> I am suggesting that you are preferring a rate based on an even more biased method, because the method yields a bigger number.
What method? I merely stated that PCR has limitations, I never said I was using it to calculate IFR.
For the record, my assumptions are that IFR is around 1%. That is largely based on the Diamond Princess data. The original paper suggested it was 0.5%[1], but at the time there were only 7 fatalities. The current number is 13. The raw IFR is now up to 1.8%. Crudely adjusting for demographics based on the ratio in the paper puts the IFR around 1%. This is the only population that was both comprehensively tested and occurred long enough ago for most of the cases to resolve (although some cases are still active/critical[2]). And I fully acknowledge that this approach has limitations. It is a small population, it happened earlier in the crisis when we knew less about treatment, etc... But I sill think it is the best data point we have.
> If a handful of internet denizens can quickly point out "methodological flaws", it usually means that the "methodological flaws" they have discovered are well-known and accounted for.
Except the Santa Clara study did not do that. They acknowledged potential bias, but did nothing to adjust for it. From the source:
"Other biases, such as bias favoring individuals in good health capable of attending
our testing sites, or bias favoring those with prior COVID-like illnesses seeking antibody confirmation are
also possible. The overall effect of such biases is hard to ascertain."
What method? I merely stated that PCR has limitations, I never said I was using it to calculate IFR.
For the record, my assumptions are that IFR is around 1%. That is largely based on the Diamond Princess data. The original paper suggested it was 0.5%[1], but at the time there were only 7 fatalities. The current number is 13. The raw IFR is now up to 1.8%. Crudely adjusting for demographics based on the ratio in the paper puts the IFR around 1%. This is the only population that was both comprehensively tested and occurred long enough ago for most of the cases to resolve (although some cases are still active/critical[2]). And I fully acknowledge that this approach has limitations. It is a small population, it happened earlier in the crisis when we knew less about treatment, etc... But I sill think it is the best data point we have.
> If a handful of internet denizens can quickly point out "methodological flaws", it usually means that the "methodological flaws" they have discovered are well-known and accounted for.
Except the Santa Clara study did not do that. They acknowledged potential bias, but did nothing to adjust for it. From the source:
"Other biases, such as bias favoring individuals in good health capable of attending our testing sites, or bias favoring those with prior COVID-like illnesses seeking antibody confirmation are also possible. The overall effect of such biases is hard to ascertain."
The LA and NY papers have not been published yet.
[1] https://cmmid.github.io/topics/covid19/diamond_cruise_cfr_es... [2] https://www.worldometers.info/coronavirus/