New paper claims 11.9 lakh excess deaths in India in 2020: What is the debate over Covid deaths?

Why the debate over Covid deaths remains unsettled

This latest episode is just another chapter in the continuing controversy over the number of deaths caused by the Covid-19 pandemic in India.

The official figures put this number to 1.49 lakh in 2020, and about 5.33 lakh overall. But every study published during, and after, the pandemic has suggested that the actual number could be five to ten times higher. In a 2022 global study of excess mortality, the World Health Organisation (WHO) said about 47 lakh excess deaths had happened in India during the pandemic years (2020 and 2021). That number was also strongly contested by the government.

Data on how deaths are estimated in India.

Assessing excess mortality is an indirect way of attributing a cause to deaths, in this case, Covid-19. It is the difference between the number of deaths expected in a year, as evident from trends from previous years, and the actual number of deaths. For example, official government data show that between 82 lakh and 86 lakh people die in India every year. If the actual number of dead during the Covid-19 years turned out to be significantly higher than this bracket, and is unexplained by annual variability, it would be considered excess mortality. Since Covid-19 was the main aggravating factor during this time, it is implied that most, if not all, of the excess deaths were caused by the pandemic.

The debate over Covid-19 deaths exists mainly because the total number of deaths in India is never a definitive count. Instead, it is an estimate that comes from a regularly conducted survey, called Sample Registration Survey (SRS), carried out by the Office of the Registrar General of India.

There is another way to keep a tab on the number of deaths. Indians are mandated by law to record births and deaths in their families with a government registrar. Once the birth or death gets registered, it becomes part of what is known as the Civil Registration System (CRS). Unlike SRS data, CRS numbers are actual counts. However, not every birth or death in India gets registered. That is why there is a need to rely on SRS numbers.

However, the registrations have risen sharply in the years immediately preceding the pandemic (see box), bringing SRS and CRS data in close convergence. In 2019, for example, 92 per cent of all estimated deaths in the country were registered. However, data after the pandemic is shaky or unavailable.

Doubtful claims

The government claims that 99 per cent of the deaths were registered in the year 2020. However, this data are not available in either CRS or SRS reports for that year. There have been no CRS or SRS reports after that.

Till 2019, the CRS reports used to contain not just the number of registered deaths, but also the estimated number of total deaths from the SRS exercise. Then, the percentage of the registered deaths to the total deaths would be provided. The 2020 CRS report does not have the estimated total deaths or the percentage of registrations. It only gives the number of registered deaths as 81.15 lakh. The 2020 SRS report also does not have an estimate for the total number of deaths.

Further, if 99 per cent of deaths indeed got registered in 2020, then the total number of deaths in that year would have to be less than 82 lakh (since registered deaths were 81.15 lakh). That would mean that in 2020, the first year of the Covid-19 pandemic, fewer people died in India compared to the previous year when 83 lakh deaths are estimated to have happened.

Notably, the crude death rate (CDR) in 2020, as reported in SRS, has remained unchanged from 2019. CDR is the ratio of the number of deaths to the total mid-year population in any given year. CDR has been showing a declining trend over the last few years, suggesting an increase in life expectancy. Fewer people from the population were dying. The decline stopped in 2020 but it also did not show any rise.

The study in the journal Science Advances has not used CRS or SRS data. It has instead relied on the latest round of the National Family Health Survey (2019-2021), a large-scale periodic survey carried out by the Health Ministry. For the first time, NFHS collected death-related data from the respondents. However, it asked the respondents not for deaths in one year but in three years before the survey. As a result, it is not comparable to CRS or SRS.

The NFHS survey was carried out between 2019 and 2021. The researchers used a subset of the data, collected in 2021, to conduct their study about deaths in 2020. The government’s statement said this was a “critical flaw” in methodology.

“The most important flaw is that the authors have taken a subset of households included in the NFHS survey between January and April 2021, compared mortality in these households in 2020 with 2019, and extrapolated the results to the entire country. The NFHS sample is representative of the country only when it is considered as a whole,” the government said.

Time to settle the matter

The issue of the number of deaths caused by the Covid-19 pandemic is not a trivial matter. It is a vital piece of health and demographic data that has important economic and policy implications. Surprisingly, the government has not taken any initiative to make a realistic assessment of the Covid-19 death toll.

While rubbishing the study and claiming that the excess mortality in 2020 was “markedly less than 11.9 lakh”, the government did not offer the 1.49 lakh number as the final and authentic figure. That shows that even the government believes that the 1.49 lakh number is an undercount.

But there is an opportunity to set the record straight. The 2021 Census, which could not be conducted due to the pandemic, might be held in 2026. This is the most authentic and reliable source of demographic data in India, involving a direct headcount of almost every individual. This Census could be utilised to collect the most definitive data on Covid-19 deaths as well.