Publication bias in antidepressant trials

Erick Turner was a former US Food and Drug Administration (FDA) reviewer. At the FDA he was aware of publication bias due to negative results of trials tending not to be published. As he says in an interview with Ayurdhi Dhar from Mad in America (see edited transcript and podcast), there was a “disconnect between what clinicians were seeing and what the FDA reviewers were seeing, and what was known to the FDA and the pharmaceutical industry”. 

I’ve always argued that the apparent small difference between antidepressant and placebo in clinical trials could be due to placebo amplification because of unblinding (see eg. previous post). Although maybe unlikely, I suppose it’s possible that publication bias could completely explain this artefact. Certainly the pharmaceutical industry has taken advantage of the clinical reluctance to consider this issue, although publication bias also plagues psychotherapy trials. As Erick says, the decision as to whether a clinical trial is “good science shouldn’t depend upon whether the results were statistically significant”.