COVID casualties: The uncounted dead

A Lancet study estimates that more COVID-19 deaths occurred in India than official statistics show, but the Central government refuses to acknowledge the possibility and instead attacks the report’s credibility.

A hospital staffer waits beside the body of a person suspected to have died of COVID-19 complications, in Ahmedabad on January 25, 2022.

A consistent feature in the Central government’s management of COVID-19 has been the underplaying of the mortality figures caused by the coronavirus. It has never been able to come to terms with the proposition that the number of actual deaths due to COVID-19 might actually be far higher than what the official figures have stated.

A recent study published in the medical journal Lancet titled “Estimating excess mortality due to the COVID-19 pandemic: A systematic analysis of COVID-19 related mortality, 2020-21” has estimated excess deaths in 191 countries and 252 subnational units in selected countries, including India, from January 1, 2020, until December 31, 2021. The study has understandably raised the hackles of the government, even though it admits that the study has been published in a “reputed international journal”.

This is the second occasion in recent times that the government has found issue with findings of studies published in Lancet . The first time it raised objections was when a Lancet study in early March 2022 projected that 19 lakh children had been rendered orphans and had lost their primary caregivers as a result of the pandemic in India. A strong reaction came from Smriti Irani, Women and Child Development Minister, who said that it was at best “sophisticated trickery to create panic among the masses”.

In a separate statement, her Ministry rubbished the findings and said that they had no correlation with the ground reality. It added that the actual number, based on data from State governments and Union territories, was around 1.5 lakh. The National Commission for Protection of Child Rights (NCPCR) also shot off a missive to the journal demanding the sources of the study.

Global toll 3.07 times higher

The Lancet study said that in total some 1.82 crore people may have died globally because of the virus—3.07 times the reported number of deaths. The authors of the study said that it was the April-August 2021 surge in South Asia that pushed the cumulative excess mortality rates to levels that were observed in some high-income countries. There were limitations in separating deaths due to COVID-19 and from other causes as only 36 countries had released the cause of death data for 2020.

According to the paper, which was published on March 10, the full impact of the pandemic went beyond the reported deaths due to COVID-19 and official statistics have provided only a partial picture of the true burden of mortality. Their estimates suggest that the mortality impact from the pandemic was more devastating than the number of officially reported deaths. Further, the gap between estimated excess mortality and reported COVID-19 deaths was much larger in South Asia and sub-Saharan Africa as compared to other regions. This difference could be because of under-diagnosis owing to insufficient testing, reporting challenges, higher-than-expected mortality due to other diseases caused by pandemic-related changes in behaviour, or simply a function of reduced access to health care and other essential services.

Explaining the gaps

Excess mortality due to COVID-19 is the net difference between the number of deaths during the pandemic and the number of deaths that would be expected as per past trends in all-cause mortality. According to the authors, this would be a crucial measure of the true toll of the pandemic. The gaps were there as a result of at least five reasons as explained by the authors.

First, health reporting systems did not list COVID-19 as the cause of death without a positive test and hence deaths were likely to be missed in countries with low testing capacities. For instance, even before testing facilities were set up, in many high-income countries the deaths among the older populations after the onset of the pandemic were not attributed to COVID-19.

Second, different countries had different systems for registering deaths that varied in both quality and comprehensiveness.

Third, there was no universal agreement in the global community about when the death of an infected person should be reported as a COVID-19 death.

Fourth, factors such as political considerations also played a role in under-reporting of deaths.

Fifth, social, economic and behavioural responses to measures like lockdowns would have altered the magnitude and distribution of deaths due to other causes.

Therefore, excess mortality was the only available tool for an accurate assessment of the total mortality due to COVID-19. The authors acknowledge the problem with ascertaining the exact contributors of excess mortality. Using empirical excess mortality estimates, the authors developed a statistical model to predict excess mortality for 191 national locations and 252 subnational locations.

For the purpose of the study, the authors collected all-cause mortality reports for 74 countries and 266 subnational locations, including 31 locations in low- and middle- income countries that had reported weekly or monthly deaths due to all causes during the pandemic in 2020 and 2021, including excess mortality rates for 12 States in India.

