OBJECTIVE: The authors reviewed data involving the impact of providing psychotherapy for psychiatric disorders on costs of care. METHOD: In a search of the MEDLINE database limited to peer-reviewed papers published from 1984 through 1994, 686 articles were identified. Forty- one articles, covering 35 studies, were found in which the intervention tested was psychotherapeutic and the study included measures of outcome that had some implications for cost. The exclusion criteria for reviewing these studies included absence of a comparison group, a focus on medical disorders instead of psychiatric illnesses, and outcomes that did not include cost data or measures from which costs could be inferred. On this basis, 18 of the 35 studies were selected for analysis. The studies were categorized according to whether or not subjects were randomly assigned to study groups. Two reviewers independently read each study to identify the following characteristics: inclusion criteria, exclusion criteria, types of interventions, main outcome variables, sample size, and statistical tests for significant differences between treatments. Outcomes had to include actual cost accounting or data on medical care utilization or work functioning. RESULTS: The findings of eight (80%) of the 10 clinical trials with random assignment and all eight (100%) of the studies without random assignment suggested that psychotherapy reduces total costs. CONCLUSIONS: Psychotherapy appears to have a beneficial impact on a variety of costs when used in the treatment of the most severe psychiatric disorders, including schizophrenia, bipolar affective disorder, and borderline personality disorder. Much of that impact accrues from reductions in inpatient treatment and decreases in work impairment.