key: cord-1017772-jab7vp33 authors: Cheatley, Jane; Vuik, Sabine; Devaux, Marion; Scarpetta, Stefano; Pearson, Mark; Colombo, Francesca; Cecchini, Michele title: The effectiveness of non-pharmaceutical interventions in containing epidemics: a rapid review of the literature and quantitative assessment date: 2020-04-10 journal: nan DOI: 10.1101/2020.04.06.20054197 sha: 0fa14715bed171a363a291bb3fea00a3399937f5 doc_id: 1017772 cord_uid: jab7vp33 The number of confirmed COVID-19 cases has rapidly increased since discovery of the disease in December 2019. In the absence of medical countermeasures to stop the spread of the disease (i.e. vaccines), countries have responded by implementing a suite of non-pharmaceutical interventions (NPIS) to contain and mitigate COVID-19. Individual NPIs range in intensity (e.g. from lockdown to public health campaigns on personal hygiene), as does their impact on reducing disease transmission. This study uses a rapid review approach and investigates evidence from previous epidemic outbreaks to provide a quantitative assessment of the effectiveness of key NPIs used by countries to combat the COVID-19 pandemic. Results from the study are designed to help countries enhance their policy response as well as inform transition strategies by identifying which policies should be relaxed and which should not. school closures, travel restrictions, contact tracing and quarantine, public information campaigns, and environmental and personal hygiene measures. Findings from the study are designed to assist policy-makers adjust, if necessary, their strategy to fighting the outbreak, by outlining the effectiveness of NPIs, in isolation or as part of a comprehensive policy package, on key outcome measures. Once the number of new cases falls and the impact of the disease lessens, the findings can also be used to inform a country's transition strategy, for example, by identifying which interventions to relax and which to not. A rapid review of the literature was undertaken to identify salient NPIs used to combat influenza outbreaks and their associated effectiveness. Academic, peer-reviewed articles were sourced from major databases including those focused on public health e.g. MEDLINE, and Google Scholar. The research was carried out on 19 March 2020. The search for articles concentrated on a combination of key terms including 'influenza', 'pandemic', 'non-pharmaceutical measures', 'non-pharmaceutical interventions', 'social distancing', 'school closures', 'workplace closures', 'mass gatherings', 'hygiene', 'contact tracing', 'travel restrictions', 'quarantine' and 'public health campaigns'. Additional articles were sourced using a snowball approach. To the extent possible, systematic reviews and meta-analyses pertaining to the effectiveness of NPIs were referenced in order to provide an up-to-date summary of the literature. To compare the relative effectiveness of policies, two common outcome measures frequently cited in the literature were identified (i.e. change in the overall attack rate and delay in the peak of disease). Data on these two dimensions was collected from the reviewed papers and compared using standard graphical approaches. Data on other dimensions was also collected and reported in the paper, however, a systematic comparison was not possible due to heterogeneity in the measures used. Finally, policies being utilised by countries to combat COVID-19 were obtained from the grey literature, in particular official government websites and documents. Social distancing refers to policies that deliberately increase physical space between people (John Hopkins Medicine, 2020 [1] ). Such policies come in many forms including banning large gatherings; encouraging people to work from home; and closure of nonessential stores such as restaurants and cafes. These policies may be implemented across a community or target specific at-risk groups such as the elderly and those with pre-existing health conditions (Anderson et al., 2020[2] ). The primary objective of social distancing is to prevent transmission thereby flattening the peak of the disease (Anderson et al., 2020[2] ). In the short-term this will ease pressure on the health care system, while in the long-term it provides time for new treatment and vaccines to be developed (Anderson et al., 2020[2] ). Several studies analysing the impact of social distancing on disease outbreaks exist (Chen et al., 2020[3] ). For example, a systematic review of workplace social distancing found the policy reduces the influenza attack rate (i.e. the proportion of individuals in a population who contract the disease) by 23% in the general population (Ahmed, Zviedrite and Uzicanin, 2018 [4] ). This is supported by an earlier study which found working-from-home was 'moderately effective' in reducing influenza transmission by 20-30% (Rashid et al., 2015[5] ). Further, social distancing in Sydney, Australia, during the 1918-19 influenza pandemic is estimated to have reduced the attack rate by nearly 40% (i.e. from 60% to 37%) indicating 22% of the population were spared infection (Caley, Philp and McCracken, 2008[6] ). One approach to social distancing is the banning of mass events (e.g. music festivals or large spectator sporting events). While often seen as a logical element of containment strategies, the evidence suggest that this intervention is most effective when implemented