Re: AstraZeneca-University of Oxford vaccine

One word could explain the differential results: Blinding

Blinding is allocation concealment in a randomized trial, such that the recipient of the vaccine (and others) don't know whether they received the vaccine or placebo 1/6
If the people in the 1/2-dose "arm" of the trial had far fewer side effects and couldn't deduce whether they were vaccinated or received placebo, they might have been more "careful" than vaccinated groups in the regular (planned) dose arm who had higher levels of side-effects 2/6
And if people in the 1/2-dose arm couldn't deduce they were vaccinated and were more careful (sheltering, wearing masks) they might have had lower rates of COVID-19 infections compared to those who "knew" they were vaccinated and thus less careful in the regular dosed arm 3/6
Thus, this 90% efficacy in the lower dose arm could be a caused by better "blinding" in the lower dose arm.

Put another way, poorer blinding in the higher side-effect regular dose arm, produced a more biased, lower estimate of 62% 4/6
So strangely, if there was less biased behavior in the 1/2 dose arm, 90% could be the more accurate estimate of efficacy!

Of course, there could be biological reasons for the discrepancy (62% vs 90%) which will come out later, but differential blinding should be considered. 5/6
As an aside, given the high side effects reported in the mRNA trials, it is encouraging that efficacy was ~95%. Many vaccinated in the mRNA trials would have known they were vaccinated (unblinded) and could've been more lax in mask use and distancing 6/6 https://twitter.com/eliowa/status/1325832200282107907?s=20
You can follow @eliowa.
Tip: mention @twtextapp on a Twitter thread with the keyword β€œunroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.