The objective of this double-blind, placebo-controlled, multicentre randomized withdrawal study was to evaluate the efficacy and tolerability of sertraline for 16 weeks in treating Japanese patients with major depressive disorder (MDD) who had achieved a response to 8 weeks of sertraline treatment. Patients (n=361) were initially treated with 8 weeks of open-label sertraline treatment followed by 16 weeks of double-blind treatment with either sertraline (50-100 mg/day) or placebo. Responders during the open-label phase were eligible to be entered into the double-blind phase. A total of 235 patients (65.1%) were entered to the double-blind phase and randomly assigned to receive sertraline (n=117) or placebo (n=118). A significantly (P=0.016) lower relapse rate was found for sertraline (8.5%) compared to placebo (19.5%) during the double-blind phase. Examination of time-to-relapse showed that the relapse free rate curve was significantly higher for sertraline (log-rank test, P=0.026) than placebo. Mean changes from beginning to end of the double-blind phase on measure of depressive symptoms, quality of life and global improvement also significantly favoured sertraline over placebo. Sertraline was well-tolerated, with similar adverse events as found in previous studies. These results confirm the efficacy of sertraline in preventing the relapse of MDD in Japanese patients.