Objective: to estimate the costs and effects of long-acting risperidone (LAR) in the treatment of schizophrenic patients in Italy, as compared to conventional and oral atypical antipsychotics. Methods: a discrete event model was used. The model simulates patients. history for every single therapeutic alternative and selects incident events, on the basis of pre-defined probability distribution-powered, randomized repetitions. The model operates on two types of parameters: patient characteristics and time-dependent variables. Patient characteristics (age, sex, illness profile and severity, probability of incurring in an adverse event and potential dangerousness) remain fixed during the 5 simulated years. Time-dependent variables are subject to changes and include outpatient visits, severity of psychotic episodes, symptom-scores, compliance, incidence of adverse effects, site of treatment and dangerousness. Three treatments have been selected: scenario 1 begins with LAR, switches to olanzapine and then to clozapine; scenario 2 starts with olanzapine, switches to oral risperidone and ends with clozapine. Direct medical costs have been computed on the basis of psychiatric visits, drug costs and costs of the institution in which the patient is treated (hospital, rehabilitation clinic, etc.) Outcome measures were number of psychotic episodes in 5 years, total time spent during these episodes and cumulative score of positive and negative symptoms at 5 years. Information on alternatives, transition probabilities, model structure and health resources utilization were derived from the literature and from a panel of experts. Results: it has been estimated that LAR is economically dominant (more effective at lower cost) respect to oral atypical antipsychotics, being able to prevent 0.87 psychotic episodes per patient, with a net cost saving of 4,773 euro per patient. Sub-group analysis indicate that LAR is always more effective than the considered alternatives and, in general, also less costly than oral atypical antipsychotics. Sensitivity analyses confirmed the robustness of the model to several variations of key parameters. Conclusions: LAR therapy dominates oral atypical antipsychotics.