Hoover-affiliated physician Scott W. Atlas wrote:

More importantly, it was never a policy goal to eliminate all cases of COVID-19. That is impossible, unnecessary and illogical, when 99 percent of infected people have no significant illness from it.

I suspect this one of those carefully crafted, politically minded pieces that is true for some definition of "significant" (which he does not define), but possibly misleading in some other sense. (Atlas et al. are also somewhat famous for claiming that the lockdown has killed more people in the US than Covid-19.)

But to focus on my question, for what definition of "significant" is that quoted statement true?

2 Answers 2


Edited to correct misunderstanding of the question (thank you, @Tobias Fritz.)

According to the CDC, 2,886,267 people in the US have tested positive for SARS-CoV-2, of which 129,811 have died (which is likely a conservative number).

That means that 4.5% of infected persons die from the infection (129811 is 4.4975395554188% of 2886267.)

You ask

...for what definition of "significant" is that quoted statement [that 99 percent of infected people have no significant illness from it] true?

It would need to be more significant than death for that statement to be true, since the death rate exceeds 1% by about 4.5 times.

  • 1
    This seems to answer a totally different question, namely whether less than 1% of the total US population have had significant illness from covid, while the question asks whether at most 1% of infected people have significant illness (and the US is not explicitly mentioned in the question itself). Jul 6, 2020 at 12:17
  • @TobiasFritz - Good catch. Jul 6, 2020 at 18:42
  • 2
    Still not right? While the CDC numbers are somewhat imprecise (it should be clear that 'only posPCR'≠case, although some do count in that botched way), your calculation seems to be for CFR, yet you (and Atlas) speak of infections (IFR). The 'death-risk' or IFR is substantially lower than the CFR rate, variously estimated at 0.1/0.3/1%. Lot of arguing over the exact numbers, how to count, other factors, etc, but "4.5%" is most certainly too high. Aug 11, 2020 at 16:54
  • @LаngLаngС - Sounded too high to me as well, but I ran numbers from several sites (CDC, WHO, Hopkins, etc.) and all came out about the same. Obviously subject to change. In any case, an IFR of 1% would make the statement invalid. Ifr doesn't take into account those who didn't die but who were/still are "significantly" ill. Aug 11, 2020 at 18:41
  • Your part in bold isn't under criticism ;) – But do you have any reliabl numbers on 'significantly ill', at all, I see some hospitalisation, some ICU, some ventialtion, and death. But when I had it I was ill, quite unwell ill, for a few days, quarantined at home. I did not search too deep into this, only at my own local district level, but I feel that such numbers aren't properly counted or coded anywhere? Aug 12, 2020 at 7:15

It's an absurd statement, mainly because "significant" is extremely vague. Conceivably he meant "life-threatening or resulting in serious permanent lung damage". It would be an odd meaning to attach to "significant", but just about defensible.

If he believes 99.9 percent of infections by covid are survived, arguable using these figures --thanks for your comment @LangLangC then he might figure that for every covid death, there were nine covid infections that nearly resulted in death, and/or resulted in serious permanent lung damage. I'm doing it now: what is "serious"?

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.