First, tell me about these numbers

The WHO report: On May 5, the World Health Organisation released a global estimate of the number of people who died due to Covid between between January 1 2020 and December 31 2021. The worldwide total: 14.9 million. The number for India: 4.7 million—which is 9.8 times the official government tally of 481,000.

Counting excess deaths: The report aims to document deaths that remained undocumented or unreported during the pandemic. It measures these ‘excess deaths’ by comparing the total number of people who died before the pandemic—with those who died during it. That difference is expected to give us a reasonable estimate of the Covid death toll—since a big spike in mortality in 2020/2021 would likely be caused by the virus. 

The WHO numbers for India, therefore, indicate the following:

“So India is among the 20 countries—representing approximately 50% of the global population—that account for over 80% of the estimated global excess mortality for this period. Almost half of the deaths that until now had not been counted globally were in India.”

The Indian government numbers: As noted, our official count for 2020/21 is only 481,000. The count as of May 3 is 522,676—which includes the gruesome second wave. The government also released the number of total registered deaths—from all causes—for 2020. The number: 8.1 million—a 6% rise over the previous year. But the government claims that the 474,806 excess deaths in 2020 cannot all be attributed to the pandemic. And it insists only 149,000 of them were Covid-related. 

Key data point to note: The WHO estimate is similar to other published studies on India’s pandemic toll. For example, a study published in Science magazine estimated that 3.2 million people died due to Covid-related diseases between June 2020-July 2021. Here’s a graphic that sums up their findings and methodology:

What’s the government’s argument?

Indian health officials insist that the WHO numbers are bad because they rely on bad methodology. The estimate is based on faulty statistical modelling—a “one-size-fits-all approach”—not actual data. The government dismisses it as “fallacious, ill-informed and mischievous in nature.” This is also why the government released the registered death data ahead of the WHO report—pointing to it as a more accurate measure of actual deaths.

Point to note: Indian Express supports this critique, arguing that other such estimates of India’s numbers share the same weaknesses:

“It is often argued that, because multiple studies have been pointing to similar estimates, they must be reflective of the true death toll in the country. What is being overlooked is that these studies have been throwing up similar estimates probably because they have all been using similar mathematical models and statistical methods. The researchers, and those doing peer reviews, belong to an overlapping set of people.”

But, but, but: here’s why many are sceptical of the government’s pushback—and its interpretation of the registered death numbers.  

One: For starters, as the WHO points out, the government flatly refused to share mortality data with its researchers. It did not even share monthly estimates—as first promised. But contrary to the government’s claims, the WHO clarified that it did not use a statistical model to arrive at its tally:

“[F]or India, we use a variety of sources for the registered number of deaths at the State and Union Territory level. The information was either reported directly by the States through official reports and automatic vital registration, or by journalists who obtained death registration information through Right to Information requests.”

And a number of other allegations made by New Delhi about the WHO model have also been proven untrue. 

What remains debatable: is the accuracy of registered death numbers used by the WHO—as reported by the states and media houses in 2020. These were only “provisional” at the time. That said, given the lag in death registrations in India, these numbers ought to go up rather than down over time. 

Two: The government claims that the 6.2% increase in the number of registered deaths in 2020 over 2019 is not just due to Covid—but reflects “increase in population, increase in reporting (registrations), more awareness, etc.” And government welfare schemes like Jan Dhan Yojana—which encourages people to open bank accounts—have incentivised reporting deaths. In fact, the health ministry claims registration levels have jumped from 92% in 2019 and 99.9% in 2020. 

But the government’s own National Family Health Survey shows that only 71% of deaths were formally registered between 2019 and 2021. In the two most populous states, Uttar Pradesh (47.4%) and Bihar (36.4%) the reporting numbers are even lower. And that 99.9% number seems all the more bizarre in a pandemic year—when so many were under lockdown—and less likely to register deaths in time. Most states also reported huge staff shortages at the time.

Three: The most puzzling is the government’s insistence that only 149,000 of 474,806 excess deaths in 2020 are due to Covid-related reasons. As Hindustan Times points out, deaths due to other reasons—like traffic accidents, for example—fell sharply in that year due to the lockdown. And more notably, 45% of those who died did not receive medical attention—which is the highest percentage ever. So it is likely they would never have been diagnosed with Covid—even if they died due to the disease.

Data point to note: India has long had a problem collecting reliable death data:

“About half the total deaths occur at home, especially in villages. Poor record-keeping means that out of 10 million deaths every year—based on demographic studies and estimated by the UN—seven million do not have a medically certified cause of death and three million fatalities are simply not registered. Women are undercounted and registration is especially low in the poorest states such as Uttar Pradesh and Bihar.”

It is unlikely the pandemic would have improved matters. 

Also telling: Covid compensation claims in many states are far higher than the official number of deaths.

Why does the actual number matter?

WHO experts claim that the government is intent on preserving its official narrative—that India has been far more successful in battling the pandemic than other countries. Bad numbers ruin that feel-good story:

“I fear that by now even if [all] the data is available, the government would be hesitant to make it public as it conflicts with their published [death] figure and the narrative that India beat Covid due to various reasons.”

Medical experts argue that this stubborn attitude makes for bad public health policy: 

“Every government does not want bad news. But India has done very well in areas such as its vaccination program. Accurate mortality data would only help judge how this program has helped reduce mortality… Counting the dead due to the pandemic would help the living. It’s a social responsibility to know where we are going with this health crisis. Why should we battle Covid blindfolded?”

The bottomline: It's a sorry situation when accurately counting our dead depends on party loyalty.

Reading list

BBC News has the best overview of the controversy—with context on death registrations in India. IndiaSpend offers a detailed rebuttal to the government’s claims about the WHO numbers. The Hindu does a good job of summing up all the other estimates of our mortality numbers—including its own analysis. Indian Express critiques the WHO estimate—pointing to problems in relying on registered death data (which is ironic). Hindustan Times offers a measured analysis of the registered death data—and argues we need to wait for the 2021 numbers before jumping to conclusions.