OBJECTIVES: The occurrence of dyslipidemia, glucose intolerance and weight gain may lead to an increased risk for developing Diabetes and coronary heart disease (CHD) during therapy with atypical antipsychotics. These complications directly influence an antipsychotic agent's cost-effectiveness. One method for outcome assessment is to capture these effects with the combined endpoint of metabolic syndrome (MetS) and, using epidemiological risk prediction methods, estimate a predicted risk of diabetes and CHD. The economic evaluation thus captured patient-relevant outcomes of MetS or diabetes/CHD events avoided. METHODS: MetS was defined according to NCEP (ATPIII) guidelines, with modifications reflecting clinically meaningful changes in risk factors. Lab measures were derived from a randomised double-blind, head-to-head trial. The cumulative incidence of MetS was computed using Kaplan-Meier analysis and compared across treatment arms for 6 months of follow-up. The 5-year diabetes and CHD risks were estimated from each individual patient's risk factor profile for diabetes and CHD using a published logistic regression and Weibull model, respectively. Resource utilisation and costs were derived from the underlying trial and published data. Cost analysis was based on a Swedish third party payer perspective. RESULTS: The results showed a 65% relative risk reduction (RRR) in MetS incidence at 6 months, a 30% RRR in diabetes and 22% RRR in CHD incidence at 5 years for Aripiprazole versus Olanzapine. For a hypothetical cohort of 1000 patients switched from Olanzapine to Aripiprazole, at study endpoint 124 events of MetS, 37 events of diabetes and 6 events of CHD are avoided, showing cost savings due to reduced medical treatment of side-effects of SEK 101.360 for MetS, 3.035.000 SEK for Diabetes and 880.000 SEK for CHD. CONCLUSIONS: The results highlight a medical and economic benefit from maintenance therapy of Aripiprazole versus Olanzapine, reflected by lower costs and a reduced incidence of MetS, Diabetes and CHD.