Recall that in October of 2003, the FDA issued a public health advisory about the possibility that antidepressants cause suicidal ideation in children and adolescents. A year later, the agency asked drug-makers to add a black box warning about this issue to their package inserts. In an article in the American Journal of Psychiatry, researchers reported that suicide rates increased by 14% in 2004, the largest such increase since the Center for Disease Control (CDC) began tracking this in 1979. The authors conclude that the black box warning not only failed, but may have had the opposite effect from what was intended (Gibbons RD et al., Am J Psychiatry 2007;164:1356-1363).

TCPR’s Take: Unfortunately, the study reported the data in a confusing manner, leading many to conclude from the paper that a drop in SSRI prescriptions was corre- lated with this rise in suicides. In fact, there was no drop in SSRI prescriptions in 2004; that drop did not occur until 2005, and the paper did not report suicide rates for that year. As it turns out, soon after the article was published, the CDC released preliminary suicide figures for 2005, showing that the overall population suicide rate actually decreased by about 3% (they have not yet released age-specific figures). This has become something of an embarrassment for the American Journal, because it took the New York Times to clarify the meaning of the data – something the journal should have required of the authors during the peer review process. Compounding the controversy, three of the authors have significant conflicts of interest, most conspicuously the first author, who has testified in product liability trials for Wyeth Pharmaceuticals (i.e., he makes money to argue that anti- depressants do not cause harm). At any rate, it is certainly too early to determine whether the black box warning “failed”; we have to wait for the CDC to provide its final 2005 numbers, which may not occur for several months.