For example, in the US: Utah has the lowest IFR in the country, with our estimate putting it at exactly 0.5% Indiana has a much higher IFR, at roughly 1.1%pic.twitter.com/PuG8VpL3Tu
You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
For example, in the US: Utah has the lowest IFR in the country, with our estimate putting it at exactly 0.5% Indiana has a much higher IFR, at roughly 1.1%pic.twitter.com/PuG8VpL3Tu
But this is LARGELY explained by differences in the age breakdown of infections - in Utah ~50% of all infections were in people <45yo when we ran our analysis compared to 40% in Indianapic.twitter.com/luJwPXFMGh
This might seem like a small difference, but even relatively minor changes in the age breakdown of infections can have an outsized impact
This does not mean that no other explanations are possible, but correcting for age explains 90% of the difference between IFR in different areas, which is pretty significant!
Also worth noting - given that there are very few longitudinal samples, we cannot exclude the possibility that other things are impacting IFR more recently (i.e. better treatments)
One interesting consequence of the exponential gradient is that you can't divide people into "old" and "young" easily! There's no easy delineation between 'low' and 'high' risk, it's a spectrum
We limited our analysis to seroprevalence studies conducted in OECD countries, but as far as I've seen this does explain a lot of the low death rate in Africa (although that's also a more complex question)
So on that basis, could one say, ironically..the areas with highest COVID19 mortality are the 'healthiest' (nebulous term klaxon) areas, as they have populations that manage to live to ripe old ages in the first place? Or am I making zero sense there? (wouldn't be first time!
)
I note Japan isn't on there, it wouldn't fit that line.
It does, kinda! The problem with Japan is that the only good serosurvey found so few infections that the confidence interval for seroprevalence included 0 (0.1, 0-0.3%) so we couldn't include in the metaregression
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.