covid- pandemic and its containment measures on mental health of the italian population; b) to identify the main areas to be targeted by supportive long-term interventions for the different categories of people exposed to the pandemic. data will be collected through a web-platform using validated assessment tools. participants will be subdivided into four groups: a) group —covid- quarantine group. this group includes the general population which are quarantined but not isolated, i.e., those not directly exposed to contagion nor in contact with covid- + individuals; b) group —covid- + group, which includes isolated people directly/indirectly exposed to the virus; c) group —covid- healthcare staff group, which includes first- and second-line healthcare professionals; d) group —covid- mental health, which includes users of mental health services and all those who had already been diagnosed with a mental disorder. mental health services worldwide are not prepared yet to manage the short- and long-term consequences of the pandemic. it is necessary to have a clear picture of the impact that this new stressor will have on mental health and well-being in order to develop and disseminate appropriate interventions for the general population and for the other at-risk groups. the covid- pandemic and its related containment measures-mainly physical distancing and isolation-are having detrimental consequences on the mental health of the general population worldwide. in particular, frustration, loneliness, and worries about the future are common reactions and represent well-known risk factors for several mental disorders, including anxiety, affective, and post-traumatic stress disorders. the vast majority of available studies have been conducted in china, where the pandemic started. italy has been severely hit by the pandemic, and the socio-cultural context is completely different from eastern countries. therefore, there is the need for methodologically rigorous studies aiming to evaluate the impact of covid- and quarantine measures on the mental health of the italian population. in fact, our results will help us to develop appropriate interventions for managing the psychosocial consequences of pandemic. the "covid-it-mental health trial" is a no-profit, notfunded, national, multicentric, cross-sectional population-based trial which has the following aims: a) to evaluate the impact of covid- pandemic and its containment measures on mental health of the italian population; b) to identify the main areas to be targeted by supportive long-term interventions for the different categories of people exposed to the pandemic. data will be collected through a web-platform using validated assessment tools. participants will be subdivided into four groups: a) group -covid- quarantine group. this group includes the general population which are quarantined but not isolated, i.e., those not directly exposed to contagion nor in contact with covid- + individuals; b) group -covid- + group, which includes isolated people directly/indirectly exposed to the virus; c) group -covid- healthcare staff group, which includes firstand second-line healthcare professionals; d) group -covid- mental health, which includes users of mental health services and all those who had already been diagnosed with a mental disorder. mental health services worldwide are not prepared yet to manage the short-and long-term consequences of the pandemic. it is necessary to have a clear picture of the impact that this new stressor will have on mental health and well-being in order to develop and disseminate appropriate interventions for the general population and for the other at-risk groups. keywords: pandemic, global mental health, post-traumatic stress disorder, burn-out, anxiety, depression, resilience background the ongoing covid- pandemic represents an unprecedented event in terms of consequences for physical and mental health of individuals and for the society at large ( ) ( ) ( ) ( ) . in order to reduce the spread of the virus, national and international bodies and institutions have ordered quarantine, physical distancing, and isolation almost everywhere in the world. however, the psychological consequences of quarantine, such as frustration, loneliness, and worries about the future are well-known risk factors for several mental disorders, including anxiety, affective disorders, and psychoses ( ) ( ) ( ) . from a medical and sociological viewpoint, the pandemic caused by covid- represents a unique event, since it does not resemble any other previous traumatic event, such as earthquakes or tsunamis ( ) . in those cases, the traumatic factors are usually limited to a specific area and to a given time; affected people know that they can "escape" from the event. on the contrary, in the case of covid- pandemic, the "threat" can be everywhere and can be carried by every person next to us ( ) ( ) ( ) . therefore, people living in cities most severely impacted by the pandemic are experiencing extremely high levels of uncertainties, worries about the future and fear of being infected. the only comparable studies are those carried out during the sars outbreak ( ) ( ) ( ) ( ) ( ) . those studies showed that people experienced fear of falling sick or dying, feelings of helplessness, increased levels of self-blame, fear, and depression ( ) ( ) ( ) ( ) . during quarantine and physical distancing, internet and the social media can be useful in reducing isolation and increasing opportunities to keep in contact with family members, friends, and co-workers at any time ( , ) . however, internet may also represent a risk factor for mental disorders, in particular internet gaming disorder. moreover, internet can also have a negative impact on mental health of the most vulnerable people, such as those who live alone or the elderly, since it spreads an uncontrolled amount of information (a situation known as "infodemic"). in the current pandemic, the impact of quarantine and physical distancing on the mental health of the general population has been explored only in a few studies, mostly conducted in china, where the pandemic started ( ) ( ) ( ) . qiu et al. ( ) found that % of the population experienced psychological distress; in particular, those more vulnerable to stress and more likely to develop post-traumatic stress disorder were women and individuals aged between and years or older than years. moreover, people were more concerned about their own health and that of their family members, while less concerned about leisure activities and relationships with friends ( , ) . after china, italy has been the first country to face the contagion of covid- and one of the countries with the highest number of deaths due to this coronavirus (http://www. salute.gov.it/portale/nuovocoronavirus/). on march , the lockdown status has been declared by the italian government. this status included the definition of specific containment and quarantine measures, such as the interdiction of all public meetings and strict movement restrictions (i.e., possibility to go out only for working, serious health reasons, or other urgent needs). these containment measures have been prolonged until may . moreover, the expected psychosocial and emotional reactions to the pandemic observed in the general population may be significantly different in the chinese and italian populations due to their socio-cultural characteristics and historical contexts, which obviously impact on people's behaviors and attitudes. furthermore, the organization of public health system is different in italy compared to china and other eastern asian countries, also due to financial constraints. in fact, although in those countries the model of care has shifted in the last years to become more similar to a western model of care, it has to be acknowledged that years is a relatively short period of time, and differences may still persist. methodologically rigorous studies are needed in order to evaluate the impact of covid- and quarantine measures on the mental health of italian population. these data will help us to develop appropriate interventions for managing the psychosocial consequences of the pandemic ( ) ( ) ( ) . the present study has been developed with the aims to: a) evaluate the impact of covid- pandemic and its containment measures on mental health of the italian population; b) to identify the main areas to be targeted by supportive long-term interventions for the different categories of people exposed to the pandemic. the "covid-it-mental health trial" is a no-profit, not-funded, national, multicentric, cross-sectional population-based trial involving the following eleven sites: university of campania "luigi vanvitelli" (naples), università politecnica delle marche (ancona), università milano bicocca, università "statale" (milan), university of perugia, university of pisa, sapienza university of rome, "cattolica" university of rome, university of trieste, university of ferrara; the center for behavioral sciences and mental health of the istituto superiore di sanità (rome). the department of psychiatry of the university of campania "luigi vanvitelli" in naples is the coordinating center, which has originally conceived the study idea and design. an online survey has been set up through eusurvey, a web platform launched in by the european commission. the application, hosted at the department for digital services (dg digit) of the european commission, is available to all eu citizens at https://ec.europa.eu/eusurvey. the survey will be online from march to june , (https://ec.europa.eu/ eusurvey/runner/covidsurvey ). the survey takes approximately - min to be completed. participants can stop the survey at any time and save their answers as "draft" on the web-platform. furthermore, participants can interact with the principal investigator of the study and with all researchers through email messages at any time during and after study participation. participants will be subdivided into four groups: a) group -covid- quarantine group. this group includes the general population which are quarantined but not isolated, i.e., those not directly exposed to contagion nor in contact with covid- + individuals; b) group -covid- + group, which includes isolated people directly/indirectly exposed to the virus; c) group -covid- healthcare staff group, which includes firstand second-line healthcare professionals; d) group -covid- mental health, which includes users of mental health services and all those who had already been diagnosed with a mental disorder. the survey addresses the italian population aged over years through a multistep procedure: ) email invitation to health professionals and their patients; ) dissemination of the link through social media channels (facebook, twitter, instagram) and the mailing lists of national psychiatric associations; ) involvement of national associations of stakeholders (e.g., associations of users/carers); ) official communication channels (e.g., university websites; websites of the hospitals directly involved in the management of the pandemic). the invitation letter includes information on study purposes and confidentiality. the provision of the informed consent is mandatory in order to start the survey. the snowball sampling procedure-without the definition of strict inclusion/exclusion criteria (except that of age limit)-will give us the opportunity to recruit a large sample of the italian population and to evaluate the effect of the studied variables on the outcome measures. the survey includes the following self-reported questionnaires: the general health questionnaire - items (ghq- ) ( ); the depression, anxiety and stress scale - items (dass- ) ( ); the obsessive-compulsive inventory -revised (oci-r) ( ); the insomnia severity index ( ) ; the severity-of-acute-stress-symptoms-adult ( ); the suicidal ideation attributes scale (sidas) ( ); the impact of event scale - items ( ); the ucla loneliness scale -short version ( ) ; the brief cope ( ); the post traumatic growth inventory short form ( ) ; the connor-davidson resilience scaleshort form ( ) ; the multidimensional scale of perceived social support ( ); the pattern of care schedule (pcs)-modified version ( ); the maslach burnout inventory (only for health professionals) ( ) . respondents' main socio-demographic characteristics, as well as data on their internet use, will be collected through an ad hoc schedule. all assessment instruments used for the study are detailed in table . the primary outcome of the study is the global score at the dass- . this choice is due to the fact that this assessment measure has already been used in a large population study carried out in china, thus giving us the opportunity to compare the italian situation with the chinese one ( ) . our study hypothesis is that the pandemic and the related containment measures are associated with higher levels of depressive and anxiety symptoms in the surveyed population compared to a community italian sample not exposed to the pandemic ( ) . furthermore, a significant difference between groups will be identified (covid- quarantine group = covid- healthcare professional second-line < covid- + group = covid- healthcare professional first-line group < covid- mental health group). in the covid- quarantined group, the severity of obsessivecompulsive symptoms, evaluated through the oci-r, the perceived loneliness and suicidal ideation will be considered as secondary outcome measures. in the covid- + patient group, the severity of post-traumatic symptoms at the severity-of-acute-stress-symptoms-adult scale will be considered. the hypothesis is that post-traumatic symptoms are more severe in this group compared to the other ones. in the covid- health staff group, the presence of burn-out symptoms, in particular mental exhaustion, and suicidal ideation will be considered. we anticipate that first-line professionals will report higher levels of mental exhaustion and suicidal ideation compared to second-lines staff members. in the covid- mental health group, the secondary outcome measures will include the adoption of maladaptive coping strategies (e.g., drinking alcohol) and a poor resilience style. patients with pre-existing mental disorders are expected to adopt more maladaptive coping strategies and poorer resilience styles compared to the other three groups. the use of internet and social media will be tested as possible moderator of the impact of pandemic and quarantine ( figure ) . moreover, the exposure time to covid- and to the related containment measures will be tested as possible mediators of the severity of the clinical symptomatology. finally, the other exploratory outcomes will include the variety of coping strategies and resilience styles as well as the different levels of post-traumatic growth. statistical