OBJECTIVES: To assess the relative cost-effectiveness of olanzapine compared to risperidone in the treatment of acute episodes and long-term maintenance of schizophrenia in Italy. METHODS: A decision analysis approach (based on semi-markov modelling techniques) was used to consider the costs and health outcomes of initiating treatment on patients with an acute episode of schizophrenia and a history of relapsing with no prior atypical antipsychotic treatment. The model allowed two alternative atypical antipsychotics to be used before considering patients as being treatment resistant. Clinical response rates were based on changes in BPRS/PANSS scores (>=40% improvement) taken from randomised clinical trials. Relapse rates for the first year of treatment were based on literature estimates. Resource use data covering periods in acute episode, stable maintenance and acute relapse health states were based on a combination of published data and clinical opinion. The model was used to compare the costs and health outcomes of olanzapine versus risperidone as 1st line treatment choices over a 3-year period. RESULTS: The base case analysis showed that 1st line use of olanzapine resulted in a reduction of relapses over the 3-year period (37 versus 38 per 100 patients treated). The olanzapine 1st line strategy was associated with an overall cost saving over risperidone of around €50,822 per 100 patients treated (approximately a 2% cost reduction) despite the increased drug costs (€6.64 versus €4.19 per day). CONCLUSIONS: With the context of the Italian health care services the use of olanzapine as a 1st line atypical antipsychotic appears to provide cost-effective health outcomes advantages over risperidone.