Objective: Attention-deficit/hyperactivity disorder (ADHD) is typically treated over extended periods; however, few placebo-controlled, long-term studies of efficacy have been reported. Method: In a global multicenter study, children and adolescents who responded to an initial 12-week, open-label period of treatment with atomoxetine, a nonstimulant treatment for ADHD, were randomized to continued atomoxetine treatment or placebo for 9 months under double-blind conditions. Results: A total of 416 patients completed acute atomoxetine treatment and were randomized. At end point, atomoxetine was superior to placebo in preventing relapse defined as a return to 90% of baseline symptom severity (proportion relapsing: atomoxetine 65 of 292 [22.3%], placebo 47 of 124 [37.9%], p = .002). The proportion of patients with a 50% worsening in symptoms post-randomization was also lower on atomoxetine (atomoxetine 83 of 292 [28.4%], placebo 59 of 124 [47.6%], p < .001). Compared with patients in the placebo group, atomoxetine-treated patients had superior psychosocial functioning at end point. Discontinuations for adverse events were low in both groups, and tolerability was similar to that observed in acute treatment trials. Conclusions: In patients who responded favorably to 12 weeks of initial treatment, atomoxetine was superior to placebo in maintaining response for the ensuing 9 months. This result supports the value of maintenance treatment with atomoxetine in patients with ADHD who respond to initial treatment.