| It is now clear that most Indians do not blame the government for pandemic-related losses of lives and livelihoods. That is understandable. I do not think any government in the world managed the fallout of the coronavirus disease pandemic well — and I do not think any could have. There were simply too many unknown unknowns (in addition to the known ones). I also believe that India has managed its vaccine drive exceptionally well (after a patchy start). It has covered almost the entire adult population and while I do think the government took too much time to expand the eligibility for booster shots for all adults, there are scientists who believe that there is still no evidence in favour of a third dose in India (I completely disagree). But more on vaccines anon. Given this, and given that most people also believe that the death toll from the pandemic in India is definitely more than the government's number of half a million, New Delhi's prickliness on this count may be misplaced — although, given its avowed subscription to the narrative that it managed the pandemic perfectly, not surprising. That prickliness was in evidence this week in India's response to a New York Times article that claimed India's opposition to WHO's estimate of the real toll of the pandemic was preventing the release of these numbers. NYT claimed these numbers were "at least 4 million". My colleague Jamie Mullick captured the methodological aspects of the controversy in one of our Page 2 stories (Pick of the Day, we call them). So, how many people in India died from the pandemic? The answer to this question depends on assumptions. One, how many infections were there for every recorded case? Was it 10, 15, or 20? Seroprevalence (the presence of antibodies in the population) data would suggest a number of 20 – working backward that would mean India's 43 million recorded cases to date translate into 860 million infections. That, in turn, translates into an exposure rate of 66%, which is actually lower than that cited by various experts on the proportion of the Indian population exposed to the virus. A conservative estimate, in India, would be 15 – and that would translate into 645 million cases and an exposure rate of almost 50%. Two, what is the infection fatality rate? Is it, as Murad Banaji and John PA Ioannidis said in separate papers (both well regarded, but dated), 0.23%? Or is it lower? Does it change by demographic profile of a nation? Does it change by per capita income? Could India have a lower than 0.23% IFR, say 0.115%? I mentioned these assumptions in May 2021 in the column on Covid I used to write. What does this mean for India's death toll? For an infection fatality rate of 0.23%, at the higher end of the estimate (20 infections for every recorded one), it would mean almost 2 million dead. At the conservative end (15 infections), it would mean almost 1.5 million dead. And at the lower end of the estimate (10 infections), it would mean a million dead. And for an infection fatality rate of 0.115%, the number of dead would range from 645,000 to 1.29 million, still higher than the official records. That's a broad range — 645,000 to 2 million. This is not to suggest this was the actual death toll from the pandemic — although it is very likely that the actual number falls in this range — but merely to highlight how, in the absence of actual numbers, estimations can arrive at widely varying results. Given the issues with the civil registry system or CRS database, which has used by some to arrive at death estimates — Shamika Ravi listed them in an article in Hindustan Times — and the need (both scientific and moral) to arrive at a reasonably accurate death toll, what should India do? ICMR already has data on the proportion of Indians with antibodies to the Sars-CoV-2 virus, as do the health departments of many states and some cities. It is theoretically possible to arrive at infection fatality rates for states (even cities); these can be estimated for urban and rural populations in provinces, even across income levels. Together, the two could help arrive at an estimate for the death toll. Perhaps the health ministry and ICMR can work with independent experts, including those advising WHO, to do this. | 
