Man, the number of people responding to this who completely misunderstand statistical adjustment is both funny and terribly sad
-
-
I understand they're an art, not a science. If they're relying on other poor epidemiology, may not even get to art.
1 reply 0 retweets 2 likes -
Replying to @TuckerGoodrich @drvyom and
You misunderstand. And this study used fairly standard methodologyhttps://www.ncbi.nlm.nih.gov/pmc/articles/PMC4017459/ …
1 reply 0 retweets 1 like -
Adjusting for confounders is not perfect. You're adjusting bad data using other bad data.
1 reply 0 retweets 0 likes -
Replying to @TuckerGoodrich @drvyom and
Unless you are attacking their data collection, that is not accurate. The main issue with controlling for confounding variables is residual confounding, which is not what the tweets were about at all
2 replies 0 retweets 2 likes -
Where do they get the values to use to adjust the raw data?
1 reply 0 retweets 0 likes -
Replying to @TuckerGoodrich @drvyom and
A variety of places, it's all in the methods section
1 reply 0 retweets 0 likes -
Yes, and there seems to be an unfounded confidence in that data, and no, it's typically not listed in methods. They just mention that ut has been done. No raw data, no data for adjustments.
1 reply 0 retweets 0 likes -
Replying to @TuckerGoodrich @drvyom and
They briefly go over the data collection and reference their earlier publications where the data collection is more broadly described
1 reply 0 retweets 0 likes -
So no raw data.
1 reply 0 retweets 0 likes
Indeed. I imagine you could get it if you applied to the study authors and had a valid reason, but it wasn't published publicly Standard practice with potentially identifiable patient data such as this. It's usually an ethical requirement, in fact
-
Show additional replies, including those that may contain offensive content
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.