Nondepressed outpatients with a compulsive buying disorder were recruited by advertisement and word of mouth for inclusion in a controlled treatment trial. Following a 1-week single-blind placebo washout, subjects were randomly assigned to fluvoxamine (n = 12) or placebo (n = 11). Subjects received fluvoxamine (up to 300 mg daily) or placebo for 9 weeks. There were few dropouts. Outcome measures included the Yale-Brown Obsessive-Compulsive Scale--Shopping Version (YBOCS-SV), three Clinical Global Impressions (CGI) ratings, the Hamilton Rating Scale for Depression (HRSD), and the Maudsley Obsessive-compulsive Inventory (MOI). At the conclusion of the trial, 50% of fluvoxamine recipients and 63.6% of placebo recipients achieved CGI ratings of "much" or "very much" improvement, while 33% of fluvoxamine recipients were "very much" improved compared with 18% of placebo recipients (by endpoint analysis). Subjects in both treatment cells showed improvement as early as the second week of the trial, and for most, improvement continued during the 9-week study. There were no significant differences between fluvoxamine- and placebo-treated subjects on any of the outcome measures, with the exception that fluvoxamine recipients achieved greater improvement than placebo recipients on the MOI (p = .02). Adverse experiences were more frequent in the group receiving fluvoxamine, particularly nausea, insomnia, decreased motivation, and sedation. We conclude that in a short-term treatment trial of compulsive buying, subjects receiving fluvoxamine or placebo respond similarly.