together with other social distancing measures (Ishola and Phin, 2011[7] ; Markel et al., 2007[8] ). This is because contact-time at such events is relatively small compared to the time spent in schools, workplaces, or other community locations such restaurants (Ferguson et al., 2020[9] ). As with other containment strategies, the earlier bans on mass gatherings are enforced, the greater their impact (Hatchett, Mecher and Lipsitch, 2007[10] ). Several challenges are associated with social distancing. Salient examples include reduced economic activity caused by closures and reduced social interaction (Rashid et al., 2015[5] ; OECD, 2020 [11] ); neglect of vulnerable populations, such as the elderly (Boddy, Young and O'Leary, 2020 [12] ); and psychological damage such as acute distress disorder, anxiety and insomnia (Brooks et al., 2020[13] ). Table 1 lists certain policies implemented by OECD countries to enhance social distancing in response to the COVID-19 pandemic. Examples range in their intensity and thus their impact on day-to-day life for affected populations. School closures aim to decrease the number of contacts by school children, and thus reduce transmission of the disease throughout the community. There are two types of school closures: proactive school closures before any infection is associated with a school; and reactive school closures in response to a student, parent or staff member falling ill. School closures, either proactive or reactive, reduce influenza transmission but with a wide range of effectiveness varying between pandemics, cultures and setting (e.g. rural or urban), and depending on the level of transmission in schools and To prevent or delay the entry of a disease into a country, it is common for policymakers to implement several travel restrictions. Screening, for example, is common in airports during a disease outbreak, and may include thermal scans to identify passengers with a high external body temperature as well as health questionnaires to detect symptoms and possible exposure (by providing information on travel history) (Selvey, Antão and Hall, 2015 [20] A major impact of travel restrictions is the flow-on effect this has on trade and business. As a result, travel restrictions dampen economic activity by reducing demand in tourism-dependent industries such as hotels, restaurants and aviation (Rashid et al., 2015[5] ). In addition, screening may further stretch limited resources by isolating or quarantining travellers with symptoms unrelated to the disease of interest (Priest et al., 2015[27] ). Contact tracing aims to identify, list and closely watch people who have been in contact with an infected person, even if they do not display symptoms. This helps the traced persons to get care and treatment early, and prevents further transmission of the virus (World Health Organization, 2017 [28] ). Quarantine (i.e. isolation) can be spent either at home, in hospital or specifically equipped structures. There is limited data on the effectiveness of contact tracing and quarantine. Evidence that is available is largely based on modelling studies, which suggest quarantine decreases the peak case load, the attack rate and would also delay the peak. Household quarantine is potentially the most effective measure to reduce attack rates in the community, but only if compliance is high (Rashid et al., 2015[5] ). Voluntary or self-isolation of infected people is moderately effective. For instance, voluntary quarantine of households with an infected individual may delay the peak of influenza by two to 26 days, and reduce the peak daily attack rate from 1.9% to 1.5%, or even down to 0.1%, depending on the associated interventions (e.g. such as treating infected people and applying prophylaxis to their households the day after the symptoms start, school closure) (Ferguson et al., 2006[14] ). Modelling work simulating the isolation of cases and contacts in the case of the COVID-19 pandemic concluded that about 70% of cases had to be traced to successfully contain the outbreak, assuming a reproduction number of 2.5 (Hellewell et al., 2020[29] ). Implementation of contact tracing and quarantine measures bears economic costs, and is associated with psychological, legal, and ethical issues. First, contact tracing requires substantial resources to sustain after the early phase of the epidemic since the number of infected people and contacts grow exponentially. While there is no obvious rationale for a routine use of contact tracing in the general population, it may be adapted in some circumstances (e.g. if there was an infected person on an aircraft) (Fong et al., 2020 [31] ). In addition, isolation is likely to cause distress and mental health problems, requiring additional services such as creating support lines and advice, helping people create plans, encouraging messages and calls and maintaining some routine (Lunn et al., 2020 [32] ). During an epidemic, policymakers can use campaigns to communicate with the public. These campaigns need to inform the public about the development of the epidemic and the risk it poses, with the aim of encouraging them to take the appropriate protective measures, such as hand washing or social distancing. In addition to saving lives, clear and timely information can also help preserve a country's social, economic and political stability in the face of emergencies (World