1/n A cautionary tale of bad #scicomm, in *four* sections, the fourth part being where I learn to take my own advice, the first part where I retro-implement doing so. What are your risks of dying from #COVID19 if you& #39;re under 50? Turns out, that& #39;s quite tricky.
2/ natural temptation is to look at #CFR — how many died (for each age group) out of all confirmed #COVID19 *cases*, which in many places means out of those ill enough from it to end up in hospital. But that wasn& #39;t the question. What the question implies is, what& #39;s the * #IFR*?
3/ This is what age-group CFR looks like for #COVID19. But we& #39;ve already noted that this *doesn& #39;t* tell you your chances of dying from it when you& #39;re under 50, it only tells you your chances if you get sick enough from it to end up in hospital (roughly).
#case-fatality-rate-of-covid-19-by-age">https://ourworldindata.org/mortality-risk-covid #case-fatality-rate-of-covid-19-by-age">https://ourworldindata.org/mortality...
4/ This is what age-group IFR looks like, if you examine as many hard-data studies as you can get your sticky mittens on. This is from a preprint ( https://www.medrxiv.org/content/10.1101/2020.07.23.20160895v4),">https://www.medrxiv.org/content/1... the best immediate guide I could find, but many big caveats apply! But let& #39;s run with this for a moment.
5/ Ignoring all the many (important) caveats that apply, we could say from that:
a/ if you& #39;re under 35, you have one chance in 10,000 of dying from #COVID19 if you catch it at all.
b/ 35<you<44: 6 chances in 10,000.
c/ 45<you<54: 2 chances in 1,000 (note the *big* jump in risk).
6/ Over 54? Gets dismal, start w/ 0.7 of a chance out of *100*, go up to 27 chances out of *100* if you& #39;re over 85, in other words, #COVID19 is bad news. But that wasn& #39;t our question. Our question was, under 35, & under 50.
Under 35? One chance in 10,000 https://twitter.com/Gurdur/status/1303182915501805568">https://twitter.com/Gurdur/st...
7/ From that meta-study, if you& #39;re between 45 & 54, your chances of dying from #COVID19 if you catch it *at all* are 2 chances out of 1,000, say 1 out of 500. But that included those between 51 and 54 inclusive, which mucks it up a little.
That& #39;s the first section concluded.
You can follow @Gurdur.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: