A double-blind, placebo-controlled trial was undertaken to compare the safety and efficacy of venlafaxine and trazodone in patients with major depression. Two hundred twenty-five patients entered an initial 6-week treatment phase, and 149 completed it. Ninety-six patients who were responders continued in a 1-year, double-blind, long-term phase during which they received the same medication and doses they had during the short-term phase. Both active treatments were significantly more effective than placebo on some measures during the short-term study, but venlafaxine produced more improvement in the cognitive disturbance and retardation factors on the Hamilton Rating Scale for Depression. Trazodone was more effective against the sleep disturbance factor. Patients on venlafaxine were most likely to enter the long-term phase and to remain in the trial longest. The side effect profiles of the three treatment groups were compared. Venlafaxine was most likely to cause nausea, whereas trazodone was associated with the most dizziness and somnolence.