I took a deep dive into the data for ivermectin for covid-19. Promising data, made less promising by a series of red flags in the key studies.
If ever there was a need for open data, this is it.
https://www.medscape.com/viewarticle/954681?src=soc_tw_share … via @medscape
-
Show this thread
-
2. The meta-analysis I discuss shows promising results... but it includes a bunch of pre-prints. There's nothing wrong with that, but it means some of the studies haven't undergone peer review...pic.twitter.com/fTp5WWFrep
2 replies 1 retweet 15 likesShow this thread -
3. And when I checked out two of the trials (Niaee et al from Iran and Elgazaar et al from Egypt) that drive the results, I found some red flags...
2 replies 1 retweet 13 likesShow this thread -
4. The Iran trial has a surprisingly high number of patients who, despite being sick with COVID are PCR negative... What's worse, they are differentially PCR negative between the treatment and control groups. (P-value calculated by me, binomial). Not c/w good randomziation.pic.twitter.com/j8MRGDhQ7y
3 replies 3 retweets 14 likesShow this thread -
5. The Egypt trial, another one with big beneficial effects, has multiple statistically significant differences in baseline factors between treatment and control groups (p-values again calculated by me, t-test).pic.twitter.com/zXPFjIIF1s
3 replies 1 retweet 11 likesShow this thread -
6. I'm not saying these trials are fake, or aren't really randomized, but peer review would give these authors a chance to address these concerns, figure out what's up, re-analyze their data, etc.
1 reply 1 retweet 12 likesShow this thread -
7. Peer review takes time though, so I am strongly encouraging the authors of these studies to release their full, deidentified datasets to a neutral service like http://datadryad.com so we can all take a look and see.
3 replies 1 retweet 14 likesShow this thread -
8. Open data fixes problems like this. Without it, we literally don't know what to believe. And without good science, we are adrift. /rant
1 reply 1 retweet 12 likesShow this thread -
PS. shoutout to
@GidMK who has already flagged some issues with the Iran trial.2 replies 1 retweet 12 likesShow this thread
Health Nerd Retweeted Health Nerd
Turns out the issues may be bigger than we first thoughthttps://twitter.com/GidMK/status/1415764372362649601?s=20 …
Health Nerd added,
-
-
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.