analyses will be conducted according to a multistep plan. missing data will be handled using the multiple imputation approach ( ) . descriptive statistics will be calculated for the dependent and confounding variables. a bilateral alpha of . is considered, and error and confidence intervals are calculated at %. the analytic plan will include: ) data cleaning of the online dataset and replacement of missing values; ) descriptive statistics of the general characteristics of the recruited sample, in terms of levels of depressive and anxiety symptoms, posttraumatic and stress-related symptoms, insomnia, satisfaction with life, suicidal ideation, hopelessness, post-traumatic growth, resilience, coping strategies, and social support; ) sub-groups analyses based on the level of exposure to the pandemic (i.e., covid- quarantine group vs. covid- + patients group vs. covid- healthcare staff group vs. covid- mental health group); ) calculation of a propensity score, in order to adjust our findings for the likelihood of being exposed to the pandemic and to the quarantine ( , ) . this method is adopted since it produces a better adjustment for differences at baseline, rather than simply including potential confounders in the multivariable models. the independent variables used for calculating the propensity score will include gender, age, socio-economic status, and geographical region. the obtained propensity score will be used to weight the observations in the multivariable analyses. in the final regression model, the inverse probability weights, based on the propensity score, will be applied in order to model for the independence between exposure to the pandemic/ quarantine and mental health outcomes and estimation of causal effects ( , ); ) development of a structural equation model (sem), in order to evaluate the possible role as mediators and moderators of coping strategies, post-traumatic growth and usage of social networks on the severity of depressive and anxiety symptoms, post-traumatic and stress-related symptoms, suicidal ideation, and hopelessness. in order to improve the external validity and generalizability of our findings, all analyses will be controlled for the impact of confounding variables, such as age, gender, and geographical region. data will be stored in an online dataset by the coordinating center. for safety reasons, the dataset will be protected by a twostep password. it will be possible to export data in compatible formats with common calculation software (e.g., microsoft access and excel) and in specific softwares (e.g., spss and stata) for the statistical analyses. this study is being conducted in accordance with globally accepted standards of good practice, in agreement with the declaration of helsinki and with local regulations. the study protocol has been approved by the ethical review board of the university of campania "l. vanvitelli" (protocol number: /i). our survey will give us the opportunity to describe the impact of the pandemic on the mental health of different subgroups of the italian population. in fact, the analyses will be run according to the four subgroups of respondents: the general population not directly affected by the virus (covid- quarantine group); people who have had a direct or indirect contact with the virus (covid- + patients group); those working in health care units as first or second-line staff (covid- healthcare staff group); people with mental health problems, independently from the contact with the virus (covid- mental health). this choice is due to the evidence that stress and traumas have a different impact on different target groups ( , ( ) ( ) ( ) . in the covid- -quarantine group, we anticipate that the pandemic and the related containment measures will increase the levels of stress, anxiety and depression, as well as other stress-related symptoms. in particular, physical distancing has obviously changed the patterns of daily routine in order to mitigate the spread of the disease, with serious consequences on mental health and well-being in both the short-and long-term ( ) . similar consequences would require immediate efforts for developing preventive strategies as well as direct interventions aiming to mitigate the impact of the outbreak on individual and population mental health. the longer the pandemic will last the most the ordinary life of the general population will be seriously affected. in particular, zhang et al. ( ) have highlighted the need to pay attention to the mental health of people who have not been directly infected by the virus though have been forced to stop all their activities during the outbreak. these people represent the most susceptible group to the detrimental impact of quarantine and physical distancing measures adopted during the lockdown. moreover, during the current pandemic, it is reasonable to expect that the incidence of severe mental disorders will increase, but also that of other mental health disturbances not reaching the threshold for a full-blown diagnosis ( ) . however, currently available data are based on studies carried out in china and the different socio-cultural context may limit the generalizability of findings to the italian and western contexts. therefore, we consider essential to collect italian data in order to develop data-driven guidelines for an adequate management of mental health problems during the emergency and the post-emergency phases. in fact, this survey will represent the starting point for developing, validating, and implementing psychosocial supportive interventions ( , ) , as discussed later in this paper. we hypothesized that internet and social media can play a buffering role in the development of psychiatric symptoms ( , ) . it may be that online contacts and interactions will limit the detrimental effects of social isolation ( ) . moreover, internet can represent the ideal setting for providing supportive interventions through tele-mental health applications ( - ). however, the positive effect of internet and social media has to be confirmed yet, since it is only speculative at this stage. in the covid- + patient group (i.e., those with a direct or indirect contagion), the impact on mental health has been mostly neglected during the acute emergency phase. of course, this has been due to the fact that the infection is a potentially lifethreatening condition, as confirmed by the need for hospitalization in intensive care units for many patients ( ) . in particular, the experience of being isolated in the hospital, the perceived danger, uncertainty about own physical conditions and the fear of dying alone can be considered risk factors for the development of post-traumatic, anxiety, and depressive symptoms ( , ) . the only study conducted in china so far has documented that over % of covid+ patients admitted to the hospital reported significant post-traumatic stress symptoms ( , , ) . furthermore, the authors found that providing patients with psychoeducational intervention is well received and perceived as helpful and useful by users. as regards the effects on mental health of those working in health care units as first-line or second-line staff (covid- healthcare staff group), we expect that many health professionals will experience symptoms of burn-out, including mental exhaustion, irritability, detachment from reality, and insomnia. in a survey involving medical and non-medical health workers, zhang et al. ( ) found a higher prevalence of insomnia, anxiety, depressive symptoms, somatization, and obsessive-compulsive symptoms in mental health staff. moreover, front-line medical staff working in close contact with infected patients (e.g., staff professionals working in the departments of respiratory, emergency, infectious disease, and intensive care unit) showed higher scores on depressive/anxiety symptoms and had a twofold increase in risk to develop a mental health problem ( ) ( ) ( ) ( ) . however, the effect on suicidal ideation of health professionals has not been investigated yet and will be the focus of one of our work-packages. finally, the pandemic will affect the mental health status of people who already suffer from mental health problems, independently from the contact with the virus (covid- mental health group). although the effects of the coronavirus on mental health have not been systematically studied, it is likely that the covid- will have detrimental effects on patients with pre-existing mental health problems. many patients with severe mental disorders have been overlooked during the pandemic, although they can have a higher risk of contracting the virus and of death considering the higher prevalence of somatic comorbidities compared to general population and the difficulties in accessing health services ( ) . however, if protracted, social isolation may increase the risk of recurrences of episodes of mental disorders, beyond triggering the onset of new mental disorders in most vulnerable people. moreover, objective social isolation and subjective feelings of loneliness are associated with a higher risk of suicidal ideation and suicide attempts ( ) . for many persons with mental disorders, being alone is a heavy burden, far beyond that experienced by many other persons ( ) . in patients with pre-existing anxiety disorders or obsessivecompulsive disorder, we expect an exacerbation or worsening of their clinical symptoms. moreover, the fact that there is not (yet) a definitive treatment for the covid infection represents another potential stressor, further increasing the levels of anticipatory anxiety and reducing personal functioning. in our study, both obsessive-compulsive and anxiety symptom clusters will be evaluated through reliable and validated questionnaires. we believe that our study has several strengths, which should be highlighted. first, this is the first national multicentric, noprofit study carried out in italy with a rigorous methodology for evaluating the impact of pandemic and quarantine on mental health. second, the development of a web-based platform for data collection will give us the opportunity to recruit a high number of participants. based on previous population surveys carried out in italy, an ideal target would have been , participants, but this target has been reached in only days. therefore, we expect to reach more than , people within the study period. a third relevant strength of our study is the selection of validated and reliable assessment instruments, which are available and validated in several languages. the next step of the project will be to adapt our survey to the european level, by involving several countries. fourth, several psychopathological dimensions will be evaluated, not only those usually assessed following natural disasters, such as the post-traumatic and depressive-anxious dimensions. in this study, we will also evaluate the obsessive-compulsive spectrum, the suicidal ideation, the maladaptive use of internet, among the others, which represent novel targets for psychiatrists ( , ) . our study has obviously also some limitations. in particular, the study sample includes the adult population only, due to existing restrictions related to the provision of informed consent of children and adolescents in italy. however, it is likely that the pandemic will have a detrimental impact on the mental health of adolescents as well ( , ) . moreover, being exposed to a traumatic event during early life is associated with alterations in the social, emotional, and cognitive development and could determine a variety of impairment in the adulthood. the effects of the pandemic on children and adolescents will be evaluated in an ad hoc study, in which we will explore the relationship between parents and their underage children during the pandemic. another limitation is related to the recruitment process, which might partially bias our findings, since only persons interested in the topic of the survey may have voluntarily participated. however, we expect that most people are interested in participating in the survey given the global magnitude of the current traumatic threat with collective psychological and social reactions. another possible limitation of our study is the choice to use a web-based online survey, which may have limited the participation of people not having access to the internet or not familiar with online tools, particularly the elderly. the cross-sectional design of the study does not allow an evaluation of changes over time as regards the levels of severity of symptoms. however, in order to overcome this possible bias, we will compare our findings with those already available from the italian population ( ) and will adopt a propensity score approach in order to understand the impact of the duration of exposure to the pandemic on the risk of developing psychiatric symptoms. with this methodology, we will be able to evaluate the levels of post-traumatic growth and the type of resilience styles in the study population in order to identify possible critical areas to be targeted in the post-acute phase. however, these psychological constructs are slow to change, and this is why we will promote a second wave of the survey, which will start six months after the end of the "lockdown phase" in italy. finally, the survey link can be used multiple times in order to allow sharing and re-posting it. this methodological choice could bias the findings, since the same person can potentially compile the survey several times. however, this methodological choice was due to the adoption of the "snowball" sampling, and it is rather unlikely that someone can compile the same long survey more than once. based on the findings of this study and on our previous work in the development of psychosocial interventions ( - ), we aim to develop a psychosocial intervention which will include elements of classic psychoeducation, cognitive-behavioral therapy, and motivational intervention ( ) ( ) ( ) ( ) ( ) . in particular, we are developing an experimental intervention which includes information on the mental health consequences of the pandemic and on strategies to prevent them; practical advices for promoting healthy lifestyle behaviors (e.g., healthy eating, regular sleeping patterns, physical activity, etc.); stress-management techniques; communication strategies; problem-solving skills. based on participants' needs, additional sessions on suicide prevention, burn-out, and internet dependence may be provided. the intervention will include face-to-face sessions and telemental health sessions ( , ) . information will be provided through instant messages (e.g., chatbot), email contacts, and the development of an ad hoc app. the modules of the intervention will be adapted according to the characteristics and the needs of the four above-mentioned target groups. in