Friday, December 15, 2006


I have pointed out on many occasions that psychiatric research showing minimal to modest effects gets overhyped heavily by both drug companies and academic researchers. For example, see my post on injectable Abilify from a couple days ago. It made me think about a somewhat snarky letter to the editor by Bernard Carroll published in the American Journal of Psychiatry a couple years ago. It's worth reading both his letter (page 4) and the authors' response (page 5). Go here to check it out.

Remember folks, if a trial has a huge number of participants, a trivial difference between the drug and placebo will be labeled as statistically significant, but don't be fooled. We must also know the magnitude of the effect, and, as I've noted on many occasions (such as here, here and here), the size of the treatment effect is often fairly unimpressive.

I must include one quote from Carroll's letter: "The study has all the hallmarks of an “experimercial,” a cost-is-no-object exercise driven by a corporate sponsor to create positive publicity for its product in a market niche." This is reminiscent of the ARISE-RD (Risperdal as an add-on treatment for depression) study that I have written much about recently (here, here, and here), in which relatively minimal treatment effects were seen, yet the accompanying press release and, indeed, much of the discussion in the article itself, is overly optimistic regarding the treatment.


james gaulte said...

Thanks for introducing me to term "experimercial".One of the recurrent themes I have thought about and blogged about is the at times very sophisticated and not that easy to detect mis-use of evidence based medicine to cook the books of clincial trials and hype the results.

CL Psych said...

Thanks for your interest Dr. Gaulte. I also enjoyed looking around your site. I'll be back for more.