1/4 Appreciate the argument this piece has stimulated even if I think we all probably agree more than disagree. My mildly tangential, but I think important, 2 cents after reading the piece...authors argue RCT can determine causality given 3 criteria:https://twitter.com/ADAlthousePhD/status/1088056397349179393 …
-
Show this thread
-
2/4 1) large enough sample size; 2) ITT analysis; and 3) near-complete follow-up. Cent number 1: ITT analysis of an RCT with an active comparator and non-adherence can be biased toward or away from the null (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3731071/ …) and
2 replies 1 retweet 8 likesShow this thread
Replying to @leskocar
I guess proper randomization, equal dropout, and other selection bias control were assumed then 
12:07 PM - 24 Jan 2019
0 replies
0 retweets
1 like
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.