particular, in the covid- quarantine group, the main focus of the intervention will be the improvement of healthy lifestyle behaviors; for the covid- + patients group, the intervention will include a specific focus on post-traumatic symptoms and on the risk of being socially stigmatized; for the covid- healthcare staff group, specific sessions will be dedicated to the burn-out syndrome and the management of stressful situations; for the covid- mental health group, sessions on resilience, coping strategies, and the detection of early warning signs of relapses will be included. the proposed experimental intervention will be tested in a randomized controlled trial which will start when the acute phase of the pandemic will be over, and the control group will be represented by an informative group intervention on the effects of the pandemic on mental health. moreover, our survey is going to be translated into different languages in order to assess the impact of the pandemic in other european countries. the pandemic and the quarantine may have a detrimental impact on mental health. an increase of psychiatric symptoms and of mental health problems in the general population is expected. most health professionals working in isolation units and resuscitation departments very often do not receive any training or support for their mental health care. mental health services worldwide are not prepared to manage the short-and long-term consequences of pandemic. it is necessary to have a clear picture of the impact that these new stressors are having on mental health and well-being in order to develop and disseminate appropriate preventive interventions for the general population as well as for the different atrisk groups. this study is being conducted in accordance with globally accepted standards of good practice, in agreement with the declaration of helsinki and with local regulations. the study protocol has been approved by the ethical review board of the university of campania "l. vanvitelli" (protocol number: /i). vg, gais, ml, vv, and af designed the study and wrote the protocol. ua, gc, cc, fc, bdo, mn, mp, gabs, at, and uv revised the draft of the paper. all authors contributed to the article and approved the submitted version. we are very grateful to the healthcare professionals, patients, and general population who have dedicated their time to participate in our study. the consequences of the covid- pandemic on mental health and implications for clinical practice the psychological impact of quarantine and how to reduce it: rapid review of the evidence the emotional impact of coronavirus -ncov (new coronavirus d i s e as e) . p s y c h ia global mental health: how are we doing public responses to the novel coronavirus ( -ncov) in japan: mental health consequences and target populations mental health care measures in response to the novel coronavirus outbreak in korea the mental health of medical workers in wuhan, china dealing with the novel coronavirus ecological disasters and mental health: causes, consequences, and interventions. front psychiatry ( ) social support, interpersonal, and community dynamics following disasters caused by natural hazards development of a comprehensive, sustained community mental health system in post-earthquake haiti the psychological effects of quarantining a city prevalence of psychiatric morbidity and psychological adaptation of the nurses in a structured sars caring unit during outbreak: a prospective and periodic assessment study in taiwan the immediate psychological and occupational impact of the sars outbreak in a teaching hospital severe acute respiratory syndrome (sars) in hong kong in : stress and psychological impact amongfrontline healthcare workers quality of life and psychological status in survivors of severe acute respiratory syndrome at months postdischarge stress and psychological distress among sars survivors year after the outbreak posttraumatic stress, anxiety, anddepression in survivors of severe acute respiratory syndrome (sars) wen soon s. psychosocial and coping responses within the community health care setting towards a national outbreak of an infectious disease psychosocial impact among the public of the severe acute respiratory syndrome epidemic in taiwan mental health burden for the public affected by the covid- outbreak in china: who will be the high-risk group? loneliness and social internet use: pathways to reconnection in a digital world quality of life, and mental health outcomes in post-disaster urban environments: the l'aquila earthquake experience impact of the covid- pandemic on mental health and quality of life among local residents in liaoning province, china: a cross-sectional study vicarious traumatization in the general public, members, and non-members of medical teams aiding in covid- control mental health and the covid- pandemic a nationwide survey of psychological distress among chinese people in the covid- epidemic: implications and policy recommendations public mental health crisis during covid- pandemic measuring and improving the quality of mental health care: a global perspective improving the quality of global mental health care requires universal agreement on minimum national investment prevention of depression will only succeed when it is structurally embedded and targets big determinants a user's guide to the general health questionnaire manual for the depression anxiety & stress scales the obsessive-compulsive inventory: development and validation of a short version the insomnia severity index: psychometric indicators to detect insomnia cases and evaluate treatment response american psychiatric association. severity of acute stress symptoms-adult (national stressful events survey acute stress disorder short scale [nsesss]) the suicidal ideation attributes scale (sidas): community-based validation study of a new scale for the measurement of suicidal ideation brief measure of posttraumatic stress reactions: impact of event scale- a short-form measure of loneliness you want to measure coping but your protocol' too long: consider the brief cope a short form of the posttraumatic growth inventory development of a new resilience scale: the connor-davidson resilience scale (cd-risc) the multidimensional scale of perceived social support family burden in bipolar disorders: results from the italian mood disorder study (imds) maslach burnout inventory immediate psychological responses and associated factors during the initial stage of the covid- ) epidemic among the general population in china the italian version of the depression anxiety stress scales- : factor structure and psychometric properties on community and clinical samples multiple imputation for nonresponse in surveys the propensity score an evaluation of inverse probability weighting using the propensity score for baseline covariate adjustment in smaller population randomised controlled trials with a continuous outcome the outbreak of covid- coronavirus and its impact on global mental health post-traumatic growth in mental health recovery: qualitative study of narratives mental health and wellbeing among italian medical students: a descriptive study the mental health consequences of covid- and physical distancing: the need for prevention and early intervention social distancing in covid- : what are the mental health implications? digital phenotyping: a global tool for psychiatry digital interventions in severe mental health problems: lessons from the actissist development and trial the "online brain": how the internet may be changing our cognition towards a consensus around standards for smartphone apps and digital mental health the efficacy of app-supported smartphone interventions for mental health problems: a meta-analysis of randomized controlled trials baseline characteristics and outcomes of patients infected with sars-cov- admitted to icus of the lombardy region posttraumatic stress symptoms and attitude toward crisis mental health services among clinically stable patients with covid- in china timely mental health care for the novel coronavirus outbreak is urgently needed icd- ptsd and complex ptsd: structural validation using network analysis post-traumatic stress disorder: a state-of-the-art review of evidence and challenges psychological status of medical workforce during the covid- pandemic: a cross-sectional study factors associated with mental health outcomes among health care workers exposed to coronavirus disease exploring ptsd in emergency operators of a major university hospital in italy: a preliminary report on the role of gender, age, and education dsm- ptsd and posttraumatic stress spectrum in italian emergency personnel: correlations with work and social adjustment the role of mental health home hospitalization care during the covid- pandemic mental health during and after the covid- emergency in italy patients with smi in the age of covid- : what psychiatrists need to know the role of psychiatry in modern medicine unmet needs in modern psychiatry intergenerational transmission of trauma effects: putative role of epigenetic mechanisms resilience from a developmental systems perspective efficacy of psychoeducational family intervention for bipolar i disorder: a controlled, multicentric, real-world study lifestyle working group. a randomized controlled trial on the efficacy of a psychosocial behavioral intervention to improve the lifestyle of patients with severe mental disorders: study protocol psychoeducational intervention for perinatal depression: study protocol of a randomized controlled trial family management of schizophrenia: a controlled study of clinical, social, family and economic benefits a protection motivation theory of fear appeals and attitude change toward a theory of motivational interviewing scaling up psychological treatments for common mental disorders: a call to action hickie ib. the role of new technologies in monitoring the evolution of psychopathology and providing measurement-based care in young people the authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. key: cord- -p ir rr authors: vigliar, elena; iaccarino, antonino; bruzzese, dario; malapelle, umberto; bellevicine, claudio; troncone, giancarlo title: cytology in the time of coronavirus disease (covid- ): an italian perspective date: - - journal: j clin pathol doi: . /jclinpath- - sha: doc_id: cord_uid: p ir rr introduction: the coronavirus disease (covid- ) is changing the way we practice pathology, including fine needle aspiration (fna) diagnostics. although recommendations have been issued to prioritise patients at high oncological risk, postponing those with unsuspicious presentations, real world data have not been reported yet. methods: the percentages of the cytological sample types processed at the university of naples federico ii, during the first weeks of italian national lockdown were compared with those of the same period in . results: during the emergency, the percentage of cytology samples reported as malignant increased (p< . ), reflecting higher percentages of breast (p= . ) and lymph nodes fnas (p= . ), effusions (p< . ) and urine (p= . ). conversely, thyroid fnas (p< . ) and pap smears (p= . ) were reduced. conclusions: even in times of covid- outbreak, cytological examination may be safely carried out in patients at high oncological risk, without the need to be postponed. the coronavirus disease (covid- ) is changing the way we practice pathology, including fine needle aspiration (fna) diagnostics. although recommendations have been issued to prioritise patients at high oncological risk, postponing those with unsuspicious presentations, real world data have not been reported yet. methods the percentages of the cytological sample types processed at the university of naples federico ii, during the first weeks of italian national lockdown were compared with those of the same period in . results during the emergency, the percentage of cytology samples reported as malignant increased (p< . ), reflecting higher percentages of breast (p= . ) and lymph nodes fnas (p= . ), effusions (p< . ) and urine (p= . ). conversely, thyroid fnas (p< . ) and pap smears (p= . ) were reduced. conclusions even in times of covid- outbreak, cytological examination may be safely carried out in patients at high oncological risk, without the need to be postponed. at the university of naples 'federico ii', cytopathology has been practised for many years. earlier activity was related to exfoliative cytopathology. then, following the example of the karolinska institute in sweden, a cytopathologist-run fine needle aspiration (fna) biopsy clinic was established. thus, since , cytopathologists have been performing fnas by themselves, ensuring correct sample management, rapid on-site evaluation (rose) and, more recently, ultrasound guidance. this activity, carried out for more than four decades, is now being challenged by the extraordinary outbreak of the coronavirus disease (covid- ) in italy. our country has overtaken china as the nation with the most coronavirusrelated deaths. from the epicentre, located in the north of the country, the disease is spreading more towards south, also involving the crowded naples, the third-largest city of italy after rome and milan. in a very dramatic speech delivered to the nation on march , the italian prime minister described the outbreak as the most difficult domestic crisis of the postwar period, deeming aggressive containment measures necessary, even if limiting freedom, privacy and individual rights. the national lockdown was imposed, and people were required to remain in their home. italians transformed their lives. accordingly, also our practice of cytopathology changed. due to the potential presence of the virus in cytology specimens, we adopted more stringent safety procedures, according to recent recommendations. - a policy of sample prioritisation was also implemented. while cytological screening activities were reasonably postponed, our cytology laboratory continued to function providing the best possible service to (potential) oncological patients. as a result, rate of malignancy increased. how long this pandemic will last is uncertain. however, we argue that in all cases, featuring a not negligible risk of a malignant disease, the cytological examination may be safely carried out without the need to be postponed. this study reports on the activity of the cytopathology laboratory at the university of naples 'federico ii', during the extraordinary covid- outbreak in italy. the first weeks of italian national lockdown were taken into consideration. to this end, all cytological reports issued from march to march were reviewed. to assess how this extraordinary time differed from a normal period, data were compared with those relative to march- march in . in both