Was the prediction for infection fatality rate (IFR) or case fatality rate (CFR)? And high-income or all countries? Globally, the CFR is 2% (3.7M/173M), but the IFR is <0.66%, because <1/3 of cases were detected.
IFR (but back in February/March 2020, a lot of people called everything “CFR”). I think he was talking about high-income countries (that’s what my 0.9% estimate for 2020 referred to – note that it’s lower for 2020+2021 combined because of better treatment and vaccines). I’d have to look it up again, but I doubt that Adalja was talking about a global IFR that includes countries with much younger demographics than the US. It could be that he left it ambiguous. Here’s the Sam Harris podcast in question; I haven’t re-listened to it yet.
Seems unlikely that <1/3 of all cases were detected at this point, since the recent outbreaks had far higher detection rates than the initial ones.
Ah. If global IFR is worse than rich-countries’ IFR, that seems to imply that developing countries had lower survival rates, despite their more favourable demographics, which would be sad.
It’s still unclear, and the developing world detection and survival rates are a bit uncertain. I think you could probably get a decent approximation by looking at test positivity rates and testing volume compared to death rates over time in different countries, but I’m not going to put together the model to do it. We’re doing something related with IFR estimates by age at 1DaySooner, but using seroprevalence data, i.e. only where there is really good data for the estimate. I don’t have results of that yet.
that seems to imply that developing countries had lower survival rates, despite their more favourable demographics, which would be sad.
This isn’t impossible because there seems to be a correlation where people with lower socioeconomic status have worse Covid outcomes, but I still doubt that the IFR was worse overall in developing countries. The demographics (esp. the proportion of people age 70-80, and older) make a huge difference.
But I never looked into this in detail, and my impression was also that for a long time at least, there wasn’t any reliable data.
From excess deaths in some locations, such as Guayaquil (Ecuador), one could rule out the possibility that the IFR in developing countries was incredibly low (it would have been at least 0.3% given plausible assumptions about the outbreak there, and possibly a lot higher).