Does COVID-19’s Death Rate Have a Blind Spot?

Virus Outbreak Dallas

Three noteworthy, as-reported averages:

  • WORLDWIDE: 3.4% 
  • SOUTH KOREA: 0.77%
  • ITALY: 7.3% 


Question: Do these figures represent the likely range of COVID-19’s death rate? If not, how far off are they, and how much further off could they become?  


THE COUNTERARGUMENT: Many serious people have been launching a pushback against the above listed figures, arguing that COVID-19’s true death rate is likely around half of what it looks like as-reported. The core of this argument is that we only know with confidence the number of infected who are dead; this number is then being divided by the number of confirmed infected cases so as to calculate the percentage of COVID-19 patients who’ve died. However, since the number representing “confirmed cases” is undoubtedly now dwarfed by the total number of infected people, the denominator in this equation is artificial and far lower than the true number of cases… Hence their conclusion that the actual death rate from COVID-19 is only around half the size the rate from this equation, or about ~ 1.5%. 

REBUTTAL: In a vacuum, in a snapchat of this moment, given the present conditions, the above argument is sound and makes logical sense. However, insofar as it is intended to serve as a reliable projection of the rate to be expected in the future, it contains a fatal error of logic. From the beginning of the outbreak up to the moment, the vast majority of “confirmed cases” have had access to appropriate health service. The availability of hospital beds, ventilators, and the attention of doctors and nurses to confirmed COVID-19 patients has presumably decreased their rates of death. Indeed, the quality of care that confirmed patients have received up to this point is emblematic of unburdened healthcare systems. So while it’s true the global death rate average of ~ 3.4% has been inflated by a misleading denominator, it is also artificially deflated, because thus far it has been measured in our “normal” treatment context — which assumes access to proper resources — and this context is not a given.

In other words, if and when (almost certainly, it’s when) this “normal” context changes in the face of overwhelming COVID-19 cases, then this assumption about COVID-19 patients having access to the treatments they may need to fight the illness will be shattered. What takes its place will be whatever the functional death rate COVID-19 renders to patients in a decidedly non-normal, likely drastically inferior treatment context. 

PREDICTION: COVID-19’s functional death rate in the context of an overburdened U.S. healthcare system will likely be on the order of ~ 2.1% – 3.6%.

The performance of the Italian healthcare system provides a cautionary tale of what COVID-19 looks like if it manages to spread beyond control. The fact that the lethality in Italy is very nearly double that of the current world average is not by random accident — the nation’s set up triage tents and is currently selecting who will live and who will die of the infected. However, it has been noted that a plurality of confirmed infected cases and majority of those deceased from illness in the North Italian region have been in the 80+ years old age cohort. For this reason, the observed Italian death rate as it stands around ~ 7% is, practically speaking, a bit inflated, given the significantly elevated risks posed to this cohort by the virus. For that reason, I would not expect to see a rate this high in the U.S., even in the context of an overburdened healthcare system. 


Curator: BPeelen (B.A. Political Science & M.P.P. Public Policy, Cal Poly)

Published: |7:31 PST // Sunday, March 15 // ’20|

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Up ↑

%d bloggers like this: