Intended for healthcare professionals

Rapid response to:

Head To Head

Are antidepressants overprescribed? Yes

BMJ 2013; 346 doi: (Published 22 January 2013) Cite this as: BMJ 2013;346:f191

Rapid Response:

Re: Are antidepressants overprescribed? Yes

In a recent analysis (Preda 2012) we discussed how one the most cited studies about the lack of difference between placebo and antidepressant (AD) response is actually not supported by the study findings.

The study under discussion is Kirsch et al. 2008. The non-spinned conclusion of Kirsch et al. 2008, as formulated by its authors was that:

“Drug–placebo differences in antidepressant efficacy increase as a function of baseline severity, but are relatively small even for severely depressed patients” (Kirsch et al., 2008)

In other words Kirsch et al. stated in non-ambiguous terms that ADs are statistically better than placebo and also that the response rates correlated with depressive symptoms severity.

As we discussed in our article this conclusion was not big news to the scientific community. In their 2002 meta-analysys of 45 studies Kahn et al. reported that AD effect correlated with the depressive scores severity at the beginning of the study.

For reasons that had little to do with science and lots to do with language (mis)use and media newsworthy selection bias the Kirsch et al. 2008 study became “the poster study of the anti-antidepressant movement” (Preda 2012), indeed a paradox considering that the study only replicated already known positive findings confirming an AD effect.

We recognize that the antidepressant – placebo literature can be confusing. We actually think that depression is not a singular case here – the history of medicine is one of trials and tribulations of diagnoses and treatments that are building momentum or loosing grounds based on cumulative evidence.

The issue with depression is that we are looking at a diagnosis complicated by a lack of hard biological markers which is used as a battle ground for experts who appear to be (over)committed to their particular version of the story (ADs do not work or they work for sure, no doubt about it).

Non-ambiguous, definitive expert conclusions on a major public health issue are then further amplified by uncritical media outlets and trumpeted to public that has its own opinion on what mental illness might be (very different than how the public thinks about other “hard” medical issues that the public is comfortable acknowledging its lack of expertise such as Ebola, GERD, cardiac ischemia etc.)

We then face a snow-balling effect that further inflames an already uninformative discourse.

A careful reading of the literature clearly makes the case that there is too much feeling and too little objectivity when it comes to taking positions on the antidepressant prescription issue.

Over or under prescribed? The many different answers to this question might be only because the question is too simply broad and non-specific.

Chance is that there are cases when either scenario is true as there are cases when either scenario is wrong.

To properly answer this question we might need to move into more nuanced questions about specific diagnostic criteria, populations, treatment duration, response and remission.

Further, we encourage the experts to always nuance the context of their findings and emphasize the ambiguity and relativity of the scientific discovery.

When it comes to such an important issue as depression this is not only a requirement for good science but also, and most importantly, the ethical thing to do.


Khan A, Leventhal RM, Khan SR, Brown WA (2002) Severity of depression and response to antidepressants and placebo: an analysis of the Food and Drug Administration database. J Clin Psychopharmacol 22: 40–45.

Kirsch I, Deacon BJ, Huedo-Medina TB, Scoboria A, Moore TJ, Johnson BT (2008) Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med 5: e45–e45.

Preda 2012: The Antidepressant Wars, a Sequel: How the Media Distort Findings and Do Harm to Patients, December 26, 2012, Mind the Brain, PLOS Blogs

Competing interests: No competing interests

29 January 2013
Adrian Preda
Professor of Psychiatry
University of California Irvine
101 The City Drive S.