Nearly one and a half years after the Covid-19 pandemic began, not a single state government has released its all-cause death registration statistics from the Civil Registration System (CRS) in a systematic and timely manner. These statistics enable one to understand how many ‘excess deaths’ occurred in a particular period when compared with the pre-pandemic period, after factoring in the usual growth rate of deaths over time. On current data release schedules, the data for 2021 will come out only in 2023 in most states. Many other countries have moved to a daily or monthly public release of these statistics to track the pandemic more effectively in real-time but not India.
As a result, researchers and journalists are scraping through government websites and offices to get a hold of this data and over the past two weeks, various articles have been written, confirming what many always suspected: that India’s official Covid-19 death toll is lower than the actual by an order of a magnitude. For just three states of India which have good completeness of death registration statistics and where recent death statistics were uncovered — Andhra Pradesh, Karnataka and Tamil Nadu — excess deaths in the first six months of 2021 calculated by the most conservative method exceeded 300,000 as compared to around 50,000 reported Covid-19 deaths in this period for those states. The potential under-reporting factors for Andhra and Madhya Pradesh (the latter has a weaker death registration system), were well over 15. Gujarat has been placed at over 10 for the period April 1-May 10, 2021. These are staggering under-reporting factors by any yardstick. In the US and many other rich countries, these under-reporting factors have been estimated to lie under 2.
The under-reporting can occur even when there is no data fudging simply because to be counted Covid dead, one needs to have tested Covid positive at some time. The official numbers for Covid deaths are thus hugely dependent on testing capacity which varies from state to state and from urban to rural areas. The data across states consistently shows a massive spike in deaths registered in May 2021, similar to what was seen in November 1918 during the raging influenza pandemic.
While not all ‘excess deaths’ have to be Covid-19 deaths, it is important to point out that one does not observe such large spikes in normal times. Simply put, these are pandemic-related deaths, whether directly by the disease or due to the collateral damage caused by the collapsed health infrastructure during the second wave.
Under-reporting factors will vary from place to place and from wave to wave. In 2020, it is likely that the under-reporting factors were much smaller, though recent data uncovered from the CRS shows a significant number of ‘excess deaths’ in states like Assam, Andhra Pradesh, and several cities such as Hyderabad, Chennai, Mumbai around the peak of the first wave. The ‘excess deaths’ method is useful in those states with near-complete death registration but needs additional verification in states with poor death registration like in Uttar Pradesh, Bihar and Jharkhand. Detailed surveys and the next Census will provide valuable information on uncounted deaths in those states.
The implications of such under-reporting are stark. If you saw the headline Covid-19 death numbers over the past few months, you would think that Andhra Pradesh was a far safer place to reside than Karnataka but it turns out that the reverse was true. The ‘excess deaths’ per capita figures for several states for 2021 alone are also high by global standards, exceeding 2,000 per million. Adjusting for under-reporting, India in all likelihood, is now the most affected country in the world by pandemic deaths, surpassing the US and Brazil, and on per capita terms, one of the worst affected countries in Asia and Africa.
It is imperative that all state governments immediately release daily/monthly death registration statistics at the district level for the past five years at once. Like in the UK, they can mark the statistics as ‘provisional’ for the more recent weeks. This is no longer a matter of mere academic interest for it has direct policy relevance. The allocation of resources, whether it was oxygen or other material, was often based on the reported Covid-19 cases and death statistics which we now know were showing a distorted picture of the ground reality. Let us not be blindsided again in the future. Our business-as-usual approach on data dissemination in a pandemic is proving to be fatal. The question is, which state government will be the first one to do this?
Views expressed above are the author’s own.
END OF ARTICLE