Frens I posted this thread at the weekend and it has gained some traction.
https://twitter.com/Arkancideisreal/status/1264032084944814082?s=20
If you want to get to the punchline quickly go to the end of the thread
(1) pretty much proven that the data is fake because there were more deaths in the study (only 5 Aus hospitals) than the whole of Australia, also the statistical technique used should reduce the number of patients available, but it didn't
(2) the authors posted another manuscript with the New England Journal of Medicine at the same time as the Lancet article, on a different subject but using the same data. Manuscripts like this take a huge amount of work and doing two in a couple of weeks without a huge team is impossible
(3) the authors are linked to (?owners of) Surgisphere - a "health informatics company" and the main author is a cardiothoracic surgeon who seems to be behind the venture. It appears that this is a health informatics artificial intelligence network software company.
The main author/Surgisphere CEO is supposedly a cardiothoracic surgeon whilst at the same time touring the world promoting his company.
https://event.icebergevents.com.au/whc2018/speakers/dr-sapan-desai-united-states-of-america
My (educated) guess is that they have access to health information and they are using this health information without permission to generate artificial datasets. It could be the biggest medical research fraud in history. Certainly two major journals are involved in publishing this stuff.
view the rest of the comments →
[–] Hoppinmad 0 points 4 points 4 points (+4|-0) ago
Wow. Great work.
This is how they cheat with the voting as well. They just assume that most will not look to closely at the fake data.