A large, well-done observational trial just found a big mortality benefit for ondansetron in the treatment of COVID-19 I think this is a fascinating case study in how we use observational researchhttps://academic.oup.com/ofid/article/8/7/ofab336/6320906?searchresult=1#275722395 …
-
-
Thing is, we've been using studies a LOT worse than this one to drive treatments for the whole pandemic. The initial use of ivermectin, HCQ etc in 2020 was based on much less robust evidence than this research
Show this thread -
Why is this interesting? I think it shows just how important RCTs of interventions for COVID-19 are If you test a bunch of drugs, you'll see some positives
Show this thread -
If 1,000 people run studies on their databases, and you only really see the positive results published, you'll get the false impression that these drugs work against coronavirus even when the evidence is not good at all
Show this thread -
This is in no way a critique of this specific paper, whose authors appear to have done a fantastic job, but it is a pretty worrying note for the rest of us
Show this thread -
Anyway, ondansetron is a cheap, low-side-effect drug available pretty much everywhere in the world, so if it works for COVID-19 it'd be a brilliant addition to the arsenal
Show this thread
End of conversation
New conversation -
-
-
Paging Drs Benjamini and Hochberg
-
In this context, I don't think it's necessary to run that sort of correction - they are very open about this being essentially a fishing expedition - but yeh it's interesting how close to the mathematical false positive rate 6/84 is
- Show replies
New conversation -
-
-
But isn't there a statistical problem with this kind of study design?
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.