groups, the total number of samples processed was recorded. the number of pap smear, urine samples and effusions was detailed. the rate of thyroid, breast, lymph node, salivary gland and soft tissue ultrasound-guided fnas performed at our clinic was obtained. the distribution of the diagnostic classes was evaluated, taking into consideration the rate of inadequate/insufficient for diagnosis, negative for malignancy, atypical, benign neoplasm, suspicious for malignancy and malignant samples. between-years variations in percentages of sample type, site and rate of malignancy were evaluated using the χ test or the fisher's exact test, when appropriate. no adjustments were made for multiple comparisons. p values of less than . were deemed as statistically significant. data relative to the period from the march- march of were compared with those of the same days of when the covid- outbreak took place. during the healthy emergency, less samples (n= ) were processed compared with (n= ). exfoliative cytological samples dropped from n= to n= , while ultrasound-guided short report figure distribution of cytological sample types during the first weeks of italian national lockdown amid covid- outbreak compared with data relative to the activity from the same period in . distribution of the diagnostic classes relative to cytological samples processed during the first weeks of italian national lockdown amid covid- outbreak compared with data relative to the activity from the same period in . patients prioritisation in cytopathologist-run fna clinic. in patients at low oncological risk (ie, unsuspicious thyroid nodules), fnas are postponed or performed only after the clinicians confirmed its opportunity. in patients at higher oncological risk (ie, breast lumps, lymph nodes enlargement), fnas are performed as usual. to identify patients with covid- symptoms, a telephone triage is performed. fna, fine needle aspiration. fnas were reduced from n= to n= . in particular, differences were very evident when considering pap smears (n= , . % vs n= , . %; p= . ). urine (n= , . % vs n= , . %; p= . ) and effusion (n= , . % vs n= , . %; p< . ) also differed among the two groups. the number of thyroid fnas was greatly reduced (p< . ) dropping from n= ( . %) to n= ( . %). a relative increase in breast (n= , . % to n= , . %; p= . ) and in lymph nodes fnas (n= , . % to= , . %, p= . ) was also evident; less marked was the variation in salivary gland (n= , . % to n= , . %; p= . ) and soft tissue fnas (n= , . % to n= , . %; p= . ) (figure ). as far as the diagnostic category distribution is concerned, inadequate/insufficient for diagnosis samples were n= ( %) in and n= ( %) in (p= . ). negative for malignancy were n= ( %) in and n= ( %) in (p= . ); atypical cases were n= ( %) in and n= ( %) in (p= . ); neoplastic, benign neoplasms were n= ( %) in and n= ( %) in (p= . ). suspicious for malignancy were n= ( %) in and none in (p= . ); malignant cases were n= ( %) in and n= ( %) in . this latter difference was statistically significant (p< . ). data are summarised in figure . the covid- emergency forced in italy the major healthcare institutions to revisit their organisation and practices. while research and laboratory activities were suspended and administrative staff were allowed to work from home, our hospital management recommended to limit, as much as possible, outpatient visit, rescheduling clinic appointments that were not urgent ( figure ) . thus, the priority of periodic cervical cancer screening decreased and, compared with march , the reduction of the pap smears was marked (p= . ). similarly, in patients at low oncological risk, such as those with unsuspicious thyroid nodules, fna was postponed or performed only after the clinicians, including the general practitioners, confirmed its opportunity (figure ). relative increases were evident for enlarged lymph nodes (p= . ) and breast lumps (p= . ). considering the higher oncological risk, in these cases, fnas were directly performed; patients were contacted by telephone the day before the fna procedure to triage the presence of covid- symptoms (figure ). salivary gland (p= . ) and soft tissue (p= . ) fnas relatively increased only slightly. in contrast to pap smears reduction, a relative increase was observed for both effusions (p< . ) and urine (p= . ), which is conceivable since these specimens are usually collected from symptomatic or oncological patients during follow-up. the safety issues recently described by pambuccian in a detailed review on the implication of the covid- pandemic for cytopathology laboratories were carefully addressed. n many aspects our practice was revised to comply, as closely as possible, with the laboratory biosafety guidelines. - - briefly, the fna clinic used to be dailynow runs on a weekly basis. this allows accurate disinfection and sterilisation and nurse rotation. working in pairs, cytopathologists are faster, thus limiting the time spent in the clinic by each patient. staff wear personal protective masks with a filter respirator and face shields to protect the eyes. t t as a general rule, smears are ethanol fixed to avoid that fluid and tissue materials obtained by fna, expelled onto slides and allowed to dry, generated aerosols and droplets; these latter might contain viable and transmissible viruses. lymph nodes smears are still air-dried as rose is needed to triage for flow cytometry or cell-block preparation. safety procedures are also mandatory in laboratory when processing body fluids. these should be received, instead than fresh, ethanol fixed to ensure adequate virus inactivation. finally, our informed consent form was changed to allow the possibility for the patient to receive the final diagnostic report by email. in conclusion, here we report that, despite a drastic reduction in the number of the patient samples, sample prioritisation policy led our activity to yield a significant increase in the percentage of malignant cases diagnosed by cytological means (p< . ). despite the challenges of the covid- extraordinary outbreak, cytological examination may be safely carried out coronavirus disease (covid- ) in italy decree of the president of the council of ministers interim laboratory biosafety guidelines for handling and processing specimens associated with coronavirus disease (covid- ) laboratory biosafety guidance related to the novel coronavirus ( -ncov): interim guidance severe acute respiratory syndrome-related coronavirus ) -canada. ca safety considerations in the laboratory testing of specimens suspected or known to contain the severe acute respiratory syndrome coronavirus (sars-cov- ) the covid- pandemic: implications for the cytology laboratory raccomandazioni del comitato di citologia siapec-iap per la citologia dell'apparato respiratorio e del cavo orale in corso di emergenza da infezione covid- gestione del rischio biologico correlato alla epidemia di covid- nella manipolazione dei campioni tissutali e citologici, con particolare riguardo ai campioni a fresco o non adeguatamente fissati rcpath advice on histopathology frozen sections and cytology fine needle aspiration during infectious disease outbreaks key: cord- - bboc z authors: parola, anna; rossi, alessandro; tessitore, francesca; troisi, gina; mannarini, stefania title: mental health through the covid- quarantine: a growth curve analysis on italian young adults date: - - journal: front psychol doi: . /fpsyg. . sha: doc_id: cord_uid: bboc z introduction: health emergencies, such as epidemics, have detrimental and long-lasting consequences on people’s mental health, which are higher during the implementation of strict lockdown measures. despite several recent psychological researches on the coronavirus disease (covid- ) pandemic highlighting that young adults represent a high risk category, no studies specifically focused on young adults’ mental health status have been carried out yet. this study aimed to assess and monitor italian young adults’ mental health status during the first weeks of lockdown through the use of a longitudinal panel design. methods: participants (n = ) provided self-reports in four time intervals ( -week intervals) in month. the syndromic scales of adult self-report - were used to assess the internalizing problems (anxiety/depression, withdrawn, and somatic complaints), externalizing problems (aggressive, rule-breaking, and intrusive behavior), and personal strengths. to determine the time-varying effects of prolonged quarantine, a growth curve modeling will be performed. results: the results showed an increase in anxiety/depression, withdrawal, somatic complaints, aggressive behavior, rule-breaking behavior, and internalizing and externalizing problems and a decrease in intrusive behavior and personal strengths from t to t . conclusions: the results contributed to the ongoing debate concerning the psychological impact of the covid- emergency, helping to plan and develop efficient intervention projects able to take care of young adults’ mental health in the long term. the novel coronavirus disease (covid- ) is a highly infectious disease that began as a viral pneumonia in late december . in march , the world health organization (who) declared the state of pandemic. as rapidly pointed out (fiorillo and gorwood, ; jakovljevi et al., ) , the covid- global pandemic has affected-and is still affecting-not only physical health but also individual, family, and collective mental health. in line with recent studies (horesh and brown, ; masiero et al., ) , the covid- pandemic should be classified as a critical event with a potential traumatic nature, which may be overwhelming and could lead to complex emotional responses that can negatively affect individuals and collective psychological systems. starting with china and followed by other states, extraordinary measures and containment efforts (e.g., lockdown) aimed to prevent the high risk of contagion and limit the covid- outbreak have been adopted. in europe, italy was the first country that had to face the pandemic. here, on march , , strict lockdown measures were imposed by the government. a series of decrees imposed restrictions on the movements of individuals in the entire national territory from march until may . during the lockdown, people were allowed to leave their homes only for limited and documented purposes. schools, universities, theaters, and cinemas, as well as any shops selling non-essential goods were, therefore, temporarily closed. as previous studies demonstrated (tucci et al., ) , health emergencies, such as epidemics, have detrimental and longlasting consequences on people's mental health. concerning the covid- pandemic, initial studies carried out in china reported high levels of anxiety, depression, and trauma-related symptoms (qiu et al., ) , both during the epidemic peak and month later . moreover, the detrimental effect of epidemics on mental health seems to be higher during the implementation of strict lockdown measures. specifically, previous studies have associated quarantine with higher levels of trauma-related disorders (wu et al., ) , depression (hawryluck et al., ) , irritability and insomnia (lee et al., ) , acute stress (bai et al., ) , and avoidance behaviors and anger (marjanovic et al., ) . in a recent review, brooks et al. ( ) individuated major stress factors as being the long duration of quarantine, the fear of infection, the inadequate supplies and information, boredom, and frustration. in a recent italian study carried out during the third week of lockdown, cellini et al. ( ) have highlighted that italians reported high levels of depression, anxiety, and sleep disturbances. similarly, have found that high rates of negative mental health outcomes were seen in the general population weeks into the covid- lockdown. within the stream of research investigating the impact of quarantine during epidemics on individual's mental health, there have been very few longitudinal investigations aimed at understanding and monitoring the changes in the mental health status during quarantine (brooks et al., ) . where longitudinal research designs were carried out, they were limited to investigating people's mental health during and after quarantine (jeong et al., ; wang et al., ) . recent psychological research on covid- has also highlighted that specific target groups are more at risk than others to develop a wide variety of psychological problems, such as medical workers, marginalized people (i.e., homeless and migrants), and young adults. regarding young adults ( - years old), recent researches have highlighted that they present higher levels of anxiety, distress, and depression than do other adult groups (cao et al., ; huang and zhao, ; qiu et al., ) . these findings have also been confirmed in italy (rossi r. et al., ) . according to cheng et al. ( ) , one of the possible reasons can be found in young adults' tendency to obtain information from social media, which can represent a high stress factor for mental health. these initial findings strongly suggest the need to assess and monitor young adults' psychological situation during the epidemic and the weight of their mental health outcomes. to the best of the authors' knowledge, there are no previous studies specifically aimed at evaluating the impact of lockdown measures on italian young adults' mental health and monitor the changes in their mental health status. to fill this gap, the current study presents a longitudinal panel design aimed to assess the italian young adults' mental health status and monitor their mental health trends during the firsts weeks of lockdown imposed from the italian government during the covid- outbreak. on the basis of recent literature on the general population, an increase in mental health problems among young adults during quarantine was hypothesized. participants were enrolled online and provided self-reports over month ( -week intervals, t -t -t -t ). participants were considered eligible for participation if they met the following inclusion criteria: (a) were between and years and (b) were in a lockdown condition. exclusion criteria were as follows: (a) diagnosis of psychiatric disorder and/or psychopharmacological treatment (assessed with filter questions in the survey) and (b) not "absolute" lockdown condition (workers who were allowed to work outside their home during the lockdown measures). from the initial sample size of t (n = ), nine participants did not participate at t (n = ); four other participants did not participate at t (n = ); and other participants did not participate at t . these participants were, therefore, excluded from the data analysis. the final simple-size was composed of participants. approval from the university research ethics committee was obtained for collecting data. data collection took place