Also, a very simple question that HAS to be answered for the analysis to mean anything: what proportion of people in each country actually got HCQ?
Simple question. Without it, the HCQ vs non-HCQ groups mean nothing 
-
-
This Tweet is unavailable.
-
Replying to @c19analysis
Lol, no you didn't. You cannot answer it, which is yet another fundamental issue. If you don't know whether someone in Algeria was more likely to get HCQ than France - and CAN PUT A NUMBER TO THAT LIKELIHOOD - then the treatment groups are meaningless
3 replies 0 retweets 1 like -
This Tweet is unavailable.
-
Replying to @c19analysis
You haven't answered this ever. All you say is that imperfect adherence is common in RCTs, which is not an answer. This isn't an RCT, and it's not about adherence
1 reply 0 retweets 0 likes -
This Tweet is unavailable.
-
Replying to @c19analysis
You didn't answer this, you just gave a non-response. Presumably you have no idea how many people in any country actually received HCQ, which as I said completely undermines the analysis
1 reply 0 retweets 0 likes -
This Tweet is unavailable.
-
Replying to @c19analysis
I feel like I'm in a Monty Python skit. I've answered them, and you still haven't even come close to addressing mine!
1 reply 0 retweets 1 like -
This Tweet is unavailable.
Answered: - no study is missing, just misclassified - early treatment unlikely to have benefit based on robust evidence - cbf re-reading your references to find the study you mean - any treatment 'may' be effective, this one appears unlikely to be
-
-
Replying to @GidMK @c19analysis
Now, what proportion of your intervention group actually received HCQ? Simple question!
1 reply 0 retweets 1 like - Show replies
New conversation -
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.