- - Sunday, December 27, 2020

Some people think that the number of COVID-19 deaths has been greatly exaggerated by the mistaken inclusion of people who died from other causes, while coincidentally infected by COVID. They’re wrong.

It is true that some people “with COVID-19” (who test positive), die from other causes, and some of those deaths could be mistakenly attributed to COVID-19. But it is simple to prove, with a bit of arithmetic, that such cases are a small percentage of the total reported COVID-19 deaths.

In normal times, the U.S. records an average of about 230,000 deaths per month, from a population of about 332 million. 230K/332M = 0.07% of the nation’s population. In other words, an average U.S. resident already has about a 0.07% chance of dying within the next month, without a COVID-19 diagnosis.

So, how does that compare to his chance of dying within the next month with a COVID-19 diagnosis?

Let’s calculate that. Here are the latest U.S. figures (as of Dec. 23): 18,687,330 known cases, 330,841 deaths and 10,948,136 recoveries.

So the number of deaths as a percentage of known resolved cases (Case Fatality Rate, or “CFR”) is 330,841 / (10,948,136 + 330,841 deaths) = 2.9%.

(Aside: It is likely that as many as half of all recoveries were mild cases which went undetected, so the true infection fatality rate [“IFR”] might be as little as half the calculated CFR, i.e., it might be as low as 1.5%. But that does not affect these calculations, and it is still more than 10 times as deadly as typical seasonal flu.)

Another way to estimate the CFR is just from recent cases, by comparing the average daily death rate (2736 per day, calculated over the week of Dec. 15-21) to the average daily number of new known cases (212,990 per day calculated over the preceding one to two weeks). 2736 / 212,990 = 1.3%. (The good news is that the CFR is declining, but the bad news is that it’s still much worse than typical seasonal flu.)

About 88% of COVID-19 deaths occur within 28 days of diagnosis. So, multiplying the CFR by 0.88 yields the percentage of people who have a positive Covid diagnosis, who can be expected to die from COVID-19 within the next 28 days: That’s 2.6% averaged over the entire epidemic, or 1.1% averaged over a single recent week.

Dividing 0.07% by 1.1% = 6.4%. In other words, if you have a positive COVID-19 diagnosis, you’re 16 times more likely to die from COVID-19 in the next month than to die from something else.

Most coroners and medical examiners are competent, and so official causes of death are correct more often than not. But even if every case in which someone “with COVID-19” who died from a different cause were misattributed to COVID-19, it still would have only a small effect on the COVID-19 death statistics.

So, if someone tells you that the number of reported COVID-19 deaths is greatly exaggerated, due to misattributions of people who test positive for COVID-19, but who actually died from a different cause, they are wrong. They need to “do the math!”

• David A. Burton is a computer scientist in Cary, N.C. References for the statistics cited in this article can be found on his website, sealevel.info.