Health Organization, 2018 [33] ). Especially in the case of COVID-19, which is a rapidly evolving situation with little knowledge about the disease, effective communication is crucial. Without it, the many unknowns can give space for rumours to develop and panic to set in (World Health Organization, 2018 [33] ). On the other hand, it is important to strike a balance between preventing panic and encouraging action. A study of the public perception of the influenza A/H1N1 (swine flu) outbreak in the United Kingdom found that few people changed their behaviour, and linked this to a believe that the outbreak had been exaggerated (Rubin et al., 2009[34] ). The authors suggest that convincing the public that the threat is real may be a more pressing task for public health agencies than providing reassurance. Evidence on the effectiveness of public information campaigns is limited. As mentioned above, the perceived risk can affect whether people change their hygiene behaviour, which means that the effectiveness of campaigns differs from one disease or epidemic to the next. Other studies have shown that the impact of campaigns can be increased by using trusted spokespeople like public health officials and through a role model effect from officials (Quinn et al., 2013[35] ). The 2003 SARS outbreak hit Singapore in late February, prompting the government into action. An important element of Singapore's containment strategyin addition to the closure of schools and case isolationwas an integrated mass media campaign. This campaign included advertisements in the four languages (English, Chinese, Malay and Tamil) in newspapers and on television, a dedicated website, a toll-free hotline, booklets, posters and stickers. At the local level, town councils and other organizations organised discussion sessions and demonstrations for residents. A survey looking at whether the public followed the recommended behaviour found that adherence to the compulsory temperature checks was high: 85% of the respondents said that they monitored their temperature daily (Karan et al., 2007[36] ). The evidence to date suggests COVID-19 is primarily spread from person-to-person via small respiratory droplets (GAVI Alliance, 2020 [37] ). The disease may also be transmitted through fomites, that is, objects that can carry infections (e.g. furniture) (Center for Disease Control and Prevention, 2020 [38] ). In order to reduce the risk of transmission through fomites, policy-makers may require public and private spaces where infected individuals are likely, or known, to have frequented be cleaned and/or disinfected (e.g. schools, offices, day care centres) (Otter and Galletly, 2018 [39] ). This process is referred to as environmental decontamination (ED) and results in contamination levels that do not harm the health of individuals (Otter and Galletly, 2018 [39] ). Given that the virus causing COVID-19 can remain on surfaces for an extended period of time, and that cleaning, as well as disinfectant, can reduce contamination levels, theoretically, ED can reduce transmission rates. Using these assumptions, a modelling study to estimate the impact of regular cleaning of high-touch surfaces in an office found the measure reduces the infection risk of influenza by 2.14% (Zhang and Li, 2018 [40] ). The academic literature on environmental hygiene is limited with a recent systematic review identifying three studies (Xiao et al., 2020[41] ). These studies all focus on younger children (i.e. those of school age) and do not accurately reflect ED efforts implemented during a pandemic. For example, one study measured the impact of disinfecting toys every two weeks (Ibfelt et al., 2015[42] ), while another estimated the impact of bleach use in the home on respiratory illness (Casas et al., 2015[43] ). For this reason, these studies should not necessarily be taken as evidence on the impact ED has on coronaviruses given SARS-CoV-2 can stay on a surface for up to three days. The spread of COVID-19 has led countries to implement various ED policies. This is evidenced by the number of governments with online resources dedicated to best practice environmental and disinfection practices (e.g. (Australian Government Department of Health, 2020 [44] ; Center for Disease Control and Prevention, 2020 [38] )). In addition to environmental hygiene, individuals are encouraged to enhance their personal hygiene through hand washing, sneezing or coughing practices, and the use of protective facemasks. Various studies have shown these interventions can protect individuals by reducing their risk of getting infected (Figure 1 ) (Jefferson et al., 2011[45] ). A study of the SARS outbreak in Hong Kong, China found that people who got infected were less likely to frequently have worn a face mask in public (odds ratio 0.36) or to have washed their hands 11 or more times per day (Lau et al., 2004[46] ). A randomised trial during the influenza A(H1N1) pandemic showed a 35% to 51% reduction in the incidence of influenza-like illness when using both masks and exercising proper hand hygiene practices and cough etiquette (Aiello et al., 2010[47] ). Similarly, a meta-analysis found that combining masks and hand hygiene reduced the risk of influenza infection by 27% (Wong, Cowling and Aiello, 2020[48] ). The impact of pandemic policy measures on outcome variables of interest are summarised in the figures