Deciphering the Covid death toll

Delhi: The WHO report, which is not yet released, has used techniques of modelling, to estimate that by the end of 2021, as many as 15 million people died of Covid across the world. The official figure, if one adds the deaths confirmed by individual countries, stands at only six million. Controversially, the WHO report has suggested that three million of those unaccounted deaths occurred in India.

A few things need to be underlined when discussing the WHO’s estimates. The organisation has used a mathematical modelling process when it comes to India and other countries classified as Tier-II. For Tier-I countries, it has relied on mortality figures put out by the governments. It is no small wonder that other member-states have written to WHO with specific questions about the use of data and its interpretation.

There are two truths to be acknowledged here. First, it is very likely that in a country such as India, the official number of Covid deaths has been underestimated. The fact that large numbers of people in a populous country live in areas where medical facilities are lacking is bound to result in a situation where the cause of many deaths remains unknown. At the same time, mathematical modelling, which relies on unofficial data and makes projections on a range of parameters, is far from perfect.

When it comes to Covid, we have seen a range of forecasts about incidence turn out to be fantastically wrong – so wrong, in fact, they were not worth the paper they were printed on. Given this, India has every right to expose the deep-seated biases in the methodology. The world needs to be cautious while using demographers, statisticians and data scientists to offer estimates about the Covid toll.

The only reliable way to determine how many excess deaths were a result of Covid is the old-fashioned one. This involves calculating the number of deaths over a particular period and then comparing them with the figures that prevailed over the same period in the previous year. This methodology obviously would need to make minor adjustments for changes in population and even lifestyle (less road accidents during lockdown). But the data it throws up is robust.

A study done by one media house on these lines revealed that there was a sharp increase in the number of deaths in the three or four months that constituted the second wave. But this spike was immeasurably modest in comparison to the wild numbers that were tossed around by mathematical modellers at the time. As for the WHO projection, it is highly unlikely that India could have hidden three million Covid deaths. Our system of reporting deaths may be flawed, but it is certainly not broken to such an extent.