Briefly, a drug company manipulated a study to claim that the anti-depressant Paxil was good for children, and then made $11 billion in sales. It later paid $3B to settle govt claims about fraudulent research.
A p-value of 0.11 was moved to less than the required 0.05 by restructuring the data.
He ends by saying that the journal should retract the publication. There must be a better solution. The study data were not faked. They were just artfully presented, with some relevant data omitted. This was considered acceptable at the time. Possibly someone could figure that out by reading the study.
We need higher standards for medical study publications.
In the meantime, you should know that the research behind psychiatric drugs is questionable.
No comments:
Post a Comment