Pattern analysis shows beneficial effect of fluoxetine treatment in mild depression

In a multicentered study, 372 patients with mild major depressive disorder with a Hamilton Rating Scale for Depression (HAM-D) score of 15 to 19 were randomly assigned to 6 weeks of treatment with placebo or 20 mg, 40 mg, or 60 mg/day of fluoxetine. Patients were rated weekly for improvement and the appearance of side effects. Pattern analysis of treatment response showed more patients in the active drug treatment groups having a persistent or a delayed persistent response, the types of response specifically associated with active treatment. Analyses of mean changes in treatment measures showed little difference among treatment groups. This may be explained in part by different distributions in outcome, placebo patients having had a higher frequency of mild improvement with fewer negative and very positive outcomes. Response rate analyses favor the active treatments numerically, but only one of the comparisons is statistically significant. These findings suggest a specific role for fluoxetine treatment in mildly depressed patients who do not respond promptly or who respond inconsistently to nonspecific treatment.