Objective: To carry out a metaanalysis of antidepressant studies to ca
lculate the effect sizes for antidepressant effect in depressive disor
der. Method: A metaanalysis of all antidepressant studies that include
d an active comparison drug as well as placebo was used to calculate t
he effect size. Articles were selected from a MEDLINE search for the p
eriod January 1966 to June 1995. Forty-nine studies were included in t
he metaanalysis. Results: The effect sizes for antidepressant treatmen
t are moderately larger than for placebo. A larger effect size was obs
erved in studies where objective diagnostic criteria for depression we
re used. Conclusions: We conclude that the superior efficacy of antide
pressants over placebo can be demonstrated.