View Single Post
Old 10th November 2021, 01:14 PM   #300
marting
Illuminator
 
marting's Avatar
 
Join Date: Sep 2003
Posts: 3,982
Originally Posted by Klimax View Post
Quote:
Findings: Vaccine effectiveness of BNT162b2 against infection waned progressively from 92% (95% CI, 92-93, P<0·001) at day 15-30 to 47% (95% CI, 39-55, P<0·001) at day 121-180, and from day 211 and onwards no effectiveness could be detected (23%; 95% CI, -2-41, P=0·07).
Hm, I would ignore that last part about no protection. P-value is high. (Null hypothesis cannot be dismissed)
It's really standard stat. speak when p > .05. The last part does indicate waning after 211 days with a point Ve of 23% but a large CI. For that matter, detecting no protection is never possible. At best one can determine that a benefit or harm is statistically likely. Actually determining no effect can't be done. So what the paper is saying is that the p value is to high to exclude no efficacy in the 95% sense. But it does show some efficacly is likely but with a broad range.

Also, these studies are observational with all the associated confounding issues. Are vaccinated people more likely to get tested because they are more health aware or are they less likely because they ascribe any symptoms to something other than Covid-19 since they are vaccinated? That's a big potential confounder.
__________________
Flying's easy. Walking on water, now that's cool.

Last edited by marting; 10th November 2021 at 02:29 PM.
marting is offline   Quote this post in a PM   Nominate this post for this month's language award Copy a direct link to this post Back to Top