during the italian lockdown from mid-march to mid-april . the administration took place in four time intervals ( -week intervals) in month. the first survey (t ) was made at the end of the first week of lockdown. the second survey (t ) coincided with the end of the second week of the lockdown. the third survey (t ) coincided with the end of the third week of the lockdown. the fourth survey (t ) coincided with the end of the fourth week of the lockdown. participants were informed about a complete guarantee of confidentiality and the voluntary nature of participation and their right to discontinue at any point. the enrollment procedure was carried out through an online advertising on social platforms. participants voluntarily accessed the online platform used for data collection once a week for the weeks of administration. to ensure anonymity, a request was made to create a personal identification code to be used for the four administrations. adult self-report (asr/ - ) the syndromic scales of adult self-report - (achenbach and rescorla, ) were used to assess the internalizing and externalizing problems. the asr is especially valuable when used routinely, as in this study design. the asr norms provide a standardized benchmark with which to compare what is reported by each individual. standardized reassessments over a regular interval enable to identify reported stabilities and changes in a group who have particular kinds of problems. in this case, the asr instrument was administered at regular intervals of week for weeks in the period of the italian lockdown. the asr was developed both to document specific problems and to identify syndromes of co-occurring problems. in this study, six specific syndromic scales, anxious/depressed, withdrawn, somatic complaints, aggressive behavior, rule-breaking behavior, and intrusive were used. anxious/depressed ( items) refers to anxiety and depressive symptoms (e.g., "i feel lonely" and "i am too fearful or anxious"). withdrawn ( items) mainly refers to attitudes of isolation and lack of contact with others (e.g., "i don't get along with other people" and "i keep from getting involved with others"). somatic complaints ( items) include physical illness, without a known medical cause (e.g., "i feel dizzy or lightheaded" and "physical problems without a known medical cause: stomachaches"). aggressive behavior ( items) includes behaviors and attitudes characterized by poor control of one's aggression (e.g., "i blame others for my problems" and "i scream or yell a lot"). rule-breaking behavior ( items) refers to transgressive behavior and violation of social norms (e.g., "i am impulsive or act without thinking" and "i lie or cheat"). intrusive ( items) refers to the difficulty faced in the interpersonal relationships and to the prevalence of intrusive behavior (e.g., "i damage or destroy my things" and "i drink too much alcohol or get drunk"). in addition, the broadband scales, internalizing and externalizing, were computed. internalizing problems reflect internal distress, while externalizing problems reflect conflicts with other people. the internalizing scale consists of the syndrome scales anxious/depressed, withdrawn, and somatic complaints, whereas the externalizing scale consists of aggressive behavior and rule-breaking behavior. moreover, the scale of personal strengths ( items) was used to assess the adaptive functioning of the individuals (e.g., "i try to get a lot of attention" and "i am louder than others"). the items are scored on a three-point rating scale: (not true), (somewhat or sometimes true), and (very true or often true); and a total score may be calculated. higher raw scores indicate more problematic behaviors on each scale. then, a normalized t score-weighted for sex and age-was assigned for the syndromic scales and to each internalizing and externalizing problem scales. raw scores of the both types of scales have been quantitatively converted in terms of gender-and agespecific t scores. clinical significant threshold is indicated by t-scores ≥ . borderline range is from to . the asr is a reliable and valid measure for the - general population (achenbach and rescorla, ) . cronbach's alpha (α) and mcdonald's omega (ω) are reported in table . statistical analyses were performed with r software (v. . . ; r core team, , and the following packages: psych (v. . . ; revelle, ), irr (v. . . ; gamer et al., ), lme (v. . - ; bates et al., ) , lmertest (v. . - ; kuznetsova et al., ), esvis (v. . . ; anderson, ) , aiccmodavg (v . - ; mazerolle, ), and ggplot (v. . . ; wickham, ) . no data were missing for any of the participants on any of the asr scales at any of the measurement points. reliability was evaluated by internal consistency analysis, using cronbach's alpha (α) and mcdonald's omega for categorical data (ω). first, the mean differences between the four time intervals (t , t , t , and t ) were performed. the unbiased sample estimate of standardized mean difference effect sizes (hedges' g; hedges, ) was performed, evaluating the magnitude of these differences. the following established ranges guide interpreting standardized mean difference magnitude: from . to . = small; from . to . = medium; and . = large (cohen, ) . growth curve analysis (gca) models were used to estimate the growth trajectories (i.e., slopes) of the syndromic scales of the asr-both internalizing and externalizing scales-and the personal strength scale. models also estimated subject variability in change across time, as represented in random-intercepts coefficients. parameters in each gca model were computed with maximum likelihood (ml) estimation. several models were estimated for each of the outcome variables, separately. specifically, it was hypothesized that the time (the week of quarantine) could have had an effect on the asr syndromic scales. in addition, it was also hypothesized that covariates, such as sex and the experience of covid- (exp-cvd ), intended as the experience of direct proximity with relatives and/or friends affected by covid- , could have had an effect on the shape of the growth curve across time. models were sequentially specified according to the guidelines (long, ; grimm et al., ) . first, a null model was estimated to provide a baseline comparison and to calculate the intraclass correlation coefficient (model -intercept only). second, a null model with covariates was specified (model -intercept model with covariates). third, a linear model with time as predictor and covariate interactions was estimated (model -linear model with covariates). fourth, a quadratic model was specified with linear interaction effects of the covariates (model -quadratic the best model fit was assessed with several indices. first of all, the likelihood ratio test (lrt) was performed between one model and the following one in a step-up approach analysis: model vs. model ; model vs. model ; model vs. model ; and model vs. model -the most parsimonious model will be preferred (long, ) . in addition, also "information criteria" indices were computed by comparing the abovementioned models. first, the schwarz bayesian information criterion (bic; schwarz, ; burnham and anderson, ) was calculated: the model with the lower bic indicated the best model-and it is recommended when model parsimony is overriding (kadane and lazar, ; long, ) . moreover, considering that the bic tends to favor simpler model (long, ) , the corrected akaike information criterion (aicc; akaike, ; azari et al., ) was also computed: even in this case, the model with the lower aicc indicated the best model. in addition, considering that-on a theoretical level-the bic is less desirable for model evaluation than the aicc (long, ) , several effect sizes based on the aicc were carried out: (i) the difference of aicc ( aicc); (ii) the weight of evidence (w h ): given a set of competing models and the unknowable true model, the w h indicates the probability that a model h is the best approximate model (the model with the large w h is the best-fitting model) (the more probable the model is, the best approximating the model will be to the true model); (iii) the evidence ratio (e h ) that expresses the difference-in oddsbetween the best-fitting model and the first worst-fitting model: the higher the e h , the more plausible is the best-fitting model. of participants, were male ( . %) and were female ( . %). the mean age of the sample was . (sd = . ; range = - ). a total of participants ( . %) had experienced proximity with a covid- -infected relative or friend. most of the participants lived with their parents during the quarantine ( . %). all participants came from the campania region, in southern italy, and attended the university. means and standard deviations between the four time intervals (t , t , t , and t ) and the effect size of means difference (hedges' g) are displayed in table . the preliminary analysis showed that the increments tended to be small from t to t for each syndromic scale and breadboard scale ( . was the highest value). from t to t , the results highlighted a medium increase for the anxious/depressed, withdrawn, and internalizing scales. from t to t , the increase was null. for somatic complaints, aggressive behaviors, rule-breaking behavior, and externalizing scales, the magnitude of the effect size was medium only considering the increments from t to t . across the weeks of quarantine, the somatic complaints scale increased with an almost null effect. finally, the personal strengths showed a small increase only from t to t and from t to t . scatterplot (figure ) showing the change of the syndromic scales and broadband scales score over time. quadratic model with linear covariates interactions quadratic model with all covariates interactions scales, as well as the related broadband scales, across the weeks of quarantine. specifically, figure was split by sex (males vs. females), and figure was split by the experience of covid- (yes vs. no). finally, figure shows the interaction between sex and experience of covid- . the broken lines demarcate a borderline clinical range from the rd to th percentiles for the syndromic scales and from the th to th percentiles for the internalizing and externalizing broadband scales. scores above the top broken line, i.e., above the th percentile for the syndromic scales and above the th for the internalizing and externalizing broadband, indicate that the individual reported enough problems to be of clinical concern. scores below the bottom broken line is in the normal range. as show in figure , the anxious/depressed scale is above the clinical threshold in t , and the withdrawn scale is above the normal threshold in t with an increase in t . preliminary analyses (m. ) revealed that the variance related to the random intercept of the participants was equal to . . the null model with covariates (m. ) revealed a nonstatistically significant effect of the interaction between sex and exp-cvd (b = − . , se = . , t = . , p = . ) or their main effects (sex: the linear model with covariates (m. ) revealed a nonstatistically significant effect of the interaction between time and exp-cvd (b = . , se = . , t = . , p = . ) or the two simple main effects (sex: b = . , se = . , t = . , p = . ; exp-cvd : b = − . , se = . , t = − . , p = . ). however, the model revealed a statistically significant interaction effect between time and sex (b = − . , se = . , t = − . , p = . ) as well as the principal effect of time (b = . , se = . , t = . , p < . ). figure shows a greater increase in males from t to t and from t to t than in females. the preliminary analyses (m. ) revealed that the variance related to the random intercept of the participants was equal to . . the suggested that this model had % probability of being the best approximate model (w h of m. was %), and the eh suggested that m. had a weight of evidence almost two times ( . ) greater than m. of being the best approximate model ( table ) . preliminary analyses (m. ) revealed that the variance related to the random intercept of the participants was equal to . . the null model with covariates (m. ) revealed a nonstatistically significant effect of the interaction between sex and the comparison of the different multilevel growth curve models suggested that the linear model with covariates (m. ) showed the lower bic and the lower aicc. the lrt showed that m. was statistically significantly different from m. (intercept model with covariates). however, although m. was not statistically significantly different from m. , it was the most parsimonious-and thus, it was chosen as the best model. however, the effect size indices suggested a negligible preference (table ) . the comparison of the different multilevel growth curve models suggested that the linear model with covariates (m. ) showed the lower bic and the lower aicc. the lrt showed that m. was statistically significantly different from m. (intercept model with covariates). however, despite that m. was not statistically significantly different from m. , it was the most parsimonious-and thus, it was chosen as the best model. however, the effect size indices suggested a small preference for m. . indeed, the aicc suggested a small difference m. and m. ( . ), the w h of m. indicates that this model had % probability of being the best approximate model, and the e h recommend that m. had a weight of evidence more than two times ( . ) greater than m. of being the best approximate model ( table ) . preliminary analyses (m. ) revealed that the variance related to the random intercept of the participants was equal to . . the ( . ) , the w h of m. suggested that this model had % probability of being the best approximate model (w h of m. was %), and the e h suggested that m. had a weight of evidence two times ( . ) greater than m. of being the best approximate model ( table ) . the comparison of the different multilevel growth curve models suggested that the quadratic model with linear covariates interaction (m. ) showed the lower bic and the lower aicc. the lrt showed that m. was statistically significantly different from m. (linear model with covariates). however, the lrt suggested that m. was not statistically significantly different from m. , but it was more parsimonious-and thus, m. was chosen as the best model. however, the effect size indices suggested a small preference for m. . indeed, the aicc suggested a small difference m. and m. ( . ), the w h of m. suggested that this model had % probability of being the best approximate model, and the e h suggested that m. had a weight of evidence more than three times ( . ) greater than m. of being the best approximate model ( table ) . as stated above, in addition to being a public physical health emergency, the covid- pandemic also implies a global mental health emergency