To assess the feasibility and effectiveness of an evidence-based pharmacotherapy algorithm in the treatment of pediatric bipolar disorder. The study reports the results of a study of 64 bipolar type I subjects who were treated according to an algorithm developed in our specialty clinic. All subjects had been diagnosed using the Washington University in St. Louis Schedule for Affective Disorders and Schizophrenia. Subjects scored an average of 28 (+/- 4) on the baseline Young Mania Rating Scale. All subjects were assessed over an 18-month period. In addition, we were able to match 17 of the 64 subjects in the algorithm sample for gender, age, ethnicity, socioeconomic status, and diagnosis with an equal number of subjects in a psychopharmacology clinic who received treatment as usual. Prescribing clinicians were able to implement primary and secondary strategies, including detailed tactics of medication choices in the algorithm group. Growth curve analysis of the total algorithm group showed strong and significant improvement in symptoms. Analyses of the matched groups also showed strong effects for the treatment algorithm over treatment as usual. Treatment adherence and family satisfaction were higher in the algorithm group. An evidence-based, problem-solving pharmacotherapy algorithm is feasible and may be associated with better outcomes in the treatment of pediatric bipolar disorder. Randomized trials will be necessary to gather additional support for the algorithm's effectiveness. Copyright 2004 American Academy of Child and Adolescent Psychiatry