Dear Wiggo said:
And studies are only published when they have positive findings. *mutter*
eg:
https://www.ted.com/talks/ben_goldacre_what_doctors_don_t_know_about_the_drugs_they_prescribe
(TED can be dodgy as f*ck, so happy to be corrected on the talk's content)
Can't watch the talk, I know of Ben though and have read most of what he writes so I'm kind of going off that.
He's completely correct in saying that negative data does not get published very often and that it should be. Of course making sense of negative data is much harder, just saying to doesn't work isn't good enough because, especially with drugs, you've already shown they have the desired effect in vitro, model organisms etc. so it can take a lot of investigation, some of which isn't possible anyway, to work out what is going on and for the publication to actually be beneficial.
In terms of drug trials I'm in full agreement that all of the data should be published regardless and so are most drug companies to be fair, it's just an old convention that started it and stops it happening.
Where I have a problem with what Ben talks about is that the people who really matter when making the decisions about what gets approved and what effects must be listed form the studies do get all of the information (at least it is a legal requirement to submit it all, companies breaking the rules is speculation but worth noting). In the UK that is the MHRA and they have several expert panels who assess all of the data submitted and decide on approval, packaging etc. It is all regulated and the information doctors require is there.
Most doctors who are not specialists have no idea what they are prescribing anyway. That's not a criticism, they can't be expected to know what every drug is, how it interacts with things and whether it is worse for a certain sub-set of a population than another, but the idea every doctor is going to research what they prescribe to everyone is insane, you could only ever expect a specialist to do that and even then most probably wont.
In an ideal world a doctor would diagnose and a pharmacist, who does have this knowledge or at least a much more in depth knowledge than the physician, would prescribe, but getting the prescription pads out of the hands of doctors isn't going to happen.
So yes, he's correct. But he's also making out like doctors have both the time and ability to make sense of the data and many of them do not.
He also attacks the MHRA over things that are completely beyond their control but I'll leave that for a different discussion.