key: cord-1029071-858c9uag
authors: Raje, D. V.; Bajaj, A.; Chakraborty, M.; Purohit, H. J.
title: Cumulative Active and Recovery Rates Based Criterion for Gradual Lockdown Exit: A Global Observation of SARS Cov-2 Management
date: 2020-06-07
journal: nan
DOI: 10.1101/2020.06.05.20123364
sha: 3391184e14d64c074cc6f5f7552142bcbd230178
doc_id: 1029071
cord_uid: 858c9uag

On March 11, 2020 the World Health Organization (WHO) had declared SARS-CoV-2 pandemic; and at present there are over 6 million cases across the globe. Based on the interim guidelines of WHO, most of the countries opted for social distancing with lockdown as the only way to control the pandemic. This led to Manufacturing shut down, which acted as a spanner in the wheel for international supply chain leading to pressure on governments to review the protocols of the lockdown. We studied epidemiological parameters for 18 countries and obtained crossover time point referring to cumulative case active and case recovery rates and the time point for the peak positive confirmation rate in a time window of 92 days; and linked with the respective governmental decisions. For countries awaiting crossover, time series non-linear models could be used for predicting the crossover point. A sample study was carried out for India. The median time for reaching crossover for 12 countries was 37 days, while peak positive confirmation rate was 30 days after their first intervention. These countries enforced strict lockdown regulations and have shown constant improvement in their recovery rate even after crossover time point. A phase wise relaxation of lockdown is evident after crossover point in most of these countries. The crossover time point with the subsequent increasing recovery rate can be a strategy for lockdown relaxation as evident from the experiences of few countries. Also, we propose a criterion based on cumulative recovery and fatality rate for micro-management of lockdown.

World economy is a network of the global value chain, where every single country is a 32 partner. By March 2020, the whole civilization realized the pandemic impact of SARS CoV2; 33 and world over, the restrictions in movement and lockdown have severely impacted the 34 industrial supply chain. 1,2 The cumulative effect of this interrupted productivity led to loss of 35 millions of jobs and this has now created uneasiness at the different levels of society; and 36 hence, governments are looking for solutions to come out of this lockdown situation. [3] [4] [5] 37 Different modelling approaches have been reported to predict the course of the viral spread, 38 of which susceptible-exposed-infected-recovered (SEIR) model, is the most explored 39 approach applied by various groups with data collected from different countries. 6, 7 However, Another issue is the diversity as reflected by genome sequence data that suggests, virus is 44 either mutating or adapting according to host in different part of the world. 11 . 45 In this paper, we have opted an observational approach using the data for monitoring test with 46 active, recovered and fatal cases and whenever the model showed some perturbations, we In this study, an observational approach was adopted in which epidemiological parameters 51 were observed with time and retrospectively analysed with reference to interventions or 52 public announcements by the governments. 53 Eighteen countries mostly from Europe and Asia were considered with at least 1000 54 diagnosed cases as on March 31 st 2020. All these countries were tracked from March 1 st for 55 the number of new cases, active cases, recoveries and number of fatalities. The data source 56 was John Hopkins CSSEDISandData, available as an open source. This data was used to 57 derive the case active rate (CAR), recovery rate (CRR) and the fatality rate (CFR) at each 58 discrete time point till May 31, 2020. These rates were obtained as the cumulative number of 59 active/recovered/fatalities on a specific day to the total number of cases diagnosed till that 60 day, considering a lag of zero days. Thus, on any day, the total infected individuals were 61 categorized into these three states, summing to unity. Moreover, the ratio of number of newly 62 . 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. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 7, 2020. 

Among the selected countries, seven were from Asia, while eleven were from European 88 continent including Russia, which is transcontinental. The CAR, CRR and CFR for each day 89 along with the PCR were obtained for each country. Simultaneously, the major events and 

. 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. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 7, 2020. . In Turkey, the spreading peak of virus was noted on April 12 th , and subsequently, a gradual 123 slowdown was observed till May 15 th . The CAR started declining after April 23 rd and the 124 crossover reached on May 3 rd with CRR (0.5010) exceeding CAR (0.4720), and CFR around 125 . 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 this version posted June 7, 2020. In Singapore, the CAR increased till April 22 nd , and then was constant till May 9 th . Later it Cross-over forecasting -A study for India 167 In India, since March, the incidence of COVID19 is on its rise. Figure 3 provides the CAR, 168 CRR and CFR, which shows that the crossover point is very close but not reached in the 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 June 7, 2020. . https://doi.org/10.1101/2020.06.05.20123364 doi: medRxiv preprint Since, it was observed that once a country crosses 50% cumulative recovery cases, the trend 190 is always followed. Hence, we first predicted the possible time point for 50% cumulative 191 recovery of cases for India; as described earlier. The next step was how to derive a criterion, 192 which can propose the strategy to gradually exit the lockdown with national data supporting 193 0.5 CRR. We have applied following disease management strategy, we propose two 194 parameters viz., CFR and CRR that can be jointly assessed for each state and eventually each 195 city/district. The thresholds for these parameters can be referred based on overall national 196 data for CRR and CFR; and accordingly, the classification can be obtained at each 197 hierarchical level. Figure 5A In this study, we adopted a strategy of learning from the experiences of select countries based 219 on epidemiological parameters. The primary focus was on crossover time and the time of 220 peak PCR. The government interventions were linked with these two parameters. The 221 . 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 this version posted June 7, 2020. There is significant variation in sequences after mid-February. The viral sequences can be 252 grouped in two different clades which are associated with European sequence data. Whereas,

. 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 this version posted June 7, 2020. . https://doi.org/10.1101/2020.06.05.20123364 doi: medRxiv preprint Asian sequences are close to two major European clades, with an additional clade having 254 different sequences representing different countries.

