Objective: To compare societal costs between patients treated with atomoxetine and placebo in Sweden. Method: Ninety-nine pediatric ADHD patients were randomized to a 10-week double-blind treatment with atomoxetine (n = 49) or placebo (n = 50). All parents received four sessions of psycho-education. Parents filled out a resource utilization questionnaire covering the 10 weeks prior to treatment and the 10-week on-treatment period. Published unit costs/prices were used to calculate costs. Results: Mean on-treatment costs in the atomoxetine group (SEK [Swedish Krona] 4,558) were significantly lower compared with placebo (SEK 7,684) after adjusting for baseline costs and site (p = 0.007). All 99 patients entered an open atomoxetine extension phase. Both groups had numerical reductions in direct and indirect costs while on atomoxetine treatment during the extension phase. The atomoxetine medication costs were offset by the reductions in direct nonmedical and indirect costs. Conclusions: These data provide preliminary evidence that atomoxetine together with parental psycho education reduces nonmedication costs associated with ADHD in Sweden. (Contains 2 tables and 2 figures.)