On July 20, GoI made two eye-catching observations in Parliament about the second Covid wave. It said states had reported that there were no deaths due to medical oxygen shortage, or due to lack of testing and treatment. This is plainly inconsistent with collective experience. April 2021 to June was the worst phase in the Covid trajectory, and large parts of the country were locked down as healthcare infrastructure was overwhelmed. GoI’s data showed that 2.35 lakh deaths, or 56% of all Covid fatalities till date, took place in those three months. It beggars belief that not one death was due to the oxygen supply crisis. The problem is states and hospitals are probably using record-keeping protocols to avoid grim facts.
A larger, related question is whether India is undercounting Covid fatalities. There’s been plenty of research on this. The most commonly used source now is the Civil Registration System (CRS), a record of deaths maintained by states. While this system does not offer data on just Covid deaths, it provides a sense of the “excess mortality” in 2020-21 that can be assumed to have been influenced largely by the pandemic. An alternative indicator is the GoI’s Sample Registration System (SRS), a demographic survey, which is available till only 2019.
Arvind Subramanian, a former GoI chief economic advisor, and his associates, recently estimated that, under some assumptions, using CRS shows that excess mortality between April 2020 and June 2021 was 3.4 million. There’s an argument against using CRS, which is that it’s not as accurate as GoI’s SRS. That’s partly true. Data of 2019 shows that in the southern states there’s no mismatch between CRS and SRS. But many other states’ CRS underestimate deaths.
Deaths are often used as a proxy for governance in political rhetoric. Politics over deaths provides perverse incentives to undercount and underplay the severity, helped by the fact that Covid death registration guidelines are too tight. That’s deeply unfair to families of victims of the second wave. They deserve an honest answer. The way out is for GoI to do a proper SRS, a survey that’s been in place for 50 years. GoI’s data shows that so far there are 4.18 lakh Covid fatalities, which is 1.34% of people who tested positive. But if sero surveys indicate that the number of people infected is far larger than test data, what’s the real scale of fatalities? The answer can only come through a nationwide demographic survey.
This piece appeared as an editorial opinion in the print edition of The Times of India.
END OF ARTICLE