For the latest updates on outcome switching at the COMPare project check out the host of blogs currently being posted.
These blogs provide an in depth analysis of the outcome switching problem and the latest responses on the COMPare site to editors and trialists responses – or in many cases non responses – to the 58 letters the COMPare team have sent.
On the 1st February, Carl Heneghan (ie. me) published on what the ICJME have to say on outcome switching. It’s worth checking out their recommendations. I am still not sure if many of the editors understand the implications of requiring registries and the impact this should have on the minimum trial data set required at the outset of the trial. Yes, you guessed it, this requires the primary and secondary outcomes to be detailed.
On the 29th January, Henry Drysdale blogged on the Post-hoc “pre-specification” and undeclared deferral of results: a broken record in the making. Referring to responses from authors and Annals’ editors on COMPare’s analysis of the ATLAS trial. ‘A great example of the problems we have uncovered,’ said Drysdale.
Ben Goldacre also blogged on the 20th of January on ‘Where does Annals of Internal Medicine stand on outcome switching? This is a lengthy detailed response – one that you should definitely read – that points out the majorly important methodological issues at the heart of the problems in outcome switching: ‘The Annals editors’ last point is extremely odd……..read on’
And if you missed it, BuzzFeed ran an excellent piece highlighting researchers are “choosing their lottery numbers after seeing the draw”, making medicine less reliable – and respected journals are letting them do it.