There is a need to understand the whether the strains/sequence data observed in the twelve 256 countries have any association with the peak PCR and the crossover times observed in these 257 countries. This may reveal the relevance of different strains and its interaction with host 258 machinery, which is supporting the viral multiplication and its dispersion factor. 24 The viral 259 genome sequence data analysis from India shows two major clades. 25,26 However, the 260 sequence data is not having enough representative sequences from the country, and hence 261 cannot be correlated with number of active cases observed in different states for the moment.

For some countries, the crossover point is before 50 days, while for others it is beyond. It will 263 be interesting to map the sequence data from different countries before and after the cross the description of various events and interventions by the respective governments before and after 364 crossover time. It seems that in eight of these twelve countries, the decision for relaxation of 365 lockdown has been taken after reaching the crossover time and deceleration of PCR. A prior peak for 366 PCR followed by a crossover could be a controlled scenario for any country to manage COVID19 367

pandemic. Subsequent to crossover, the CAR and CRR were found progressing in the same acquired 368 directions in logarithmic manner. Based on these trends, the strategies by these governments appears 369

to be rational, and can be adopted by countries in which the crossover is still awaited. 370 showing cities with more than 100 cases as on May 31 st C) A tentative state level management 393 guideline to improve upon the two parameters. The dashed lines indicate the national level CFR and 394

CRR as reference to identify the hot-spots. 395 396 . 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 this version posted June 7, 2020. . description of various events and interventions by the respective governments before and after 416 crossover time. It seems that in eight of these twelve countries, the decision for relaxation of 417 lockdown has been taken after reaching the crossover time and deceleration of PCR. A prior peak for 418 PCR followed by a crossover could be a controlled scenario for any country to manage COVID19 419

pandemic. Subsequent to crossover, the CAR and CRR were found progressing in the same acquired 420 directions in logarithmic manner. Based on these trends, the strategies by these governments appears 421

to be rational, and can be adopted by countries in which the crossover is still awaited. 422 showing cities with more than 100 cases as on May 31 st , C) A tentative state level management 445 guideline to improve upon the two parameters. The dashed lines indicate the national level CFR and 446

CRR as reference to identify the hot-spots. 447

A-C 448 . 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. (which was not certified by peer review)

The copyright holder for this preprint this version posted June 7, 2020. . https://doi.org/10.1101/2020.06.05.20123364 doi: medRxiv preprint

Variation in government responses to 298 COVID-19

How will country-300 based mitigation measures influence the course of the COVID-19 epidemic? The Lancet

If the world fails to protect the economy, COVID-19 will damage 303 health not just now but also in the future

Transition steps to stop COVID-19 305 without killing the world economy

Macroeconomic Implications of 307 COVID-19: Can Negative Supply Shocks Cause Demand Shortages? SSRN Electron

Estimating required 'lockdown' cycles before immunity to SARS-CoV

Model-based analyses of susceptible population sizes, 'S0', in seven European countries 311 including the UK and Ireland

Using ILI surveillance to estimate state-specific case 313 detection rates and forecast SARS-CoV-2 spread in the United States

Why is it difficult to accurately predict 316 the COVID-19 epidemic?

Predictive Mathematical Models of the COVID-318 19 Pandemic: Underlying Principles and Value of Projections

Modelling the COVID-19 epidemic and implementation of population-wide 321 interventions in Italy

Emergence of genomic diversity and recurrent mutations in SARS-CoV-2

How and when to 326 end the COVID-19 lockdown : an optimisation approach

Stay Nearby or Get Checked': A 328 Covid-19 Lockdown Exit Strategy

Inferring change points in the spread of COVID-19 reveals the effectiveness 330 of interventions

World Health Organization. Critical preparedness , readiness and response actions for COVID-332 19 . Interim guidance 22

Mental health care for medical staff in China during the COVID-19 outbreak

Supporting the Health Care Workforce during the COVID-19

The Propagation of the Economic Impact through Supply Chains: The 338 Case of a Mega-City Lockdown against the Spread of COVID-19

Economic Effects of Coronavirus Outbreak (COVID-19) on the World 341 Economy

Responding to Covid-19 -A Once-in-a-Century Pandemic?

Preparedness and vulnerability of African countries against importations of 345 COVID-19: a modelling study

Managing COVID-19 in Low-and Middle-Income 347

Comparative Genomic Analysis of Rapidly Evolving SARS CoV-2 Viruses 349

Reveal Mosaic Pattern of Phylogeographical Distribution

Genomics of Indian SARS-CoV-2: 354 Implications in genetic diversity, possible origin and spread of virus

Understanding SARS-CoV-2: Genetic Diversity, Transmission and 357 Cure in Human

Mar 24: Nationwide lockdown for 21 days. 2. Mar 12: Cluster related to public gathering identified

Nationwide lockdown extended till 17 May

Mar 31: Cluster linked to religious gathering identified

Nationwide lockdown extended till 3 May, Conditional relaxation from

Mar 4: Compulsory screening of all international passengers on arrival

May 17: Nationwide lockdown extended till 31 May

May 30: Lockdown extended in containment zones till