COVID-19 Deaths | WHO report questionable, but India needs to improve data systems

The release of the modelled estimates for excess deaths during the Covid19 pandemic by the World Health Organization has raised a lot of hackles within the corridors of power in New Delhi. Though it’s only in the form of data compilations and a final report is awaited, it gives India the unenviable crown of being the epicentre of Covid-19 deaths in the world.

The official tally of Covid19 deaths in the country stands at 0.52 million at the time of writing this piece; the WHO estimate puts the figure at 4.7 million during the period till the end of 2021. This implies that India had the highest mortality due to Covid-19 in the world and that our reporting systems are inefficient or fudged the data. These are unpalatable figures for a government which has been trying to champion a narrative around how the country was able to ‘successfully defeat Covid19’. It is true that the government was able to launch an extremely ambitious vaccination programme which has administered more than 1.9 billion doses since its roll-out, but the new revelations about the undercounting of deaths were not entirely unexpected.

In the middle of 2021, some activists had come out with accusations of massive undercounting of Covid-19 deaths in the country’s most populous state Uttar Pradesh. Most public health experts agree that there has been some undercounting of deaths, due to the political pressure on states to perform and the challenges in the Civil Registration System (CRS), but the scale of under counting mentioned in the WHO report has surprised many.

The Press Information Bureau (PIB) of the government published a rebuttal immediately after the modelling estimates were released. The biggest argument was about the methodology followed for the estimates. The methodology was established by the WHO in collaboration with the United Nations Department of Economic & Social Affairs (UN DESA) through a Technical Advisory Group on Covid-19 Mortality Assessment. The group incidentally had Prof Anand Krishnan of All India Institute of Medical Science, New Delhi, as one of its members.

The team looked at the deaths which would have normally occurred in the absence of Covid-19, based on mortality figures reported in previous years, and compared it with the estimates for deaths in 2020 and 2021. The difference, if any, was termed as ‘excess mortality’. The team also accounted for the possible lower mortality due to road-traffic accidents and influenza as a result of lockdowns and use of masks. They divided countries into two different groups—ones in which mortality data was available and others which did not have systematically compiled data.

India was placed in the latter group, probably because the CRS data published by the Registrar General of India for 2020 and 2021 was not available in the public domain at the time of analysis. (CRS data for 2020 has since been placed in their website)

In mid-April, The New York Times had published a scathing piece on the supposed attempts by India to stall the release of the Covid-19 mortality estimates to the public. The PIB had published a note on the same day, criticising the article and also the methodology followed by WHO. It is very similar to the one published after the release of the estimates. The primary issue raised is about placing India in the list of countries where data was not available for 2020 and 2021.

The use of ‘unofficial data’ from 17 states and extracted data from media reports in the model, has been cited as deep flaws affecting the validity of the estimates. The use of data from Global Health Estimates, which itself is synthesised information from multiple sources, in the model to predict the number of deaths was heavily criticised in the note from PIB. The validity of the extrapolation used for age-sex distribution of deaths and the process of selection of some covariates for the model are also rubbished in the note. The country claims that 99.2 percent of all deaths were registered in 2020 and implies that WHO has relied on patently unreliable data for arriving at the excess death numbers.

It is true that the mathematical models used in epidemiology are as good as the quality of data sources. Therefore, the argument that India’s stance on these modelled estimates are entirely political, does not hold water. But numbers have to be rebutted by numbers; and the government needs to come up with its own comprehensive data, if it hopes to denounce the estimates published by WHO. There are enough institutions like the Indian Statistical Institute which can possibly help medical research institutions in the country to do the data crunching. But the challenge before the government will be to justify its figures, when other modelling based studies have reported almost similar excess mortality in the country in 2020-21.

A study published in The Lancet, one of the most respected medical journals in the world, indicated an excess mortality of 4.07 million which is 8.33 times more than the reported Covid-19 deaths during the same period. The study was primarily led by a group of researchers from the Institute of Health Metrics and Evaluation of the University of Washington in Seattle, and had several venerated global universities assisting them.

Another study by the Economist had also shown 10 times higher mortality than official numbers during the 2020-21 period in India, though the confidence interval for the estimate was very wide. There can be several caveats about the unreliable sources of data and the quality of the models/algorithms, but the fact is that official mortality figures are much lower than the predictions. The open query is about the proportion of under reporting which has happened during the period and its possible reasons. Whether we can attribute all the excess mortality to Covid-19 is another question altogether.

All this show us the need to invest in improving the quality of our Civil Registration System and making real-time data available to researchers across the world. With the remarkable progress achieved in e-governance initiatives, it is quite possible to collect vital statistics without impinging on the privacy of our citizens. The spectacular success of our vaccination program should serve as a template for proactive government interventions in collecting and compiling vital statistics and epidemiological information. But above all, the government should be focusing on improving the public health infrastructure in the country to prevent a repeat of the tragedy which unfolded in the country during April-May 2021.

Philip Mathew is a physician, public health consultant and a doctoral student at Karolinska Institutet, Stockholm. Twitter: @pilimat.

Views are personal and do not represent the stand of this publication.