Suicide rates in short-term randomized controlled trials of newer antidepressants.

Concerns have been raised about the appropriateness of placebo controls in clinical trials for major depressive disorder (MDD), given that there are approved treatments for this illness. Critics have argued that patients with untreated depression would be exposed to an unnecessary risk of suicide. There is also a competing concern that antidepressant drug treatment itself may induce suicidal behavior and thinking (suicidality). To examine this question, we have evaluated the rate of suicide in placebo- and active drug-treated groups of patients with MDD and various anxiety disorders participating in short-term randomized controlled trials (RCTs). We examined data from all manufacturer-sponsored short-term RCTs of 9 commonly used antidepressants in patients with MDD and various anxiety disorders. All short-term RCTs of antidepressants in patients with MDD and various anxiety disorders were included. Individual patients' data were available for all trials. Data were available for the 207 trials conducted in patients with MDD, including a total of 40,028 patients. There were 21 cases of suicide in these patients. Forty-four trials were conducted in patients with various anxiety disorders, including a total of 10,972 patients. There were 2 cases of suicide in these patients. Overall, at least 1 case of suicide occurred in 21 of the 251 trials. Sixteen of the suicides in MDD trials occurred in trials that had only an active control comparison group, and most of these (14 cases) were observed in the non-North American trials. In the placebo-controlled MDD trials, the rate ratios of suicide in the combined drug groups compared with placebo were 1.07 (0.1-63.4) and 0.5 (0.0-36.7) for the non-North American and North American trials, respectively. In the anxiety disorder studies, the overall rate ratio of suicide for the selective serotonin reuptake inhibitors compared with placebo was 0.9 (0.0-71.4). Neither use of placebo nor of antidepressants in short-term RCTs was associated with an increased risk of completed suicide among patients with MDD or various anxiety disorders. Nonetheless, because of the small numbers of suicides in these trials and the subsequent lack of statistical power, an increased risk of completed suicide in association with either drug or placebo treatment cannot be definitively excluded.