RESUMEN
OBJECTIVE: To evaluate the cost-effectiveness of atypical antipsychotic treatment sequences for the management of stable schizophrenia in the UK. RESEARCH DESIGN AND METHODS: A Markov model was developed to assess the cost per quality-adjusted life year (QALY) gained from 12 alternative treatment sequences each containing two of four atypical antipsychotics (aripiprazole, olanzapine, quetiapine and risperidone), followed by clozapine. The main model parameters were populated with data from the Clinical Antipsychotic Trials of Intervention Effectiveness (CATIE) study and a recent trial comparing aripiprazole with olanzapine. Patients enter the model with stable schizophrenia and may relapse, discontinue or continue and experience adverse events (AEs), or develop diabetes. Population mortality was adjusted for schizophrenia and diabetes. Utility decrements applied to stable schizophrenia, relapse, diabetes and treatment-related AEs were taken from a direct UK utility elicitation study. Resource use and unit costs were taken from published sources. A time horizon of 10 years was adopted. Results are based on 10,000 probabilistic iterations of the model. RESULTS: Aripiprazole followed by risperidone produced the greatest number of QALYs, an additional 0.03 compared with risperidone followed by olanzapine, at an incremental cost of £257 (incremental cost/QALY: £9,440). Aripiprazole followed by risperidone had the greatest probability among evaluated sequences of being cost-effective at a threshold of >£10,000/QALY. All other strategies were dominated by at least one of these strategies. The impact of lower pricing for risperidone (based on generic availability) did not impact results. CONCLUSIONS: Modelling the cost-effectiveness of different treatment sequences for stable schizophrenia is appropriate given that patients rarely remain on one treatment for long periods. The treatment sequence aripiprazole followed by risperidone was the most cost-effective option for patients with stable schizophrenia in the UK.