The methodology included selecting 15 covariates such as seroprevalence rates, population metrics such as health care access and quality index. The understanding behind the study was that an accurate measurement of COVID-19 deaths for each country was necessary to figure out the magnitude of the impact of the pandemic on public health—a perfectly reasonable proposition. An accurate measurement was also important to ascertain the determinants of variation in the infection-fatality ratio among populations.

Global distribution of excess deaths

One of the main findings of the Lancet paper is that although the reported COVID-19 deaths worldwide were 59.4 lakh, actually an estimated 1.82 crore people had died, accounting for excess deaths due to the pandemic between January 1, 2020, and December 31, 2021. The global all-age rate of excess deaths due to COVID-19 was 120.3 per 1 lakh population. The paper said that the number of cumulative excess deaths was the largest in India (40.7 lakh) followed by the United States (11.3 lakh), Russia (10.7 lakh), Mexico (7,98,000), Brazil (7,92,000), Indonesia (7,36,000) and Pakistan (6,64,000).

Also, the excess mortality rate was the highest in Russia, followed by Mexico, Brazil and the U.S. Clearly, India was not singled out by the study. The seven countries mentioned above accounted for more than half of the excess deaths worldwide in the 20-month period. During the same period, there were more than 250,000 estimated excess deaths each in six countries: Bangladesh, Peru, South Africa, Iran, Egypt and Italy.

Among regions, the number of excess deaths due to COVID-19 was the highest in South Asia, North Africa, West Asia and Eastern Europe. Whether there was an economic correlation to the observed trend was not clear, but it was evident that countries falling in these regions were low-income ones.

The study said that Bolivia saw the highest estimated excess mortality rate, at 734.9 deaths per 1 lakh population. In as many as 21 countries, the excess mortality rate was more than 300 deaths per 1 lakh population.

There were wide variations in the ratio of excess mortality rate to reported mortality rate, between countries and within countries as well. India had an estimated 152.5 excess deaths for every 1 lakh persons, which was much higher than its reported mortality rate of 18.3 per 1 lakh population.

Eight States—Uttarakhand, Manipur, Maharashtra, Chhattisgarh, Haryana, Himachal Pradesh, Punjab and Karnataka—had excess mortality rates higher than 300 for every 1 lakh persons between January 1, 2020, and December 31, 2021. Seven States—West Bengal, Madhya Pradesh, Tamil Nadu, Karnataka, Bihar, Uttar Pradesh and Maharashtra—had more than 200,000 excess deaths each as of December 31, 2021. Bihar, U.P. and Maharashtra had a greater number of excess deaths than South Africa. Also, there were nine States where the excess mortality rate was lower than the global average of 120.6 per 1 lakh.

Even though the excess mortality rate due to COVID-19 was not the highest in India, it still accounted for 22.3 per cent of global excess deaths.

Govt response

The government’s response to the study, in the form of a statement as part of its “COVID-19: Myths and Facts” series issued by the Health Ministry, once again highlighted the fact that its objective was not to get at the truth of how many Indians perished in the course of the pandemic but to suppress it altogether.

While the premise of the Lancet study, that is, its rationale, was that “mortality statistics are fundamental to public health decision making”, the Government of India said that death was a “sensitive matter” in the context of the pandemic and was of the opinion that this kind of “speculative reporting has potential to create panic in the community, can misguide people and should be avoided”.

It described the study as “yet another estimate” on excess mortality due to COVID-19 by “another set of researchers”. The authors, it said, had themselves admitted to several “methodology flaws and inconsistencies”. Data sources for India, it stated, “appears to have been from newspaper reports and non-peer reviewed studies”. Data had been taken from another non peer-reviewed model as an input, which raised serious concerns about the accuracy of the statistical exercise. The likelihood of under-reporting in India, the Ministry said, was less for several reasons.