that may have a potential traumatic nature and provoke complex emotional responses that could negatively affect individual and collective mental health (jakovljevi et al., ; masiero et al., ) . therefore, this global pandemic constantly requires researchers and professionals to monitor and assess the current mental health situation, in order to plan and develop efficiency-driven strategies aimed to reduce its negative psychological impacts. this study assessed and monitored italian young adults' mental health status during the firsts weeks of lockdown imposed by the government during the covid- outbreak, from march to april . to the authors' knowledge, this is the first study specifically focused on young adults' mental health status during covid- quarantine, both in italy and worldwide. a longitudinal panel design was carried out in order to assess internalizing and externalizing problems on italian young adults living in the campania region, southern italy. a gca (jackson et al., ) was performed to monitor the changes during the first weeks of quarantine. first of all, in line with the global trend reported by previous studies carried out on the general population (cao et al., ; huang and zhao, ; qiu et al., ; rossi r. et al., ) , this study confirmed the negative behavioral and emotional responses provoked by covid- quarantine and also highlighted the high vulnerability of young adults in developing psychological distress. comparing the internalizing and externalizing domains, the results showed an analogous increase for both areas from t to t , even though higher rates of internalizing manifestations were registered. specifically, the growth curve modeling highlighted that, within the internalizing problems area, the levels of anxiety/depression, withdrawal, and somatic complaints overall increased from t to t , showing an increase while the lockdown measures were in place. in this context, in line with results obtained on medical health workers (zhang et al., ) , having experienced a closeness with a covid- -infected relative or friend resulted in an increase of somatic complaints. similarly, within the externalizing problems area, the levels of aggressive behavior and rule breaking behavior increased from t to t . among the internalizing domains, youth reported clinical-level symptoms of anxiety and depression. according to the recent review on the psychological impact of quarantine (rajkumar, ), anxiety as well as depressive symptomatology was the most common. furthermore, the results showed that withdrawal level was above the normal threshold. this finding could be related to the specific situation of quarantine and the impossibility to engage in social behaviors due to the lockdown. indeed, the physical distance can intensify feelings of loneliness that in turn trigger intense anxiety (boffo et al., ; banerjee and rai, ; . if, broadly, the results obtained confirmed the general detrimental effects of social isolation due to epidemics on young adults' mental health (hawryluck et al., ; tucci et al., ; qiu et al., ; wang et al., ) , some brief reflections need to be outlined about the specificities of young adults' condition. indeed, young adults live a specific transition period in which their identity development process is based and founded on continuous affective investments on social and extra-familiar relationships (sica et al., ) . in this context, the lockdown measures may be interpreted as a forced regression that triggers negative mental health outcomes even more. within the range from t to t , higher levels of internalizing and externalizing problems were registered at t , whereas a sort of stabilization from t to t emerged. the peak reported at t probably indicated a sort of gradual cognitive and emotional recognition experienced from young adults about the seriousness of the pandemic, which increased feelings of anxiety, depression and worry, and irritability and anger. regarding the stabilization of both internalizing and externalizing problems between t and t , these findings might need to be interpreted in relation to the specific historical context of the covid- pandemic in italy. specifically, t corresponded to the week from april to in which a double attitude was observed in italy. on the one hand, despite the lockdown, the italian "civil protection" continued to alert the general population about the very high levels of contagions; on the other hand, in that period, italians also started to receive the first information about the so-called "phase , " which followed the forced lockdown. it might be hypothesized that the high levels of viral load continued to worry participants, even though the closeness to phase assumed a sort of protective function regarding an eventual mental health worsening. in correspondence to the increase of mental health distress, the results also showed a gradual decrease of participants' perception of their personal strengths, suggesting the need for researchers to strengthen individual's psychological resources in order to mediate the individual reaction to the covid- pandemic (di giuseppe et al., ) . in conclusion, regarding gender differences, a significant increase of the levels of anxiety/depression from t to t and, to a lesser extent, from t to t in males than the females emerged. these findings were in line with previous studies that pointed out higher symptoms of anxiety and depression in condition of social isolations in boys than girls (troop-gordon and ladd, ; derdikman-eiron et al., ) . the results reported no other statistically significant differences between sex. these findings seemed to be in opposition with the recent studies that have investigated the impact of covid- on mental health and highlighted a higher vulnerability for women to develop negative mental health outcomes, as compared with men (qiu et al., ; rossi r. et al., ) . in the context of gender studies, a wide range of recent literature tended to connect these results to the reinforced gender inequalities promoted by the lockdown measures. according to these studies (adams-prassl et al., ; béland et al., ; etheridge and spantig, ) , in fact, during the lockdown measures, the increase of unemployment rates as well as the commitment into the domestic work and in the management of children has represented a high risk factor for women, compared with men. within the same interpretation field, the lack of significant gender differences as emerged by the results might be correlated to the same nature of the sample, which mostly involved university students who probably were involved in the same challenges and tasks and did not experienced greater or smaller efforts connected to specific gender roles, such as to outline differences. the present study is not free from limitations. first of all, the number of participants should be increased in future studies, and the results need to be replicated in other geographical areas to determine their generalizability. furthermore, the sample was only composed of university students who came from the campania region in southern italy where the covid- outbreak has been taken more under control. to assess the mental health of young people during the quarantine, only a self-report measure was used. consequently, the data may be influenced by a reporting bias (e.g., social desirability). moreover, despite the longitudinal panel, the study is an observational study. in this sense, experimental manipulations and a control group are lacking. future researches need to extend the young adults' mental health assessment to other italian regions, taking into consideration that in the south of italy, where the study was carried out, the covid- outbreak has been taken moderately and was under control, compared with the north. higher levels of distress might be hypothesized in places where very high numbers of losses and deaths have been registered. moreover, the present study investigated the internalizing and externalizing problems as individual responses to covid- pandemic; further investigations to measure the traumatic symptomatology and the characteristics of post-traumatic effects caused by such stressful events are needed (troisi, ; margherita and tessitore, ) . follow-up investigations are also needed. considering the high levels of withdrawal that emerged from the results, future investigations should explore the function and the role played by virtual environments and e-communities during pandemic in-depth, taking into account the roles played by the online environments and by the use of social media in terms of both risks and protective functions (faccio et al., ; gargiulo and margherita, ; procentese et al., ; boursier et al., ) . in this sense, future investigations might be also directed to investigate the changes in the dynamics of social and love relationships (mannarini et al., (mannarini et al., , a balottin et al., ; as well as the role of social support (ratti et al., ) postlockdown and post-pandemic. in conclusion, recognizing the fundamental value of qualitative investigations to shed light on the inner aspects and subjective meanings of personal experiences is also vital (margherita et al., ; felaco and parola, ; parola, ; parola and felaco, ; tessitore and margherita, ) . these are much needed actions in order to develop an indepth understanding of the emotional and affective dimensions connected to the experience of the covid- pandemic, as well as possible risk and protective factors for mental health. in conclusion, the present study could contribute to the ongoing debate concerning the psychological impact of the covid- emergency, helping to develop efficient and personcentered intervention projects able to take care of young adults' mental health in the medium and long terms, understanding their specific needs and susceptibilities (benedetto et al., ; parola and donsì, ; fusco et al., ) . this is even more urgent considering that despite the distressing and prolonged situation, a significant number of people avoid seeking psychological help . on the one hand, some of these people may be reluctant to seek professional help due to the associated stigma (mannarini et al., b (mannarini et al., , faccio et al., ; mannarini and rossi, ) . on the other hand, some individuals may deny the problem, leading them to think that it will probably resolve itself naturally (sareen et al., ; rossi ferrario et al., ; rossi ferrario and panzeri, ) , thus choosing to manage the psychological issue on their own (wilson and deane, ) . the datasets generated in this article are not readily available because to ensure the privacy of the participants. requests to access the datasets should be directed to ap, anna.parola@un ina.it. the studies involving human participants were reviewed and approved by the ethical committee of psychological research of university of naples federico ii and was carried out in accordance with the american psychological association rules. the patients/participants provided their written informed consent to participate in this study. ap developed the theoretical framework of the present study, designed the study, and developed the methodological approach. ar performed all the analyses and designed tables and figures. ft and gt led the literature search and interpretation of data. sm critically revised the manuscript. all authors read and approved the final version of the work. manual for the aseba adult forms & profiles inequality in the impact of the coronavirus shock: evidence from real time surveys information theory and an extension of the maximum likelihood principle esvis: visualization and estimation of effect sizes longitudinal data model selection survey of stress reactions among health care workers involved with the sars outbreak triadic interactions in families of adolescents with anorexia nervosa and families of adolescents with internalizing disorders social isolation in covid- : the impact of loneliness fitting linear mixedeffects models using lme the short-term economic consequences of covid- : exposure to disease, remote work and government response metacognitive beliefs and alcohol involvement among adolescents exploratory structure equation modeling of the ucla loneliness scale: a contribution to the italian adaptation objectified body consciousness, body image control in photos, and problematic social networking: the role of appearance control beliefs the psychological impact of quarantine and how to reduce it: rapid review of the evidence model selection and multimodel inference: a practical information-theoretic approach the psychological impact of the covid- epidemic on college students in china changes in sleep pattern, sense of time and digital media use during covid- lockdown in italy psychological health diathesis assessment system: a nationwide survey of resilient trait scale for chinese adults statistical power analysis for the behavioral sciences gender differences in subjective wellbeing, self-esteem and psychosocial functioning in adolescents with symptoms of anxiety and depression: findings from the nord-trøndelag health study psychological impact of covid- among italians during the first week of lockdown the gender gap in mental well-being during the covid- outbreak: evidence from the uk (no. - ). institute for social and economic research the power of weight and the weight of power in adolescence: a comparison between young and adult women young in university-work transition. the views of undergraduates in southern italy the consequencies of the covid- pandemic on mental health and implications for clinical practice from creativity to future: the role of career adaptability irr: various coefficients of interrater reliability and agreement narratives of self-harm: the experience of young women through the qualitative analysis of blogs growth modeling: structural equation and multilevel modeling approaches sars control and psychological effects of quarantine distribution theory for glass's estimator of effect size and related estimators traumatic stress in the age of covid- : a call to close critical gaps and adapt to new realities mental health burden for the public affected by the covid- outbreak in china: who will be the high-risk group? brief strategic therapy and cognitive behavioral therapy for women with binge eating disorder and comorbid obesity: a randomized clinical trial one-year follow-up covid- pandemia and public and global mental health from the perspective of global health security mental health status of people isolated due to middle east respiratory syndrome methods and criteria for model selection lmertest package: tests in linear mixed effects models the experience of sars-related stigma at amoy gardens longitudinal data analysis for the behavioral sciences using r a rasch-based dimension of delivery