OBJECTIVE: The authors studied the responses of drug-treated patients in an attempt to validate observations about abrupt and gradual improvements in patients receiving placebo. Since previous data suggested that in the first 2 weeks of antidepressant treatment specific drug effects are unlikely, the authors hypothesized that this improvement is a placebo effect. Therefore, in the first 2 weeks of antidepressant treatment abrupt and gradual improvements should have the characteristics of their placebo counterparts. METHOD: The subjects were 263 patients in controlled antidepressant trials lasting 6 weeks. RESULTS: The percentage of abrupt improvements that occurred in the first 2 weeks was higher than that for gradual improvements. Abrupt improvements during the first 2 weeks of drug treatment were also less persistent than gradual improvements with drug and no more persistent than improvements with placebo during the same period. However, in weeks 3, 4, and 5, abrupt and gradual improvements with drug were equally persistent and both were more persistent than abrupt improvements with placebo. CONCLUSIONS: These data support the authors' findings about placebo. Abrupt improvements during treatment with both drug and placebo are more likely during the first 2 weeks of treatment and are less likely to persist than gradual improvements. The fact that persistence of abrupt improvements with drug in weeks 1 and 2 appears different from that of gradual improvements but appears no different after week 3 suggests that the mechanism of action of abrupt improvement with drug changes after week 2.Abstract Teaser