First, national guidelines had been issued for reporting of deaths. Second, the reporting and updating of death data were done regularly. Third, the backlog from States was regularly reconciled with the Central Ministry data. Fourth, there was a financial incentive to report COVID-19 deaths as the deceased were entitled to compensation.

In order to paint the study as a purely speculative exercise, a statement from the Health Ministry, issued a day after the study was published, attempted to undermine the study’s credibility by questioning its methodology and its data sources.

A scientific exercise such as the Lancet study is not above scrutiny and questioning. However, a review of it can be done only by fellow experts in the field, a process which may or may not yield a scientific consensus about the accuracy of its estimates or throw up alternative ones. Most importantly, however, even if there is a debate and contestation in the process, the ultimate objective would still remain figuring out the exact number of COVID-19 deaths, and it would be hard to find experts who would deny that official statistics across the world are gross underestimates of this count.

Had the Central government been serious about addressing this issue, it would have first acknowledged the possibility that the number of COVID-19 deaths in India was higher than the official count and initiated processes to generate a more accurate count, as has been done in other countries, including compiling and releasing data available with the government, which would help generate better estimates of mortality.

A sero-survey by Indian Council of Medical Research (ICMR), based on sample data, had estimated that 67 per cent of the public was carrying antibodies in June-July 2021, the bulk of them through exposure to the infection and not vaccination. These results were announced at a Ministry briefing and their validity was not questioned. This was despite the implied number of infected people being several times the number of confirmed cases reported in the official data. Instead of accepting the obvious inference that the death numbers should also be similarly larger, in the context of the Lancet study, the government has suddenly realised that sample-based estimations “have failed repeatedly to give reliable results for a large, diverse population”.

One of the assertions made in the Ministry’s statement is that the COVID-19 death data in India have an inbuilt tendency to be accurate because there is a financial incentive for reporting such deaths owing to monetary compensation associated with it. Even if one ignores the fact that this system came very late and only after the Supreme Court’s intervention, there is the rather troubling fact for the government that in several States the number of claims for compensation received, and even the ones accepted, were greater than the number of official COVID-19 deaths. This indisputable fact, of course, finds no mention in the Health Ministry’s statement.

While updating of COVID-19 death data has been done to an extent in several States, only in Kerala has there been a continuous and still ongoing exercise linked to the process of giving compensation. It is hardly surprising, therefore, that the Lancet study found that while the estimated excess mortality rate per 1 lakh population in Kerala was much less than in several other States, the ratio between excess mortality rate and reported mortality rate of the State was significantly lower than in all States except Goa. In other words, the under-reporting of COVID-19 deaths was of a much lower order in Kerala.

The most telling parts of the Health Ministry’s statement relate to the issue of all-cause excess mortality (above average) during 2020 and 2021. In one part, the statement says this about the Lancet study: “For States where Civil Registration System was available, reported deaths during the pandemic has been compared with average reported deaths for the same period in the year 2018 and 2019 which doesn’t take into account multiple pandemic management efforts including lockdown, containment zones, testing and contact tracing, wider dissemination and implementation of clinical management protocols and world’s largest vaccination drive, which form the foundation of pandemic management in the country.”

It further questions the applicability to India of the assumption in the study that most of the excess deaths are attributable to COVID-19. Thus, the fact of excess mortality is not denied: what it denies is that these were due to COVID-19. It might then be asked, if 40 lakh, or even 30 lakh, additional deaths occurred during these two years compared to a normal year, and less than 5 lakh these was on account of COVID-19, why was the government focussing everyone’s energies on managing the pandemic instead of dealing with the mysterious other cause of these exceptional deaths? Also, to be fair to the study, it was not about comparing the COVID-19 management efforts of individual countries but about figuring out how many excess COVID-19 deaths may have occurred.

The study acknowledges that deficiencies in data were possible owing to inadequate mortality data reported by as many as 36 countries for 2020. The situation would not have been different in 2021. Data of all kinds help frame and direct policy; in the case of COVID-19, it is all the more essential in order that there is preparedness to meet any future challenges posed by new variants rather than patting oneself on the back for doing things that any government ought to have done for its people.