experience: spontaneous vs. medically assisted conception etiological beliefs, treatments, stigmatizing attitudes toward schizophrenia. what do italians and israelis think? assessing conflict management in the couple: the definition of a latent dimension the role of secure attachment, empathic self-efficacy, and stress perception in causal beliefs related to mental illness -a cross-cultural study: italy versus israel assessing mental illness stigma: a complex issue how do education and experience with mental illness interact with causal beliefs, eligible treatments and stigmatising attitudes towards schizophrenia? a comparison between mental health professionals, psychology students, relatives and patients a comparison between pro-anorexia and non-suicidal self-injury blogs: from symptom-based identity to sharing of emotions italian validation of the capacity to love inventory: preliminary results teen mothers who are daughters of teen mothers: psychological intergenerational dimensions of early motherhood from individual to social and relational dimensions in asylum-seekers' narratives: a multidimensional approach the relevance of psychosocial variables and working conditions in predicting nurses' coping strategies during the sars crisis: an online questionnaire survey from individual to social trauma: sources of everyday trauma in italy, the us and uk during the covid- pandemic aiccmodavg: model selection and multimodel inference based on (q)aic(c) novel coronavirus outbreak and career development: a narrative approach into the meaning of italian university graduates suspended in time. inactivity and perceived malaise in neet young adults time perspective and employment status: neet categories as negative predictor of future a narrative investigation into the meaning and experience of career destabilization in italian neet families and social media use: the role of parents' perceptions about social media impact on family systems in the relationship between family collective efficacy and open communication a nationwide survey of psychological distress among chinese people in the covid- epidemic: implications and policy recommendations social support, psychological distress and depression in hemodialysis patients r: a language and environment for statistical computing. r foundation for statistical computing. vienna: r core team covid- and mental health: a review of the existing literature psych: procedures for personality and psychological research the italian version of the attitudes toward seeking professional psychological help scale -short form: the first contribution to measurement invariance the anxiety-buffer hypothesis in the time of covid- : when self-esteem protects from loneliness and fear to anxiety and depression covid- pandemic and lockdown measures impact on mental health among the general population in italy. an n= web-based survey exploring illness denial of lvad patients in cardiac rehabilitation and their caregivers: a preliminary study development and psychometric properties of a short form of the illness denial questionnaire perceived barriers to mental health service utilization in the united states, ontario, and the netherlands estimating the dimension of a model i became adult when pathways of identity resolution and adulthood transition in italian freshmen's narratives pre-and postmigratory experiences of refugees in italy: an interpretative phenomenological analysis female nigerian asylum seekers in italy: an exploration of gender identity dimensions through an interpretative phenomenological analysis. health care women int land of care seeking: pre-and postmigratory experiences in asylum seekers' narratives measuring intimate partner violence and traumatic affect: development of vita, an italian scale trajectories of peer victimiza-tion and perceptions of the self and schoolmates: precursors to inter-nalizing and externalizing problems the forgotten plague: psychiatric manifestations of ebola, zika, and emerging infectious diseases a longitudinal study on the mental health of general population during the covid- epidemic in china ggplot : elegant graphics for data analysis brief report: need for autonomy and other perceived barriers relating to adolescents' intentions to seek professional mental health care the psychological impact of the sars epidemic on hospital employees in china: exposure, risk perception, and altruistic acceptance of risk mental health and psychosocial problems of medical health workers during the covid- epidemic in china key: cord- - f o l authors: carpinelli mazzi, michele; iavarone, alessandro; musella, caterina; de luca, marzia; de vita, dalila; branciforte, stefano; coppola, alessia; scarpa, rosa; raimondo, sabina; sorrentino, sara; lualdi, flavia; postiglione, alfredo title: time of isolation, education and gender influence the psychological outcome during covid- lockdown in caregivers of patients with dementia date: - - journal: eur geriatr med doi: . /s - - -z sha: doc_id: cord_uid: f o l purpose: the study evaluated the effects of the covid- emergency lockdown on the psychological outcome in caregivers (children or spouses) of patients with dementia and the loss of the welfare services in these patients. methods: zung’s depression and anxiety assessment scales and the perceived stress scale were administered by a telephone interview or a self-compilation directly on the online platform. results: the sample consisted of participants (men = ; women = ) with a mean age of . years (sd = . ). education was associated with significantly lower overall anxiety and depression scores while days of isolation and female gender were associated with the higher scores. a marked reduction of health services was observed in all patients. conclusion: the lockdown registered a particular impact on people with dementia and their caregivers. many people with dementia were deprived of care services and time of isolation had a significant negative effect on anxiety and depression in caregivers. the covid- emergency has forced many countries to implement measures to contain the epidemic. in italy, the lockdown lasted almost months with severe restrictions in an individual movement. the persons who suffered most from this covid- related lockdown were probably those affected by dementia [ ] , a syndrome involving around million persons in italy [ ] and . in the southern region of campania [ ] . since dementia is a pathology requiring continuous care and complex therapies, the measures to contain the epidemic forced to a prolonged period of isolation and the suspension of non-urgent care and health services. this condition may have been a cause of significant psychological distress for caregivers. the aim of the present study was to analyse the effects of the lockdown on the psychological outcome in caregivers of persons with dementia and on the loss of welfare services in these patients. the campania section of the italian alzheimer's diseaseassociation (aima naples) interviewed caregivers of patient with dementia living in three cities of the campania region (naples, salerno and caserta). italian versions of zung's depression and anxiety assessment scales (zdaas) [ ] [ ] [ ] were used to analyze the affective domain and the italian version of the perceived stress scale (pss) [ , ] was used to assess the level of stress. specific questions were asked to describe the sociological profile of respondents and the affective and behavioral levels in caregivers of patients with dementia. a telephone interview or a self-compilation directly on the online platform were performed by caregivers of patients with dementia. all interviewed caregivers were relatives of people with dementia (sons, daughters, spouses or other relatives) and were in isolation for the national lockdown for covid- infection. the completion of the questionnaires and interviews took place soon after the lockdown and was carried out during all this period. this allowed to observe the affective, emotional and stress state in relation to the duration of isolation. descriptive statistic was used to summarize data. the effect of demographic variables on anxiety, depression and pss scores was examined using multiple regression analysis taking anxiety, depression and pps scores as dependent variable and age, education (years of schooling) and days of isolation as the independent one(s). the significance level was fixed at p = . , based on an overall p value < . level divided by the number of independent variables on a bonferroni basis. the anxiety, depression and stress scores of the caregivers of patients with alzheimer's disease (ad) were compared with those with other forms of dementia using a non-parametric test for independent samples (mann-whitney). the same test was used to compare the scores between men and women. all participants were screened with the zdaas and the pss scales, if possible, during two separate occasions. the sample consisted of participants (men = ; women = ). the mean age of caregivers was . years (sd = . ) and their mean number of years of school attendance was . (sd = . ). the mean age of patients with dementia was . years (sd = . ) and their mean number of years of school attendance was . (sd = . ). ad was diagnosed in the %, vascular dementia (vd) in the %, mixed dementia in the %, frontotemporal dementia (ftd) in the % and lewy body disease (lbd) in the % of the patients. % of the people interviewed was living in the same home with his/her patient with dementia while the other % was living nearby. half of the caregivers reported that since the beginning of isolation behavioral disorders and mood tone worsened, respectively, by - % in particular in patients with mild-moderate dementia. in the others, who were mostly affected by a severe form of dementia, the mood ( %) and behavior ( %) remained unmodified. helps support and assistance were discontinued in % of cases. among these, % lost the assistance of professional nurses and domestic aids; % had to forego health benefits (including medical examinations, physiotherapy, cognitive stimulation, home assistance); % suspended attendance at the daily centers; % lost the help of friends or neighbors. table presents the results of the multiple regression analyses, taking the anxiety, depression and stress scores as dependent variables and demographic variables and days of isolation as the independent(s) ones. in all cases, no effect of the age was observed. education was associated with significantly lower overall anxiety and depression scores. conversely, days of isolation were associated with higher scores in the anxiety and depression scales, but not with stress score. a significant difference was observed in anxiety, depression and stress scores between men and women: non-significant differences were recorded between caregivers of patients with ad and caregivers of patients with other forms of dementia, and among caregivers who live/do not live in the same home with people with dementia. in italy, the lockdown involved millions of people in the effort to contain the covid- epidemic. people with dementia and their families found themselves forced to deal with a very complex situation. this study involved caregivers of people with dementia. they were sons, daughters, spouses or relatives of the person with dementia. about half of those interviewed were living with the patient during the isolation period, while the second half was in isolation. according to caregivers, the lockdown had a negative effect on the behavior and mood of patients with mild-moderate dementia, but not in those with most severe forms. nearly half of the sample also reported having to give up welfare services, such as professional assistance, domestic aids, health services (medical examinations, physiotherapy, cognitive stimulation, integrated home assistance) and the presence at the day-centers. the lockdown seems to led to negative consequences for an important part of people with dementia both mentally and in terms of continuity of care. regarding the effects of the lockdown on the psychological outcome in caregivers, multiple linear regression showed that education and time spent in isolation had an important role in anxiety and depression. the higher the educational level, the lower the scores of anxiety and depression. education seems a protective factor towards reactions of depression and anxiety in the presence of stressors. the effect of years of school attendance on apathy was recently described with highly educated persons obtaining lower scores during the validation of an apathy scale [ ] . it is also known that years of school attendance have a significant effect on the performance on cognitive tests [ , ] . however, the effect of education on depression and anxiety symptoms appears in original data, and not clear evidences are reported in the literature. however, more days spent in isolation, the higher the levels of depression and anxiety were reported. gender seems to play an important role in levels of depression and anxiety: women had higher scores than those in men. however, it should be considered that cultural factors could influence the answers: during the interview, men may have wanted to show minor negative effects of the lockdown on their psychological condition than those reported in the women. this aspect needs further studies. a recent review describes how socio-economic, health and psychosocial factors can be considered important to explain the gender differences in psychological health; in fact, most of the gender-related variance was found in countries where the gender gap is largest [ ] . the present study is not free from limitations. in fact, the italian versions of the scale used were not validated for telephone interviews. however, due to the high risk of infection by direct contact, it was considered inevitable to conduct remote interviews. in addition, the anxiety scale has never been validated in italian, despite having been already used in italian studies [ , ] . this limitation did not invalidate the goal of the study, which was not intended to discriminate normal from pathological scores, but to describe changes in anxiety symptoms in a given population in which each subject was the "control" of himself. in argentina, a recent study produced very similar results. in fact, obligatory social isolation has forced the suspension of many care services for people living with dementia. negative psychological effects on people with dementia and increased stress levels in their caregivers have also been observed [ ] . many other papers investigated the stress of medical and nursing staff in several health contexts, but the research on the psychological outcome of caregivers of patients with dementia is at present very limited. this study showed the effects