OBJECTIVE: A model was developed to assess the clinical outcome and costs of an intensive deinstitutionalization strategy called Intensive Case Management (ICM) in comparison to Standard Case Management (SCM) for persons suffering of chronic schizophrenia and long term hospitalised (>1 year). METHODS: A Markov Model was constructed to describe the different possibilities of case management of long term hospitalised schizophrenic patients. The Markov states were classified to one of five categories hospital, institution, outpatient, dropout, and death, followed by either “success” or “failure” of the adopted case management. Clinical, functioning improvement, and patient discharge from acute care unit define success here. The Markov process iterates in 1-year cycles until the 5-year. Model probabilities are gathered from an ongoing ambispective cohort of patients assigned to ICM or SCM in the catchment area of Clermont-Ferrand. Control patients (SCM) were/are retrospectively chosen on the premise that they could have been selected for ICM group. Service utilisation data are obtained from patient medical and administrative record while direct costs are obtained from the analytical accounting system of the catchment area. RESULTS: Based on the primary data collected (ICM: n=42, SCM: n=28), the outcomes of the two types of care showed, that the ICM was more effective in terms of success over the first one-year period 59% of the patients experienced a success as opposed to 25% in SCM. It was also shown that readmission to hospital was more likely in the SCM (67% as opposed to 22% in ICM). A threshold analysis will be performed in order to evaluate at which point ICM will be too costly for the hospital budget. CONCLUSIONS: With these outcome results, ICM is shown to be more effective than SCM. The model will be used to generate cost-effectiveness ratios, budgetary previsions for large-scale implementation of ICM, in order to help decision-makers in policy-making for mental health care organisation.