I have a new article in the New York Times discussing lack of randomized trials for masks for source-control for COVID. In my newsletter, I discuss what this debate/debacle reveals, not just about COVID denialism, but also about our public health response.https://zeynep.substack.com/p/on-randomized-trials-and-medicine …
-
-
But taken at face value, it would a great argument for increasing the quality of our own masks and for universal masking. (Because it would say cloth masks with poor compliance aren't enough to protect the wearer). Unfortunately, as authors note, too underpowered for conclusion.pic.twitter.com/ITbAJZ93ym
Show this thread -
Oh, man, it's going this now, no? No no having a confidence interval that large doesn't show "mask usage may increase the likelihood of infection." It says the study was statistically underpowered so the measurements couldn't be narrowly pinpointed. It's a CONFIDENCE INTERVAL.pic.twitter.com/N8vxKz2ABH
Show this thread -
And seriously, the reporting on this is, predictably terrible. This below is the NYT. Why this weird "both sides" framing? No, it's not just the critics that are noting these limitations. The study itself is well-written and IT'S THE AUTHORS THEMSELVES NOTING THESE LIMITATIONS.pic.twitter.com/UfAIDDngDm
Show this thread -
Look, the Danish study is weak and statistically underpowered. Too short, only measures impact of "recommending" rather than wearing masks, testing method has a false positive issue plus not measuring source-control. Okay, though. This is described in the paper. Not much to see.
Show this thread
End of conversation
New conversation -
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
It was well written, mostly. It's going to get widely misrepresented, but the authors did a fair representation of what they had which was unfortunately not that much, mostly for structural reasons. Hard topic to do a RCT on! I'm glad it was published to be honest.
End of conversation
-
-
-
Let us know if you find good fact-checks of the study citing these flaws. False conclusions will be shared on FB. It’s so helpful to have short, easy-to-comprehend explainers for everyday people who struggle w/math/stats. (I did share the good NYT Gina Kolata piece already.)
-
The study itself! There will be a lot of misinformation attempts based on the study, but it's open source and well-written, along with two editorials.
- Show replies
New conversation -
-
-
Thanks. 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.