below. Specifically, the impact on: 1) the attack rate, which represents the proportion of the population who are infected; and 2) the number of days the policy measure delays the peak of a disease. Results from the analysis indicate that social distancing is the most effective measure for both reducing the attack rate as well as delaying the disease peak. For example, work place social distancing measures (such as working from home and workplace closures) can reduce the disease attack rate by between 23-73% (Ahmed, Zviedrite and Uzicanin, 2018 [4] ; Rashid et al., 2015[5] ). Less effective measures include travel restrictions, in particular international restrictions, which is estimated to reduce the attack rate by just 0.02% (Mateus et al., 2014[21] ). . CC-BY-NC-ND 4.0 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 . Modelling studies consistently conclude that policy packages, as opposed to individual policies, are the most effective approach to reduce the impact of an epidemic (Lee, Lye and Wilder-Smith, 2009 [49] ). For example, Ferguson et al. (2006[14] ) estimated that home quarantine reduces the overall attack rate by approximately 10% (i.e. from 27% to 24%), with this figure increasing to approximately 70% when adding school and workplace closures, effective border controls, and antiviral treatment and prophylaxis. A study from the United States found the reduction in the overall attack rate was 30 . CC-BY-NC-ND 4.0 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 . prophylaxis, quarantine, isolation and school closures (90% reduction in the attack rate) when compared to community and workplace social distancing only (60% reduction in the attack rate) (Halloran et al., 2008[50] ). Finally, a modelling study from Milne et al. (2008[18] ) estimated that a combination of school closures, case isolation, workplace non-attendance and limited community contact can reduce the final attack rate by 96% (i.e. from 55% to 2%) compared to between 12-45% (i.e. from 55% to 30-48%) when policies are implemented independently. This study outlines key NPIs used by various countries to combat the COVID-19 pandemic, including an assessment of their effectiveness. The objective of this analysis is two-fold: 1) to assist countries adapt their strategy to fight COVID-19, if necessary; and b) to inform transition strategies by identifying which NPIs to relax once the number of new cases falls and the impact of the disease lessens. Based on an analysis of modelling studies, policy-makers are encouraged to implement policy packages to combat disease outbreaks such as COVID-19. Depending on the methodology and the policy package evaluated, these studies found comprehensive packages can reduce disease attack rates by at least 40%. At the individual policy level, NPIs that reduce social interaction are most effective, in particular measures encouraging or requiring people to work at home, school closures and quarantine periods for those infected or potentially infected with the disease. These interventions, however, are associated with significant consequences including severely dampened economic activity and an increase in mental health issues. Less evidence is available to support interventions such as internal and international travel restrictions, bans on mass gatherings, environmental hygiene and public health campaigns. Findings from this study broadly align with outcomes from Imperial College London's modelling study, which found more stringent policies, such as school closures, have a greater impact on reducing disease transmission (Flaxman et al., 2020[61] ). Several limitations are associated with this study. First, this study relied on findings from a rapid review as opposed to a comprehensive systematic review of the literature. Second, only a selection of NPIs were analysed based on current discussions regarding potential policy actions to combat COVID-19. Third, findings from the analysis may not be directly applicable to combatting COVID-19 given the effectiveness of NPIs differ depending on the context in which they are implemented (e.g. population compliance) as well as the characteristics of the disease (e.g. reproduction number, incubation period, infection fatality rate). Finally, evidence to assess the impact of policy packages was reliant on modelling studies, which were often specific to a geographic region (primarily the United States and the United Kingdom). Effectiveness of workplace social distancing measures in reducing influenza transmission: A systematic review Mask Use, Hand Hygiene, and Seasonal Influenza-Like Illness among Young Adults: A Randomized Intervention Trial How will country-based mitigation measures influence the course of the COVID-19 epidemic? COVID-19 (Coronavirus) and the Australian border Environmental cleaning and disinfection principles for COVID-19 COVID-19 Incubation Period: An Update Estimates of the reproduction number for seasonal, pandemic, and zoonotic influenza: A systematic review of the literature School closure during novel influenza: A systematic review Cabin fever': Australia must prepare for the social and psychological impacts of a coronavirus lockdown The psychological impact of quarantine and how to reduce it: rapid review of the evidence Empirical Evidence for the Effect of Airline Travel on Inter-Regional Influenza Spread in the United States Quantifying social