A 1-year semi-Markov model was constructed to simulate the cost-effectiveness of atypical and typical antipsychotic treatments for schizophrenia.
The core model comprised nine 6-week cycles and includes events such as survival, response, adverse events, and compliance. The nature, duration, intensity, and timing of adverse events were incorporated. Compliance was modeled as a function of health state, time, and adverse events. Three first-line treatments were considered (risperidone, olanzapine, and haloperidol oral) and the transition probabilities of switching between five different therapies (haloperidol oral, haloperidol depot, risperidone, olanzapine and clozapine) were included. Effectiveness was modeled based on a modified method of TWiST (time without symptoms and toxicity). The direct costs of utilization of medical resources are taken into account, including five different patient care settings, consultations, neuroleptic medication, laboratory tests, and treatment of side-effects.
This paper reports the methodology used to construct the model and the results obtained when it was applied to the treatment of patients with schizophrenia in the Belgian health care system.
Over the study period, risperidone and olanzapine were more cost-effective than haloperidol and of the two major atypical drugs, risperidone was the more cost-effective.