COVID-19: India’s erratic CFR makes it difficult to calculate toll over 2 years. Experts suggest alternatives

No agreement on exact number of deaths, even though it is certain that India’s CFR is one of the highest in the world

By Taran Deol
Published: Thursday 28 April 2022

Deaths due to COVID-19 in India remain a contentious topic of debate. The World Health Organization (WHO) recently analysed that the actual pandemic toll of the country is four million, a claim that the Union health ministry has denied based on the “validity and accuracy” of the modelling exercise used.  

The Narendra Modi government has also been accused of stalling WHO analysis on excess deaths. His government maintained that about half a million people have died of the SARS-CoV-2 virus, while the global health body’s analysis claimed the toll is at least eight times higher — the most any country has recorded throughout the pandemic. 

Down to Earth (DTE) looked at the case fatality rate (CFR) — total recorded deaths as a share of total recorded cases — in India between March 2020 and April 2022 to gauge how the death rate has changed with widespread primary vaccine coverage and emerging variants of concern.

Deaths typically lag cases by roughly two weeks. So, we plotted a seven-day rolling average of the CFR through the course of the pandemic. 

The findings revealed that India has an erratic CFR that increases in congruence with a surge in cases, typically triggered by a new variant of concern. Cases are on the rise in some parts of the country, following a deluge of cases triggered by omicron in December 2021 and January 2022. The current CFR is actually higher than what it was in January 2021 and the months thereafter, when the vaccination drive had just begun. 

India’s CFR is among the highest in the world (1.89 per cent), according to an analysis by Shahid Jameel, virologist and former head of the Indian SARS-CoV-2 Genomics Consortium (INSACOG). 

The United States has a CFR of 1.5 per cent, France and Italy 0.12 per cent each, South Africa 0.5 per cent and Germany below 0.1 per cent (approaching the CFR for flu).

What explains this? There are myriad reasons for India’s high CFR, but an ineffective vaccine isn’t one of them. 

“Deaths do not escape detection as much as cases do,” Jameel explained. If somebody has an infection but is not getting tested, they are not in the system. 

The 1,000-2,000 daily cases being reported in India currently is a gross underestimation. If the denominator increases, the CFR will automatically come down. 

Source: Made by: Pulaha Roy

Several other experts point to similar concerns.

A major difficulty about interpreting CFR is that the denominator — the number of cases — is by and large the number of virus test-positive samples. This is subject to change based on several variables, including how many tests are being done, whether all tests are being reported and who exactly is getting tested. 

Satyajit Rath, an immunologist and visiting faculty at the Indian Institute of Science Education and Research in Pune, told DTE:

For example, if testing becomes less widespread, and if the people being tested are mostly people who are moderately or severely ill, then the CFR will appear to be high. That would be simply a statistical accident, and not an indication that the infection is more lethal.

What makes relying on CFR to assess how deadly the pandemic is even trickier is the backlog catch up in deaths that India does in reporting fatalities. Since deaths lag cases by roughly two weeks, CFR should be calculated based on deaths recorded today and cases recorded two weeks ago.

However, some experts argue that the reporting lag makes it very difficult to use this metric as a daily measure. 

A lot of these deaths happened a long time ago and we don't really know when. We can't keep a sensible track of the trend when we see this kind of data addition,” said health economist Rijo M John. This is a key statistical issue when looking at CFR, he added. 

The United Kingdom, for instance, has a robust data monitoring and collection system. Its CFR, as per Our World in Data, an online repository of global socio-economic data, has been on a gradual decline since April 2021, remaining below 1 per cent for a year till April 25, 2022. 

The CFR variation also happens as and when COVID-19 waves are recorded. India’s CFR was higher during the omicron wave than it was during the delta wave, even though studies have revealed that the fifth variant of concern is not as virulent as its predecessor. 

Reasons for this is largely the vast figure of missed infections given omicron’s high transmissibility. A greater number of people relied on home tests during the third wave, according to news reports. A majority of them were not recorded as official figures and therefore drove up the CFR. 

What indicator, then, can we rely on to better understand how the death rate of the COVID-19 pandemic has changed with every emerging variant and start of the vaccination programme?

To do this adequately, we would need age and gender-wise time series data on cases and deaths by vaccination status, John explains. “Epidemiologists model this based on available data and infection fatality ratio (IFR) estimates from other countries where such disaggregated data are available. Unfortunately, in India, we don’t have such data at the national level except for some locations.”

A December 2021 study published in ScienceDirect found that “RT-PCR-based crude CFR with correction accounting for under-reporting of asymptomatic cases in locations where testing is extensive should be a reliable indicator of the lethality of COVID-19.” 

It also terms sero-excess IFR — where infections are calculated based on serological surveys and excess deaths used as total deaths — “the most ideal approach”. However, it’s not feasible to rely on this because of insufficient data. 

Subscribe to Daily Newsletter :

Comments are moderated and will be published only after the site moderator’s approval. Please use a genuine email ID and provide your name. Selected comments may also be used in the ‘Letters’ section of the Down To Earth print edition.