Earlier quoted context omitted.
Fun fact: the trial you referenced failed to show any benefit of COVID vaccination in reducing all-cause mortality in the population they studied: https://www.medrxiv.org/content/10.1101/2021.07.28.21261159v... > During the blinded, controlled period, 15 BNT162b2 and 14 placebo recipients died; during the open-label period, 3 BNT162b2 and 2 original placebo recipients who received BNT162b2 after unblinding died. None…
> Fun fact: the trial you referenced failed to show any benefit of COVID vaccination in reducing all-cause mortality in the population they studied: Is the implication that we should have expected it to? The study endpoints are clearly "vaccine efficacy (VE) against laboratory-confirmed COVID-19 and safety data".
It's a general principle of medicine (or at least, it used to be) that the important outcomes are the actually clinically relevant outcomes. Did people die less? Did they achieve a better quality of life? When you get into proxy metrics you get to this weird place where you can show that something is insanely effective for proxy metric X, and yet it makes no difference (or is even deleterious) in thing-you-actually-care-about Y.