Recent Illinois COVID Deaths Not Plausible
Omicron variant doesn't kill enough to cover recent Illinois COVID deaths
Omicron variant is a pussycat compared to wild-type SARS-CoV-2, or even the dreaded Alpha variant. To get out to the limits of plausibility on an Omicron infection fatality rate (IFR), I generated 45 million random IFR values and found which ones led to the results in a recent Omicron study in Ontario.
In the study, they found 12 total deaths from 37,296 Omicron infections, or 1 death for every 3108 infections. That’s an IFR of about 0.032% — under half as deadly as seasonal flu.
After putting through 45 million IFRs, almost 1200 of them resulted in 12 deaths from 37,296 infections:
The upper bound of the 99% credible interval shows the limits of plausibility, a maximum plausible IFR which is still consistent with the known evidence: 0.064%.* If the highest plausible IFR is 0.064%, then the least amount of COVID infections required in order to produce one COVID death is 1562 infections.
Omicron became dominant in the US by 27 Dec 2021 and was over 90% of all COVID positive cases just two weeks later (on 10 Jan 2022):
The rise of Omicron (sounds like a monster movie!) led to almost 30% test positivity in the US:
For two days in a row, Jan 6 and Jan 7, 29.4% of tests were coming back positive. But COVID deaths in Illinois stretch the limits of possibility, given the IFR for Omicron. Here is the accumulated COVID death there on 27 December, when Omicron began to dominate:
And from that baseline of 30,747 COVID deaths in Illinois, we can move forward to 12 Mar 2022 to find that 6563 COVID deaths had been added during the Omicron surge:
At a minimum of plausible COVID infections required to produce one COVID death of 1562 infections (upper bound 99% credible interval), in order to get 6563 COVID deaths during Omicron requires (1562*6563=) 10.25 million COVID infections.
But the entire Illinois population is just 12.7 million, so the limit of plausible IFRs on Omicron make it so that it must be the case that at least 81% of every single person in Illinois got COVID in 2.5 months.
That isn’t plausible given the other evidence on test positivity rates though.
*To investigate the probability of obtaining the sample outcome found in Ontario (12 deaths from 37,296 infections), you can use a binomial distribution with n=37,296 and look for probability that X is 12 or lower — using different IFR values.
Using an IFR of 0.07%, for instance, results in 12 or fewer deaths at cumulative probability of 0.0017 — meaning that if the true IFR of Omicron was as high as 0.07%, you’d only get the Ontario results 17 times in every 10,000 attempts.
Reference
[highest Omicron IFR consistent with the evidence is 0.064%] -- Ulloa AC, Buchan SA, Daneman N, Brown KA. Estimates of SARS-CoV-2 Omicron Variant Severity in Ontario, Canada. JAMA. Published online February 17, 2022. doi:10.1001/jama.2022.2274 Available: https://jamanetwork.com/journals/jama/fullarticle/2789408