[burger]

There's yet another Stanford preprint trying to estimate the infection fatality rate of covid-19.  This one is from a couple of people in the biology department, so no connection with the med school.  https://www.medrxiv.org/content/10.1101/...1.full.pdf

They come up with an IFR of 1.04% (95% limits 0.77-1.38%) by modelling death rates as a function of testing across countries.  Basically, they argue that deaths are reported accurately but numbers of cases depend on testing, so the best estimates of IFR should come from places with the most testing.

It's hard to argue with their logic, though I could imagine some different ways to get at the exact relationship.  Even if deaths are underreported in less developed countries, that shouldn't affect their model too much--it's the countries with the most testing that have the most influence on the results.

This looks carefully done, though I'd need to spend a couple of hours looking it over to be more sure.

[burger]

There's yet another Stanford preprint trying to estimate the infection fatality rate of covid-19.  This one is from a couple of people in the biology department, so no connection with the med school.  https://www.medrxiv.org/content/10.1101/...1.full.pdf

They come up with an IFR of 1.04% (95% limits 0.77-1.38%) by modelling death rates as a function of testing across countries.  Basically, they argue that deaths are reported accurately but numbers of cases depend on testing, so the best estimates of IFR should come from places with the most testing.

It's hard to argue with their logic, though I could imagine some different ways to get at the exact relationship.  Even if deaths are underreported in less developed countries, that shouldn't affect their model too much--it's the countries with the most testing that have the most influence on the results.

This looks carefully done, though I'd need to spend a couple of hours looking it over to be more sure.

[oldalum]

glad it's coming from Stanford researchers so that Ioannidis et al. are not seen as the official Stanford study.

[oldalum]

glad it's coming from Stanford researchers so that Ioannidis et al. are not seen as the official Stanford study.

[BostonCard]

glad it's coming from Stanford researchers so that Ioannidis et al. are not seen as the official Stanford study.

There is no such thing as an "official Stanford study" for this or any other kind of research.

BC

[BostonCard]

glad it's coming from Stanford researchers so that Ioannidis et al. are not seen as the official Stanford study.

There is no such thing as an "official Stanford study" for this or any other kind of research.

BC

[oldalum]

yes I know that, and you know that, but I don't think that much of the general public does. When the only "Stanford" paper in the news is the one Ioannidis has coauthored, it may be seen as the "Stanford" study of disease prevalence rather than a study of disease prevalence by some Stanford faculty.

[oldalum]

yes I know that, and you know that, but I don't think that much of the general public does. When the only "Stanford" paper in the news is the one Ioannidis has coauthored, it may be seen as the "Stanford" study of disease prevalence rather than a study of disease prevalence by some Stanford faculty.

[burger]

Another preprint from someone at Stanford. They used methods very similar to the paper I linked to in the OP and came up with a similar result: IFR ≈ 1%: https://www.researchhub.com/paper/817131/summary

I haven't read this one yet--just browsed through the figures.

It seems more than coincidental that we have two papers now from Stanford essentially refuting the studies by Ioannidis, Bhattacharya, etc.

[burger]

Another preprint from someone at Stanford. They used methods very similar to the paper I linked to in the OP and came up with a similar result: IFR ≈ 1%: https://www.researchhub.com/paper/817131/summary

I haven't read this one yet--just browsed through the figures.

It seems more than coincidental that we have two papers now from Stanford essentially refuting the studies by Ioannidis, Bhattacharya, etc.