(THREAD) misleading results from subgroup analysis combined with regression to the mean in crossover RCT’s. Pull up a chair, let’s have a chat.pic.twitter.com/jA86IV1CRD
-
Show this thread
-
Credit to
@Srpatelmd for spotting this:https://www.ncbi.nlm.nih.gov/pubmed/303954861 reply 0 retweets 4 likesShow this thread -
This is a crossover RCT of 20 sleep apnea patients comparing one night of atomoxetine 80mg plus oxybutynin 5mg (ato-oxy) versus placebo administered prior to sleep
1 reply 0 retweets 0 likesShow this thread -
Primary outcome: apnea hypopnea index (AHI), continuous variable. Higher AHI = more severe sleep apnea.
1 reply 0 retweets 0 likesShow this thread -
In an effort to show that ato-oxy was *especially* effective in the patients with more severe sleep apnea, the authors included the following subgroup analysis:
1 reply 0 retweets 0 likesShow this thread -
“When analysis was limited to the 15 patients with OSA (AHI≥10 events/h) on placebo, ato-oxy reduced AHI by approximately 28 events/h or 74%”
2 replies 0 retweets 0 likesShow this thread -
Replying to @ADAlthousePhD
I'm confused. Unless I'm mistaken, that's not the definition of OSA. Could this perhaps be a bit of selective data slicing?pic.twitter.com/JLorPwQRUk
1 reply 0 retweets 0 likes -
Replying to @GidMK
I don’t think it’s quite that nefarious. I just don’t think the authors really get why what they did introduces such a bias.
1 reply 0 retweets 0 likes
Fair enough. It's just a weird definition of OSA as far as I'm aware, usually you'd go with either 5 or 15
-
-
Replying to @GidMK
The overall results of the trial were positive before even doing the subgroup analysis. There’s enough here to move forward to larger confirmatory trial. I’m just sighing that nobody pointed out how dumb (and unnecessary!) this subgroup analysis was/is.
0 replies 0 retweets 1 likeThanks. 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.