New meta-analysis of 18 studies finds that smokers are *82% less likely* to be hospitalised with COVID-19 than nonsmokers.https://www.mdpi.com/1660-4601/17/20/7394/htm …
-
Show this thread
-
Replying to @cjsnowdon
I mean, that's a fairly obvious case of survivorship bias and also a remarkably terrible study
5 replies 1 retweet 29 likes -
Replying to @GidMK @cjsnowdon
Just...wow. They took a single estimate of how many smokers they expected to see based on a decade-old sample and then just assumed this applied to all places. No correction for age, no analysis of confoundingpic.twitter.com/RiUFFnHPW7
1 reply 2 retweets 11 likes -
Replying to @GidMK @cjsnowdon
No consideration of reporting bias (they used medical codes to capture smoking status in most studies, which can miss people!)
1 reply 0 retweets 4 likes
No discussion of the included study quality or of hospitalization bias, just a huge number of places that biases could've crept in. They even talk about how the lack of correction for age makes the results problematic in the study!pic.twitter.com/kmcThhcjAo
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.