Mainstream Weekly

Home > 2020 > Does Data on COVID-19 Permit Scientific Studies in India? | Indrakant (...)

Mainstream, VOL LVIII No 28, New Delhi, June 27, 2020

Does Data on COVID-19 Permit Scientific Studies in India? | Indrakant Sulibhavi, Mahendran Arumugam

Friday 26 June 2020

Brief Research Paper on COVID-19

by Indrakant Sulibhavi2, Mahendran Arumugam1

Abstract

The paper seeks to examine the suitability of the available data on COVID-19 for studying the spatial variation in transmission of the disease. It looks into the format of the available data in drawing meaningful and useful inferences for policy formulation from the estimates of the parameters. It is observed that in Inter-country analysis some of the statistical and conceptual aspects do not receive attention which they deserve. Similarly, in inter-state analysis, suitable deflators need to be used. The format of the available data on COVID-19 does not permit us to make meaningful and reliable estimates of the fatality rate and the recovery rate. Even the method adopted by the Government to estimate these parameters seems to be defective. The data has to be grouped on the basis of the date of admission of persons in the hospital. The required parameters have to be worked out separately for each group. The average of these estimated parameters pertaining to sufficient number of groups would be meaningful, reliable and useful for policy formulation.

Keywords: Fatality rate, recovery rate, scientific method, data on COVID-19.

Introduction

Very often inadequate data prevent an in-depth study of a problem encountered by a society. Sometimes huge data (but not in required format) hinders scientific study. Presently every day the data on COVID-19, at global and national levels is being released through different channels by the concerned officials. This has encouraged (and to some extent prompted) Researchers, Academicians, Journalists, People with Social Concerns and many others to make use of the available data and draw their own inferences. The data on COVID-19 has been used for various purposes which include the spatial spread of COVID-19, the pace of transmission of the disease, the vulnerability of people belonging to different age groups and different ethnic groups, effectiveness of measures taken by the Government to control the transmission of COVID-19, response of general public to the Government Measures etc. Further, it is not uncommon to notice that most of the space in daily newspapers is occupied by matter relating to COVID-19. In this context, some important questions need to be answered. Does the available data on COVID-19 permit us to conduct scientific study? Whether the estimates generated from the available data reliable? Whether the inferences drawn from such estimates reveal ground reality? Are the findings useful for policy formulation? These are some of the questions which the present Paper seeks to examine.

 inter-Country Comparison:

Many works have used the data on COVID-19 to study the extent of transmission of Corona virus across the countries and the reasons for the same. Some people are puzzled to note that in America the number of persons affected by COVID-19 is in lakhs and the number of deaths due to COVID-19 is in thousands. They even start questioning the claim of Americans that they have one of the best health infrastructures. There are few persons who appreciate the ability of China in controlling the impact of Corona virus with much smaller loss of lives. However, there are some persons who point finger at China for transmitting the disease to the continents of Europe and America. Some people even go to the extent of accusing China of using Novel Corona virus as a weapon in biological war to capture world market. It appears that most of the persons engaged in analysing the data on COVID-19 are using the data only to justify their strong belief as adequate precautions are not taken before using the data. Further, it must be noted that all the countries were not affected at the same point of time and hence for meaningful comparative study of countries this issue must also be considered. In addition, conceptual issues need to be examined. The practice of compiling the data is likely to vary from one country to another. Death of a diabetic patient affected by COVID-19 may be recorded as death due to COVID-19 in one country while another country may record it as a death due to Diabetes. Further, a person may die due to multiple complications. In such a case, it is not proper to attribute the death only due to COVID-19. No doubt, COVID-19 may have advanced the death of the person. Therefore, first step in any scientific study would be to focus attention on such short comings and clean the available data. Unfortunately, adequate attention to this aspect is not paid by the Researchers.

inter-State Comparison:

In India, people make use of State-wise data to assess the situation in their state in the background of national picture. People get panic when there is a spike in the number of positive cases and relax on days when no positive cases are reported. Indians draw some satisfaction from the fact that the number of affected people in India is only in few thousands while it is in lakhs in America and in European countries. Supporters of Prime Minister Modi give credit to him for announcing nation-wide lockdown at right time. There is a section of population, who attribute lower rate of transmission of COVID-19 in India to the better immunity power among Indians (which may be due to vaccination in their childhood). The practice of having cooked food at home, hot climatic conditions and lower addiction rate to smoking among Indians may have also contributed for the lower rate of transmission in India.

Inter-state comparison in terms of absolute numbers has one important limitation via; substantial variation in the size of the States. It is of no use to compare the number of deaths in a big state like Uttar Pradesh with the number of deaths in a small state like Uttarakhand. For a scientific study, it is necessary to neutralise for the differences in the size of the States by using suitable deflator like population of the state or the number of persons affected by the COVID-19 in the state. Many studies ignore this basic and fundamental step while undertaking inter-state comparative study.

