Examining India’s Fatality Rate
The word ‘only’ is highly problematic, but it is a word that is now used quite commonly to underplay the effect of COVID. Can the word be applied in case of India?
‘The fatality rate in India is ONLY 1.x%, while worldwide it is 4%”, is a statement commonly heard. Underlying the statement is the implication that India is somehow doing better than other countries, either because of our demographic dividend or because of some inherent immunity.
We can argue that even 1% is an extremely large number given the population of the country, but before we get to that is the fundamental question- is the fatality rate in India really lower than elsewhere?
The easiest way to calculate the fatality rate is by dividing the cumulative number of deaths that are directly attributable to COVID with the cumulative number of people who have tested positive. Using this formula, the fatality rate in India is 1.69% against a global average of 3.36%. It does certainly seem as though India is doing better.
However, this is not the correct way to calculate the fatality rate, because people do not die immediately after testing positive. The virus takes time to establish itself in the patient, and most deaths occur only a few weeks after testing positive. The correct formula, therefore, is-
Fatalilty rate = (total number of deaths) / (total number of deaths + total number of people recovered)
This data is not easily available in India, because in case of asymptomatic or mildly symptomatic patients, the person is deemed recovered two weeks after testing positive, even in the absence of a negative COVID test.
Therefore, assuming the fatality rate remains stable over time, the formula that can be used is-
Fatality rate = (total number of deaths) / (total number of people who had tested positive 4 weeks back)*
Applying this formula, the fatality rate in India, goes up to 3.17%, against a global average of 4.44%. Though the fatality rate in India is still lower than the global rate, it is no longer much lower as it was earlier.
Is 3% insignificant? Certainly not.
More importantly, if despite a younger population, the fatality rate in India remains as high as 3%, we cannot dismiss the virus as benign as some people try to claim it is.
Once the fatality rate has been shown to be comparable to the global level, the people who want to deny the severity of the virus, will then bring out their trump card. There have been multiple sero-surveys conduced across the country, all of which point to an exposure rate well over what is indicated by the total number of people who have been tested positive. On an average, according to the results of the sero-survey, about a quarter of the population has been exposed to the virus and developed antibodies against it. If the fatality rate is calculated using those estimates, it certainly falls to an extremely low percentage.
There are, however, two arguments that work against it- the possibility of contamination by exposure to a different pathogen, and the underreporting and undercounting of deaths in the country.
The first is easier to deal with. We are not sure about the accuracy of the kits used in the sero-survey. There have been questions asked about whether a previous exposure to a very different pathogen may not produce similar responses. Since we have no data, we can, perhaps, set this objection aside for now.
The second objection is more important. The deaths due to COVID are certainly underreported and under counted.
In the ideal scenario, unless a person dies in an accident, any death that occurs after a person tests positive should be counted directly or indirectly as a COVID related death. This, however, is not the case. If a person has co-morbidities and succumbs to any of them, that is given as the only reason for death. This leads to a gross undercouting.
How then can an estimate be made for the total number of deaths? Three ways have been put forth-
(a) Access the data on cremations and burials that have been conducted with COVID protocals in place, and compare that to the total number of deaths that have been registered as due to COVID. This has been attempted in Delhi, and has revealed that deaths due to COVID could be underreported by a factor of 5.
(b) Estimate the number of COVID related deaths, by extrapolating from the ratio of deaths which mention a medical cause to the total number of deaths. This methodology also throws up the fact that deaths could be underreported by a factor of 5.
© Compare the gross mortality numbers for the current year with the gross mortality numbers of previous years, to estimate the additional number of people who have died this year. Assuming all other factors being common, the difference would give the number of people who died directly or indirectly due to COVID. These numbers are not yet available for the entire nation, so this cannot be used yet.
Even though it is hard to estimate the total number of deaths, it is clear that it is much higher than the actual number reported. Correcting for the accurate mortality figures will certainly counter the increased number of people presumed to be exposed to the virus.
It is, therefore, best to go with the fatality rate calculated using the official data. That figure, is close to 3% of the population that has got infected, and it certainly cannot be dismissed with the word ‘only’.
The pandemic is still raging in India, and the fatality rate is as high as 3%. We certainly cannot afford to not take it seriously. If we do, we do so at our peril.
*the assumption made is that it takes an average of 4 weeks after testing positive for a person to succumb to the virus
Sources-
Gross mortality rate in Mumbaicompared to the previous year- https://indianexpress.com/article/cities/mumbai/covid-outbreak-64349-deaths-in-mumbai-this-year-6587188/
Mismatch in funerals conducted with COVID protocols — https://theprint.in/health/covid-death-data-mismatch-continues-in-delhi-difference-of-over-500-in-govt-funeral-figures/463072/?amp&__twitter_impression=true
Estimating deaths by extrapolating data on medically certified deaths — https://www.thehindu.com/opinion/op-ed/covid-19-deaths-are-not-as-low-as-reported/article32310814.ece
