Cost effectiveness analysis of olanzapine and risperidone

Background: Schizophrenia is a major public health problem throughout the world and known to be a major cause of disability. Antipsychotic medications are the mainstay of treatment for schizophrenia and olanzapine and risperidone are popular choices among atypical antipsychotics in Norway. Despite of the high number of prescribing olanzapine, however, cost-effectiveness of olanzapine versus risperidone is still debated.\n\nMethods: A decision analytic model combined of a decision tree model and a Markov model was developed to compare the ICER of olanzapine and risperidone and determine the optimal alternative for schizophrenia treatment in Norway. The PANSS was used to measure effectiveness of the antipsychotics and costs were measured with a perspective of providers / funders of health and social care services in five years time frame. The Cochrane meta-analysis and Norwegian cost analysis provided the main source to operate the model.\n\nResults: The total five-year expected cost per schizophrenic patient for olanzapine was NOK 1,878,072 and for risperidone NOK 1,943,868. The total expected PANSS reductions were 112.60 and 111.55 for olanzapine and risperidone, respectively, thus olanzapine was a dominant alternative to risperidone. However, the results from the PSA using Monte Carlo simulation indicated that olanzapine and risperidone are not different in terms of cost-effectiveness within 95% confidence interval. The ICE scatterplot demonstrated that the chance of olanzapine being an optimal alternative was 67.1% in the model.\n\nDiscussion and conclusion: Considering all of the uncertainties surrounding the costs and effectiveness of the olanzapine and risperidone, the model could not conclude that olanzapine was more cost-effective than risperidone in Norway. For more precise results, further clinical trials are recommended to follow up patients who drop out of the study and evaluate relapse rates categorized into compliant and noncompliant patients distinctively. However, the model seemed capable to figure the overall total costs per schizophrenic patient treated with antipsychotic medications in Norway. In addition, the model can be used to provide a basic frame of modelling patients with schizophrenia diagnosis and to facilitate further schizophrenia studies.