Fatality Rate and Recovery Rate:

A common man generally deals with figures in hundreds and sometimes in thousands. He is panic driven when he comes across figures which mention the number of affected persons in lakhs and the number of deaths in thousands. He is demoralised and loses his confidence to encounter the disease. The same data can be used to convey positive message which will boost the moral of the general public. The data on COVID-19 can be used to estimate the fatality rate and the recovery rate. Fatality rate is the percentage of deaths among COVID-19 affected persons which will generally be small in magnitude and hence boost up the confidence of general public.  In a Press Briefing on March 3, 2020, the Director-General of the World Health Organisation, Dr Ghebreyesus reported that the fatality rate due to COVID-19 is only 3.4 percent. Experts admit that calculating fatality rate in case of COVID-19 is tricky due to long incubation period. Further, a declining trend in the fatality rate and an improvement in the recovery rate indicates a movement towards victory and hence will build the confidence of the general public to fight the battle against COVID-19 and achieve complete victory in the battle.

Fatality rate and recovery rate can’t be computed scientifically from the data released by the Government of India to the general public. But the Users of the Government data generate some estimates of the fatality rate and the recovery rate. However, they do not explicitly spell out the method.

Scientific Approach:

Scientific method would be to make use of the data, based on the date of admission of the patient to the hospital. Suppose, 100 persons affected by COVID-19 are admitted to a hospital on April 1, 2020. Further, for the sake of argument, let us assume that by April 25, 2020, among these 100 persons, unfortunately 10 persons scumbed to the disease and 90 persons, thanks to the efforts of Doctors and Nurses, completely recovered and were discharged from the hospital. Now, the fatality rate and the recovery rate would be 10 percent and 90 percent, respectively and two rates would add up to100. Surprisingly, in the present context of COVID-19 pandemic this procedure seems to be not adopted either due to non-availability of the data in required format or due to the ignorance of the Users. For Users, other than Government the data may not be available in the required format.

Government Approach:

The Government has not explicitly spelt out the method adopted by it to estimate the fatality rate and the recovery rate. In this context, it is worth exploring the possible method adopted by the Government to arrive at fatality rate and recovery rate based on the data released by it. According to the Government data, out of 33,610 persons affected by COVID-19 till April 30, 2020, 1074 persons died and 8373 persons were discharged (The Hindu dated May 1, 2020) . So, the sum of number of persons who died and the number of recovered persons is 9447. It implies that out of 33,610 affected persons, still 24,136 persons (i.e., 33,610 - 9447) are still in various hospitals undergoing treatment for the disease. We don’t know how many of them will survive. It appears that the Government has not considered this issue. The fatality rate has been computed as percentage of deaths to the total number of cases (i.e., 1074/33,610) which works out to 3.2 percent. Similarly, the recovery rate has been taken as percentage of persons recovered to the total number of cases (i.e., 8373/33, 610) which is about 25.2 percent. As explained earlier, fatality rate and recovery rate should add up to hundred which is not the case in the figures released by the Government. More serious defect in this method is the fact that the Government is estimating the required rates when the outcome of over 70 percent of patients, who are under treatment, is unknown. Such estimates fail to reveal the ground reality. Therefore, the Government has to group the available data with it on the basis of date of admission of persons and work out the fatality rate and the recovery rate for each group. Average of estimates so obtained for sufficient number of groups would be more reliable and meaningful for Policy formulation. Similar procedure may be adopted for examining the vulnerability of people belonging to different age groups or different ethnic groups.

Conclusion:

It is observed that in Inter-country analysis some of the statistical and conceptual aspects do not receive attention which they deserve. Similarly, in inter-state analysis, suitable deflators need to be used. The format of the available data on COVID-19 does not permit us to make meaningful and reliable estimates of the fatality rate and the recovery rate. Even the method adopted by the Government to estimate these parameters seems to be defective. The data has to be grouped on the basis of the date of admission of persons in the hospital. The required parameters have to be worked out separately for each group. The average of these estimated parameters pertaining to sufficient number of groups would be meaningful, reliable and useful for policy formulation.

References

The Hindu paper 1st May 2020,

https://drive.google.com/file/d/1ySFt0jUabDaXpWrfmqXZTsHTy9mrkkvy/view

WHO Media Brief on COVID-19, https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-media-briefing-on-covid-19---3-march-2020

The Economic Times March 25, 2020, https://economictimes.indiatimes.com/news/politics-and-nation/india-will-be-under-complete-lockdown-starting-midnight-narendra-modi/articleshow/74796908.cms?from=mdr

The Authors:

2Indrakant Sulibhavi is Visiting Professor, Centre for Economic and Social Studies, Hyderabad, Telangana, India. Email: indrakant_s[at]yahoo.com

1Mahendran Arumugam is Assistant Professor in Economics, VIT Business School, VIT University, Vellore, Tamil Nadu, India Email: dr.a.mahendran[at]gmail.com, mahendran.arumugam[at]vit.ac.in

1Address for the correspondent:

Dr. Mahendran Arumugam

Assistant Professor in Economics, VIT Business School, VIT University, Vellore, Tamil Nadu, India. Email: dr.a.mahendran[at]gmail.com, mahendran.arumugam[at]vit.ac.in

Notice: The print edition of Mainstream Weekly is now discontinued & only an online edition is appearing. No subscriptions are being accepted