Super-lame to quote tweet your own tweet an hour later, but there's already a lot of cool discussion happening under here, so come join the fray...https://twitter.com/ADAlthousePhD/status/1115658383175102468 …
-
-
Replying to @ADAlthousePhD
There is zero per cent chance this is a randomised experiment. Zero. There is no allocation concealment. Even if investigators don’t know which day is treatment and which is control, guessing would cause systematic allocation differences.
2 replies 0 retweets 12 likes -
Replying to @jd_wilko @ADAlthousePhD
Wait, why is concealment required for it to be an RCT?
2 replies 0 retweets 3 likes -
Replying to @venkmurthy @ADAlthousePhD
By ‘concealment’ I mean prospective concealment. Distinct from blinding during the study. If the investigator knows whether treatment A or B is the next allocation, they can decide whether or not they want to enrol the next patient now on that basis. That is selection.
3 replies 1 retweet 3 likes
And it's impossible to conceal the selection, because even if the initial patients' allocation is truly random, as soon as the first people are allocated the alternating schedule becomes obvious and the concealment is broken and investigators get input into allocation again
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.