Follow Us:
Tuesday, July 05, 2022

Why ‘excess mortality’ figures for Covid must be calculated

They will not only help capture the true scale of the tragedy, but will also help in planning better for the next waves of the pandemic.

Written by Chinmay Tumbe |
Updated: May 26, 2021 7:36:08 am
In response to the current pandemic, we need to have a real-time release of death registration statistics. (Illustration: C R Sasikumar)

In his memoirs, the writer Suryakant Tripathi (1896-1961), better known as Nirala, described the river Ganga as “swollen with dead bodies” when the deadly second wave of the influenza pandemic struck India in 1918. The pandemic was a deeply traumatic experience for him, as his family “disappeared in the blink of an eye”. Other oral histories from that period describe dead bodies being thrown off cliffs or in jungles. When the Sanitary Commissioner of India, Norman White, gave the estimate in 1919, of six million deaths during the pandemic, even he acknowledged that the true number must have been much higher because the statistical system recording deaths also broke down during the pandemic. A few years later, when the Census was conducted in 1921, officials found many villages depopulated and nudged up the pandemic death toll estimate in excess of 10 million. Over the next century, several researchers used refined demographic estimation methods based on “excess deaths” or “inter-censal growth rates” to nudge the number even higher. My own research, which accounted for deaths in the princely states as well, places the death toll of the 1918 pandemic in the Indian subcontinent to 20 million or three times the official figure. This was around 6 per cent of the total population and one of the greatest demographic shocks ever recorded in history. The best estimates for the global death toll of the 1918 pandemic, which were initially pegged at around 20 million, now exceed 40 million.

A century later, we find ourselves caught in the second wave of another deadly pandemic and brave journalists on the ground in India are showing that the true death toll is far in excess of the reported Covid-19 fatality numbers that are released on a daily basis. In Gujarat, on May 14, the newspaper Divya Bhaskar broke a story showing that between March 1 and May 10, around 1,23,000 death certificates were issued as compared to only 58,000 in the same period in 2020. These “excess deaths” numbering around 65,000, when compared to an official Covid-19 death scorecard of around 4,000, point to a potential under-reporting factor of 16. The Gujarat government has not disputed the 2021 numbers but correctly pointed out that 2020 was a problematic reference year because the national lockdown halted the death certification process and that deaths increase every year by a certain amount any way. So I use the April-May period of 2019 as the reference period and project the deaths for the 40 day period of April 1-May 10 in 2021 as if it was a normal year using the average growth rate of deaths in 2017-19. The analysis is displayed in the table and chart. This shows an “excess mortality” figure of around 43,000 compared to 3,992 reported Covid-19 cases for this period or an under-counting factor of close to 11. This is lower than 16, what Divya Bhaskar reported, but is still very high and much greater than 3, the value I found for India during the 1918 pandemic. Using the excess mortality method, the under-reporting factor for the USA and many other high-income countries in the current pandemic, as per an estimate by the University of Washington, is between 1 and 2.

The under-reporting factor in Gujarat also varies substantially across municipal corporations, where death registration statistics are near-complete, and in general seem to be larger for smaller settlement sizes: Ahmedabad (4), Vadodara (16) and Rajkot (30). If we compared Ahmedabad and Rajkot on reported Covid-19 deaths per capita, one would conclude that both were similarly affected, but using “excess deaths” per capita, Rajkot was ten times more affected than Ahmedabad.

Best of Express Premium
UPSC Key-July 5, 2022: Why to read ‘Muscular Regional Policies’ or ‘Allur...Premium
As PAGD stitches up joint poll front, BJP & smaller parties face a ch...Premium
Making sense of Assam floods: Why rivers turned violent in the 20th centu...Premium
BJP and Shiv Sena allies might not go after Uddhav factionPremium

When we work with “excess deaths” rather than the reported Covid-19 death figures in Gujarat, one also observes little relation with reported Covid-19 caseloads, which are highly dependent on testing capacity. Jamnagar, for instance, has a higher reported Covid-19 caseload and death toll per capita than other municipal corporations of its size and yet its under-reporting factor is nearly a fourth of the others, suggesting relatively more accurate reporting. Simply put, while our reported figures on Covid-19 may be useful to gauge some trends in the pandemic, they are simply not able to capture the scale of the tragedy. As Covid-19 sweeps through rural India, especially in states like Uttar Pradesh, Bihar and Jharkhand where death registration statistics are incomplete, the death toll is likely to be staggeringly higher than what is reported as Covid-19 death and unrecorded until the next Census takes place. But even the most conservative extrapolations from the available excess mortality data take the all-India death toll of the second wave to over a million.

India’s death registration system started in the 1860s in response to the cholera pandemic. In response to the current pandemic, we need to have a real-time release of death registration statistics. This sounds idealistic, since our politicians and administrators have little incentive to do so but if they actually did — and it literally requires only a flick of the pen — we could still identify the places, currently invisible on the reported Covid-19 dashboards, which need immediate attention and resources to save lives. Understanding the true scale of the tragedy today will also enable better district-level planning to counter future pandemic waves. Indian policymakers should take heed of some ancient wisdom on public health surveillance mentioned in the Arthashastra: “Any doctor who is called to a house to treat a severely wounded person or one suffering from unwholesome food or drink shall report the fact to the gopa and the sthanika. If he makes a report, he shall not be accused of any crime; if he does not, he shall be charged with the same offence (which he helped to conceal).”

This column first appeared in the print edition on May 26, 2021 under the title ‘Why we must count the covid dead’. Chinmay Tumbe is a faculty member at IIM-Ahmedabad and the author of The Age of Pandemics, 1817-1920: How They Shaped India and the World

Express Explained Go beyond the news. Understand the headlines with our Explained stories

📣 Join our Telegram channel (The Indian Express) for the latest news and updates

For all the latest Opinion News, download Indian Express App.

  • Newsguard
  • The Indian Express website has been rated GREEN for its credibility and trustworthiness by Newsguard, a global service that rates news sources for their journalistic standards.
  • Newsguard
0 Comment(s) *
* The moderation of comments is automated and not cleared manually by