Objective Fluoxetine and its active metabolite norfluoxetine have long half-lives. We postulate that, owing to the long elimination half-life and the time to reach steady-state level in plasma is nearly four weeks, patients diagnosed with major depressive disorder might be treated with fluoxetine taken once every third day, after being treated initially during 4 weeks with daily doses of fluoxetine. Methods In this open label, 12-weeks, randomized, prospective study, patients diagnosed with DSM-IV major depressive disorder were randomly assigned into 1 of 3 treatment groups. Thirty-four patients took 20 mg and 32 patients took 40 mg of fluoxetine daily throughout the study. Twenty-nine patients had been taking 20 mg of fluoxetine daily for 4 weeks of the study initially, and then were switched to 20 mg fluoxetine once every third day regime. The severity of depression was assessed by Hamilton Depression Rating Scale (HDRS) and Clinical Global Impressions-Severity Scale (CGI-S). Response was defined as a 50% or greater reduction of the baseline HDRS total score. After defining a strict criterion of relapse, time to relapse was estimated using survival analyses (Kaplan-Meier method). Results The repeated measures analysis of variance (ANOVA) of HDRS found a significant time effect (F = 464.04, df = 1.00, p < 0.001), but no significant group effects (F = 0.84, df = 2.00, p = 0.433) from baseline through week 12. The proportion of responders was not significantly different between the treatment groups at the endpoint. Survival analyses showed, however, a Significant delayed mean time to relapse in patients treated with 40 mg of fluoxetine daily compared to either patients treated with 20 mg of fluoxetine daily or 20 mg fluoxetine once every third day. The mean times to relapse were 79.8, 70.8, and 70.5 days, respectively. Fluoxetine was associated with some adverse events in 46.3% of patients. The most frequently occurring adverse event was insomnia. Conclusion It is proposed that either every third day or daily dosing with the same dose of fluoxetine could treat the patients with major depressive disorder during the acute and continuation period of treatment. Nevertheless, higher daily dose of fluoxetine has a reduced relapse rate compared to that of the lower daily dose.