distancing arising from pandemic influenza Domestic use of bleach and infections in children: a multicentre cross-sectional study Closure of schools during an influenza pandemic Environmental Cleaning and Disinfection Recommendations Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study COVID-19 control in China during mass population movements at New Year The Effect of School Dismissal on Rates of Influenza-Like Illness in New York City Schools During the Spring 2009 Novel H1N1 Outbreak Coronavirus response: Transport measures Coronavirus and entry restrictions: 4 things travelers to Germany need to know Strategies for mitigating an influenza pandemic Impact of non-pharmaceutical interventions (NPIs) to reduce COVID-19 mortality and healthcare demand Estimating the number of infections and the impact of nonpharmaceutical interventions on COVID-19 in 11 European countries Nonpharmaceutical Measures for Pandemic Influenza in Nonhealthcare Settings-Social Distancing Measures Travel Advice against all non-essential travel: Foreign Secretary's statement Factors that make an infectious disease outbreak controllable What is COVID-19 and how is it spread? Modeling targeted layered containment of an influenza pandemic in the United States Public health interventions and epidemic intensity during the 1918 influenza pandemic Feasibility of controlling COVID-19 outbreaks by isolation of cases and contacts Effect of cleaning and disinfection of toys on infectious diseases and micro-organisms in daycare nurseries Could influenza transmission be reduced by restricting mass gatherings? Towards an evidence-based policy framework The effects of school closures on influenza outbreaks and pandemics: Systematic review of simulation studies Physical interventions to interrupt or reduce the spread of respiratory viruses Coronavirus, Social Distancing and Self Quarantine Emerging Victorious Against an Outbreak: Integrated Communication Management of SARS in Singapore Media Coverage and Impact of the SARS Campaign in Moving a Nation to be Socially Responsible Early dynamics of transmission and control of COVID-19: a mathematical modelling study The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly Reported Confirmed Cases: Estimation and Application SARS Transmission, Risk Factors, and Prevention in Hong Kong Interrupting transmission of COVID-19: lessons from containment efforts in Singapore Combination strategies for pandemic influenza response -a systematic review of mathematical modeling studies Using behavioural science to help fight the coronavirus Nonpharmaceutical interventions implemented by US cities during the 1918-1919 influenza pandemic Effectiveness of travel restrictions in the rapid containment of human influenza: A systematic review A Small Community Model for the Transmission of Infectious Diseases: Comparison of School Closure as an Intervention in Individual-Based Models of an Influenza Pandemic Evaluation of the Effectiveness of Surveillance and Containment Measures for the First 100 Patients with COVID-19 in Singapore OECD Economic Outlook Supporting people and companies to deal with the Covid-19 virus: Options for an immediate employment and social-policy response Environmental decontamination 1: what is it and why is it important? Assessing the impact of travel restrictions on international spread of the 2014 west African Ebola epidemic Statement From The Premier Effectiveness of Border Screening for Detecting Influenza in Arriving Airline Travelers Border restrictions Exploring Communication, Trust in Government, and Vaccination Intention Later in the 2009 H1N1 Pandemic: Results of a National Survey Evidence compendium and advice on social distancing and other related measures for response to an influenza pandemic Public perceptions, anxiety, and behaviour change in relation to the swine flu outbreak: cross sectional telephone survey Entry screening for infectious diseases in humans Current Outbreak of Coronavirus Disease Government announces further measures on social distancing School closures caused by Coronavirus (Covid-19 Aerosol and Surface Stability of SARS-CoV-2 as Compared with SARS-CoV-1 Evolving Epidemiology and Impact of Non-pharmaceutical Interventions on the Outbreak of Coronavirus Disease Hand hygiene and risk of influenza virus infections in the community: a systematic review and meta-analysis Coronavirus Coronavirus disease 2019 (COVID-19) Situation Report -46 Influenza (Seasonal Managing epidemics: Key facts about major deadly diseases WHO | Contact tracing Nonpharmaceutical Measures for Pandemic Influenza in Nonhealthcare Settings-Personal Protective and Environmental Measures SARS-Cov-2) outside of Wuhan, China: Retrospective case series South Korea is reporting intimate details of COVID-19 cases: has it helped? Transmission of influenza a in a student office based on realistic person-to-person contact and surface touch behaviour The views expressed in this paper are those of the authors and do not necessarily reflect those of the Organisation for Economic Co-operation and Development (OECD) or its member countries. The OECD program of work on public health is supported by a number of voluntary contributions by Ministries of Health or other national governmental institutions of OECD member countries or other key partners. No financial disclosures were reported by the authors of this paper.