of lockdown to contain the covid- epidemic on caregivers of patients with dementia. isolation had negative effects in about half of people with dementia, also for the suspension of essential services for their care. prolonged time of isolation had negative effects on caregivers: as the isolation time increased, higher levels of depression and anxiety were recorded. educational level appears to be a protective factor. women seem to suffer the consequences of the lockdown on anxiety, depression and stress more than the men. funding none of the authors received any funding for this work. dementia care during covid- the italian dementia national plan alzheimer's disease: costs of care and needs of families in the campania region a rating instrument for anxiety disorders a self-rating depression scale convergent and discriminant validation of the italian version of the zung self-rating depression scale psychometric evaluation of three versions of the italian perceived stress scale a global measure of perceived stress italian version of the starkstein apathy scale (sas-i) and a shortened version (sas- ) to assess "pure apathy" symptoms: normative study on individuals mini-mental state examination: new normative values on subjects in southern italy the italian version of the quick mild cognitive impairment (qmci-i) screen: normative study on healthy subjects gender, mental health and ageing prevalence and incidence of depressive and anxious symptoms in couples undergoing assisted reproductive treatment in an italian infertility department fatigue in sjogren's syndrome: relationship with fibromyalgia, clinical and biologic features living with dementia: increased level of caregiver stress in times of covid- key: cord- -p phkwo authors: cellini, nicola; canale, natale; mioni, giovanna; costa, sebastiano title: changes in sleep pattern, sense of time and digital media use during covid‐ lockdown in italy date: - - journal: j sleep res doi: . /jsr. sha: doc_id: cord_uid: p phkwo italy is one of the major covid‐ hotspots. to reduce the spread of the infections and the pressure on italian healthcare systems, since march , , italy has been under a total lockdown, forcing people into home confinement. here we present data from , people living in the italian territory (m (age) = . ± . years, females, workers, university students), who completed an online survey from march to march , . in the survey, we asked participants to think about their use of digital media before going to bed, their sleep pattern and their subjective experience of time in the previous week (march – , which was the second week of the lockdown) and up to the first week of february (february – , before any restriction in any italian area). during the lockdown, people increased the usage of digital media near bedtime, but this change did not affect sleep habits. nevertheless, during home confinement, sleep timing markedly changed, with people going to bed and waking up later, and spending more time in bed, but, paradoxically, also reporting a lower sleep quality. the increase in sleep difficulties was stronger for people with a higher level of depression, anxiety and stress symptomatology, and associated with the feeling of elongation of time. considering that the lockdown is likely to continue for weeks, research data are urgently needed to support decision making, to build public awareness and to provide timely and supportive psychosocial interventions. and increase the risk of mental health problems (leigh-hunt et al., ) . sivertsen et al. ( ) have shown that young adults in a range between and years old tend to have a high and growing prevalence of sleep problems, and young adults make extensive use of social media devices that interact with daily activities, including sleep (gradisar et al., ; orzech, grandner, roane, & carskadon, ) . social media platforms offer an opportunity to ameliorate social isolation, but the use of digital media near bedtime could have a broadly negative effect on sleep outcomes of young adults (orzech et al., ) . we were interested in characterizing the change in digital media use before going to bed, in sleep quality and timing, and in the subjective experience of time passing, and their relationship with depression, anxiety and stress levels. to achieve this, we administered a cross-sectional survey to non-hospitalized italian residents. in this survey we asked participants to think about the previous week (march - , which was the second week of the lockdown) and the period prior to the first week of february (february - , before any restriction in any italian area). participants completed an anonymous online survey, after reading the written consent form and explicitly agreeing to participate in the survey. the survey was shared via social media for a limited time window (from march to , ), targeting people from to years old to obtain a homogenous sample of responders, because young adults (e.g., ages - years) are likely to be heavy digital users and they are especially at risk of sleep disturbance (levenson, shensa, sidani, colditz, & primack, ; sivertsen et al., ) . there was no monetary or credit compensation for participating in the study. the study protocol was approved by the ethics committee of the centre for research and psychological intervention (cerip) of the university of messina and was conducted in accordance with the declaration of helsinki. a total of , participants completed the survey. from this sample, we excluded participants who reported having been infected by covid- , participants with age > or < years and one outlier. the final sample consisted of , young adults living in the italian territory. data reported in this study were part of a wider research project designed with multiple purposes regarding the psychological impact of home confinement in italy, and other data with different purposes will be presented elsewhere. perceived sleep quality in the previous week (march - , the second week of the lockdown) and prior to the first week of february (february - , before any restriction in any italian area) was assessed using the italian version of the pittsburg sleep quality index (psqi; curcio et al., ) . scores higher than indicate poor sleep. from the psqi we also derived information about how many hours they spent in bed (tib; hr), at what time participants went to bed to sleep (bedtime) and at what time participants woke up in the morning (wake time). the use of digital media was assessed by asking participants to report their frequency of digital media activity choices in the hr before bedtime in the previous week and prior to the first week of february. we selected digital media activities based on the work of orzech et al. ( ) , such as computer work, email/instant messaging, social networking sites, videogames, surfing the internet, video calls/calls on phones, watching online videos, reading e-books, watching tv/dvds, and watching tv series. response options ranged from "never" to "every day" and were transformed into weighted responses to estimate total digital media use per week (range, - ), as previously done by orzech et al. ( ) . an ad-hoc questionnaire was created to assess the ability of participants to keep track of physical time (time track, three items referring to difficulties in remembering the day of the week, remembering the day of the month and keeping track of the time of day), to assess the subjective experience of time (time expansion, two items referring to the feeling of boredom and that time is passing slowly) and to assess the regularity of daily routines (schedule, two items referring to maintaining routines such as meal times and bedtimes). the short-form version of the depression anxiety stress scales (dass- ) was used to assess stress, depression and general anxiety (bottesi et al., ) . scores from each subscale were summed up and multiplied by two to suit the original items. the cut-offs for moderate symptomatology for depression, anxiety and stress are ≥ , ≥ and ≥ , respectively (lovibond & lovibond, ) . of the , participants (m age = . ± . years), were workers (m age = . ± . years, f) and were university students (m age = . ± . years, f). one-fourth ( . %) of the entire sample reported moderate to extremely severe symptoms of depression, . % of the sample reported moderate to extremely severe symptoms of anxiety, and . % of the sample reported moderate to extremely severe symptoms of stress (figure a) . under the restriction, our sample increased the frequency of digital media use in the hr before going to bed from ~ activities a week before the restriction to ~ activities under lockdown (f , , = . , p < . , η p = . ). looking at changes in sleep habits, we observed a significant increase of the psqi score under the restriction (f , , = . , p < . , η p = . ), with no difference between students and workers (all ps > . , figure a ). the proportion of poor sleepers (i.e., psqi > ) increased from . % to . % (χ = . , p < . ; log odds ratio = . [ci = . - . ]; figure b) . sleep timing changed dramatically, with all participants going to bed later under restrictions (f , , = . , p < . , η p = . ; figure b ). on average, bedtime was delayed by ~ min in both workers and students. the restrictions had an even stronger effect on wake time (f , , = . , p < . , η p = . ), in particular in workers (f , , = . , p < . , η p = . ; figure c ), who started to wake up about hr and min later than usual, whereas students delayed their wake time by ~ min. the later wake time resulted in a generally longer tib under restrictions (f , , = . , p < . , η p = . ), which was mainly driven by workers (f , , = . , p < . , η p = . ; figure d ), who spent ~ min more in bed than before the restrictions (change for students: ~ min). regarding the experience of time, the lockdown induced a signif- also, they experienced an expansion of time during the lockdown (f , , = . , p < . , η p = . ), with people increasingly feeling a sense of boredom, and they felt stuck in time about five times in the week compared to . times before lockdown. we observed that the change in sleep quality (psqi global score) was stronger in participants with a high dass- score (i.e., participants with higher symptoms of depression, anxiety and stress; r = . , p<. ), in particular in students ( figure a) . interestingly, the increase in subjective elongation of time was moderately associated with poorer sleep quality (r = . , p < . ), in particular in workers (see figure b ), but mildly associated with the increased use of digital media (r = . , p < . ). lastly, the increase in digital media use before going to bed mildly affected sleep habits; that is, sleep latency (r = . , p < . ), bedtime (r = . , p = . ) and wake time (r = . , p < . ), but not sleep quality (r = . , p = . ). in this cross-sectional study, we investigated the impact of the italian restriction measures on individuals' daily habits, such as sleep-wake rhythms, digital media use and the experience of time. as expected, during the lockdown people increased the usage of digital media near bedtime. however, this change in digital media use did not affect sleep habits. nevertheless, already in the second week of the lockdown (march - ), sleep-wake rhythms markedly changed, with people going to bed and waking up later, and spending more time in bed, but, paradoxically, also reporting a lower sleep quality. the decrease in sleep quality was stronger for people with a higher level of depression, anxiety and stress symptomatology and also associated with an increased feeling of expansion of time. although this is a partial snapshot of the current italian situation and does not consider adults older than years and children and adolescents, our data seem to confirm the evidence already observed in quarantined individuals (brooks et al., ) , showing that people in lockdown are experiencing negative psychosocial changes, which have an impact on well-being (e.g., sleep disturbances, depression and anxiety). as stated by a task force of the european cbt-i academy (altena et al., ), during total lockdown individuals' sleep habits are challenged by several factors (i.e., reduced exposure to sunlight, reduced physical activity and psychological distress. we also think that the lack of social zeitgebers, such as regular work schedules and social activities, as well as changes in living conditions (e.g., moving to parents' house), are strongly affecting sleep habits under restrictions. of note, our data were collected before the f i g u r e (a) frequency of participants reporting moderate symptoms of depression, anxiety and stress as derived by the depression anxiety stress scales (dass)- scores. (b) frequency of participants with a pittsburg sleep quality index (psqi) score ≤ and > before and during the lockdown f i g u r e changes in (a) pittsburg sleep quality index (psqi) total score, (b) time in bed, (c) bedtime and (d) wake time as a function of the presence of the lockdown and the status (worker or student) of the participants. error bars represent confidence intervals f i g u r e change in pittsburg sleep quality index (psqi) global score (before lockdown scores minus during lockdown scores) as a function of (a) depression anxiety stress scales (dass)- total score and (b) time expansion change in students and workers due to the covid- outbreak: practical recommendations from a task force of the european cbt-i academy the italian version of the depression anxiety stress scales- : factor structure and psychometric properties on community and clinical samples the psychological impact of quarantine and how to reduce it: rapid review of the evidence validity of the italian version of the pittsburgh sleep quality index (psqi) the sleep and technology use of americans: findings from the national sleep foundation's sleep in america poll comunicato stampa n° / . studio iss. un paziente su cinque positivo al coronavirus ha tra e anni covid- in italy: momentous decisions and many uncertainties an overview of systematic reviews on the public health consequences of social isolation and loneliness social media use before bed and sleep disturbance among young adults in the united states: a nationally representative study manual for the depression anxiety stress scales digital media use in the h before bedtime is associated with sleep variables in university students sleep patterns and insomnia in young adults: a national survey of norwegian university students immediate psychological responses and associated factors during the initial stage of the coronavirus disease (covid- ) epidemic among the general population in china mental health and psychosocial considerations during the covid- outbreak retrieved from changes in sleep pattern, sense of time and digital media use during covid- lockdown in italy