I don’t know about that. We’ve drawn conclusions about obesity and Covid from similar evidence.
-
-
Replying to @cjsnowdon
Sorry, yes, I didn't put that correctly. What you'd need to do is look at the whole at-risk population, whereas this study simply compares smoking rates in one group with a survey done 10 years ago. Can't really garner much from that!
2 replies 0 retweets 1 like -
Replying to @GidMK @cjsnowdon
I mean, leaving aside the statistics for a second, can we really say much about comparing the number of smokers hospitalized today with COVID-19 to the mean rate of smokers in 2010? What would that even tell us?
1 reply 0 retweets 0 likes -
Replying to @GidMK
I don’t think it’s the 2010 date that’s the problem here.
1 reply 0 retweets 0 likes -
-
Replying to @GidMK
Take this study, for example (randomly Googled), comparing no. of obese patients in hospital with Covid to the obese population in 2014.https://www.thelancet.com/journals/landia/article/PIIS2213-8587(20)30160-1/fulltext …
1 reply 0 retweets 0 likes -
Replying to @cjsnowdon
Yes that is equally terrible. Basic ecological fallacy
1 reply 0 retweets 0 likes -
Replying to @GidMK
Fair enough. I’d say it was crude but potentially useful. I’m not sure how else you’d do a case-control with this but by treating the cases as cases and everyone else as the control.
1 reply 0 retweets 1 like -
Replying to @cjsnowdon
You'd draw from the same population and try to match on as many factors as possible. Case-control 101!
1 reply 0 retweets 0 likes -
Replying to @GidMK
How would that work in practice? How can you match for factors when it’s factors you’re trying to identify?
1 reply 0 retweets 0 likes
You match for basic things like age and gender and then connect enough information to correct for other factors in your analysis
-
-
Replying to @GidMK
Age and gender, fine (unless you want to see if older people and men are more likely to get it.) Quite a few of the smoking and obesity studies have done that, tbf.
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
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.