Fear of increased risk of attempted suicide in the placebo groups should not be an argument against performing short-term and long-term, placebo-controlled trials in major depression.

The original data from the placebo-arms and the tricyclic-arms of all parallel randomized controlled three-arm studies, which had been conducted in the period 1979-1991 for a drug under development in order to obtain marketing authorization for the indication major depression, were included in a meta-analysis. Thirty-two placebo-controlled studies including 4314 patients were analyzed. The intention to treat analysis resulted in 46% responders (at least 50% improvement on the Hamilton Depression Rating Scale) in the tricyclic antidepressant group and 31% in the placebo-group (CI 11.5-17.1%). The number needed to treat for responders was 7 95%-difference . In 10 out of 32 studies, a statistically significant difference in favor of tricyclic antidepressant compared to placebo was 95% found for responders. The responder rate in the placebo-group varied from 6 to 52%. We conclude that tricyclic antidepressants are efficacious in the short-term treatment of major depression. However, the magnitude of the effect is rather modest. Because 69% of the placebo-controlled studies with a tricyclic antidepressant did not show a statistically significant difference in favor of tricyclic antidepressant and the placebo rate varied considerably from study to study, equivalence studies with tricyclic antidepressant as comparator without a placebo-control are not sufficient for demonstrating efficacy. Therefore in major depression, placebo-controlled studies are still necessary to demonstrate efficacy.

Tricyclic antidepressant response is independent of gender.

The results indicate that lithium is an effective drug in the treatment of moderate to severe manic episode. The variability in placebo response indicates that a placebo control arm in efficacy studies among patients with moderate to severe manic episode is necessary.

No abstract