3 ms·
>if P(death|vaccine) < P(death|covid) then it seems really barbaric not to give it. No, it would have to be P(death|vaccine) < P(death|covid) * P(covid) becaus
by fauigerzigerk 6y ago
>if P(death|vaccine) < P(death|covid) then it seems really barbaric not to give it.
No, it would have to be P(death|vaccine) < P(death|covid) * P(covid) because not everyone would get covid during the pre-approval period. That's a far higher bar.
Also, let's not forget P(covid|vaccine). People would behave very differently once they had the vaccine and feel safe. So if the vaccine isn't effective, that could be catastrophic.
- barnacled 6y agoThat's a very good point, latter should be P(death ^ covid) I suppose! I still think the numbers would arguably work out well given the general safety especially of the mRNA vaccines (as I understand it). It wouldn't be a simple or easy decision but it's one that should be considered (very carefully!)