Publish or perish

Two doctors, Woodruff and Borst, we might describe them as followers of Hippocrates rather than drug company salespeople, are discussing the implications of published vs unpublished studies.

...

Woodruff: In 2006 the number of prescriptions written for antidepressant drugs was greater than for any other class of medication in the U.S. I should have gone into psychiatry.

Borst: And become a Prozac salesman?

Woodruff: It would help keep the wolf from the door.

Borst: Come on, Harry, you're an honest doc, you know the figures: 94% of published studies on antidepressant drugs report positive effects to taking the drug under investigation, but 96% of unpublished studies find questionable or no benefits to taking antidepressants.

Woodruff: I know, Bill. I’ve had patients come to me with depression and they say the medication doesn’t work. And I know that’s often the case. But sometimes I think part of the problem is with the patient. They give up taking the meds before they’ve had a chance to work. Get impatient to see results or don’t like the side effects.

Borst: The amount of negative data on drug use is remarkable, though. We know that drug companies like to see studies that show positive results and “lose” studies that show no results. They are biased.

Woodruff: Yeah, and there’s another worrying aspect to getting good data. Medical journals prefer to publish articles that demonstrate a treatment benefit. Could the drug companies be in collusion with the medical journals? Is there a huge reef of hidden data telling different stories about drug treatments that drug companies don't want the public to see?

...