PREDICTION OF OUTCOME DURING FLUOXETINE CONTINUATION

Massachusetts General Hospital, Boston MA

Investigators

Linked publications & trials

Abstract

About half of the patients who respond to antidepressants improve as a result of the natural course of the illness or due to nonspecific or "placebo" effects rather than pharmacologic effects. Since current practice dictates drug maintenance for all responders to antidepressants, much continuation pharmacotherapy may be unnecessary. If "placebo" responses to an active drug could be identified, unnecessary medication could be discontinued, decreasing morbidity and cost knowing that initial improvement is attributable to a "placebo" effect would also permit more rational treatment of relapses during continuation treatment. Multiple studies have shown that the analysis of patterns of response to acute antidepressant treatment differentiate "true" or pharmacologic responses from "placebo' or nonspecific responses. Preliminary data presented in this proposal suggest that pattern analysis also distinguishes patients who require medication during continuation from those who do not. In this study, the predictive value of the acute response pattern will be tested prospectively using a placebo-controlled discontinuation design. Success in this would allow clinically useful recommendations for continuation treatment to be made based on the collection of relatively simple and easily acquired clinical data. The present study involves the enrollment of 700 patients with major depression at two sites. Responders to a 12-week acute trial of fluoxetine will be prospectively randomized for 24 weeks of double-blind continuation treatment based on their pattern of response ("placebo" or "true drugs type). In each group, 50% of the patients will be assigned to remain on fluoxetine and 50% will switch to placebo. Differences in relapse rates will be examined in the four groups.