Objective: The efficacy of atomoxetine as treatment of symptoms of attention-deficit/hyperactivity disorder (ADHD) in patients with autism spectrum disorder (ASD) has not been established. Method: In this study, 97 patients aged 6 to 17 years with ADHD and ASD were randomly assigned to double-blind treatment with 1.2 mg/kg/day atomoxetine or placebo for 8 weeks. The primary endpoint was the ADHD Rating Scale (ADHD-RS) score; secondary endpoints were the Clinical Global Impression of ADHD-Improvement (CGI-I) and the Conners Teacher Rating Scale-Revised: Short Form (CTRS-R:S) score. Results: Baseline mean ADHD-RS scores for atomoxetine versus placebo were 40.7 and 38.6; after 8 weeks, mixed-effect model repeated-measure means were 31.6 (95% confidence interval 29.2-33.9) and 38.3 (36.0-40.6), respectively, with a difference in least square means of -6.7 (-10.0 to -3.4; p less than 0.001). The CTRS-R:S Hyperactivity subscore also improved significantly for atomoxetine compared with placebo, but not the other CTRS-R:S subscores. However, there were not significantly more patients on atomoxetine (20.9%) who improved much, or very much according to the CGI-I, than on placebo (8.7%; p = 0.14). Adverse events (mostly nausea, decrease in appetite, fatigue, and early morning awakening) were reported in 81.3% of atomoxetine patients and 65.3% of placebo patients (p greater than 0.1). There were no serious adverse events. Conclusions: Atomoxetine moderately improved ADHD symptoms in patients with ASD and was generally well tolerated. Adverse events in this study were similar to those in other studies with ADHD patients without ASD. (Contains 4 tables and 1 figure.)