In my last post, I explained why we in India need to take the Coronavirus seriously. In this post, I would like to express my concerns about the manner in which the fatality rate of this virus is calculated and why the numbers may not be telling us the right story.
According to the the World Health Organization’s director general, Dr Tedros Adhanom Ghebreyesus, the fatality rate is 3.4%. But his figure is calculated by dividing the number of deaths by the number of officially confirmed cases. To me that doesn’t make any sense. I think in order to calculate out the actual fatality rate you need to divide the number of deaths by the number of closed cases. By closed cases, I mean cases where the result has been a discharge from hospital or a death. If you divide the number of deaths by the sum of all open and closed cases you get a very large and false denominator which results in a fatality rate that is much smaller than what it really is.
Take at a look at the numbers as of today (18-Mar-2020).
If you divide the number of deaths (7,987) by the sum of active and closed cases (107,673 + 90,749), you get a fatality rate of 4%, which is close to the number that WHO is putting out. But if you look at just the closed cases, the mortality rate is 9%. Consider the number of serious or critical cases (6,415). A large number of these may end up dying. So if you are adding the number of active cases to the denominator you need to factor these future deaths in the numerator as well.
If this is confusing to you, consider this situation. Say a chess champion is playing 10 separate games with 10 different people at the same time. Let’s say that 4 games have ended and the remaining 6 are still going on. Say he won 2 of the 4 games that ended. His win rate would be 50% (2 out of 4) and not 20% (2 out of 10). The remaining 6 games are yet to finish. So adding them to the denominator to calculate his win rate would be incorrect.