How big is the placebo effect in depression?

The evidentiary landscape regarding antidepressant efficacy seems to grow ever more slippery. Now comes a study, drawn to my attention by the busy-eyed Philip Dawdy at Furious Seasons, that finds that the beneficial effects of placebo treatment of depression last longer than generally thought.

As the study's authors note, "The assumption that the placebo response in depression does not endure is widely held and often stated in writing." In particular, many seem to assume that placebo effects fade while effects of actual medications persist -- another argument for antidepressants.

The point here is not that antidepressants never work. It's that the strength and breadth of the placebo is so strong -- placebo helped 80% as many patients in this study as did real antidepressants -- that it greatly complicates evaluating real antidepressants at either the individual or population level.

I've not had time to closely the study in question, which was authored by a team dran from Duke, Brown, and Tufts. As the authors note, their meta-analysis drew on a fairly large patient base -- 3063 patients -- but included only 8 studies, apparently because they could only find that many that ran the trials long enough to test the question whether placebos worked long-term. (In this case, long-term means 12 weeks. Most antidepressant trials run pretty short, which is another of their many significant limitations.) Even with those caveats in mind, however, this seems a pretty startling and significant finding.

Furious Seasons's take:

Via Furious Seasons

Placebo Effect In Depression Treatment Much Larger Than Previously Thought

CL Psych made me aware of an explosive study in the August Journal of Psychiatric Research which contends that the placebo effect in anti-depressants is much larger than I think anyone in the research world expected. It's a meta study re-analyzing eight anti-depressant trials comprising 3,063 people diagnosed with depression. The study was done by Arif Kahn et al. Kahn is well known in the research world and runs a large clinical research facility in the Seattle area. So he's a long way from being an anti-meds advocate.

In the study, which looked at trials that went longer than 12 weeks (some went as long as 12 months), Kahn found that 79 percent of patients on placebo remained well compared to 93 percent of anti-depressant responders. That would give an overall effect size of the anti-depressants studied of 14 percent, well under the usual 25 percent to 30 percent in shorter anti-depressant trials. I cannot tell from the abstract what specific anti-depressants were involved, but for a sugar pill to perform nearly as well over time is astonishing. Eye-opening even.

Dawdy at Furious Seasons further notes:

In his study, Kahn offered this assessment:

"The widely held – and probably erroneous – belief that the placebo response in depression is short-lived appears to be based largely on intuition and perhaps wishful thinking."

That was the study's final sentence. It'll be interesting to see if this gets replicated.