[Cost-effectiveness analysis of schizophrenic patient care settings: impact of an atypical antipsychotic under long-acting injection formulation].

UNLABELLED: Schizophrenia is a disease affecting the young adults and amounts to approximately 300,000 people in France. The French public psychiatric sector takes care of approximately 150,000 adults schizophrenics: 50% benefit from ambulatory care, 50% are in partial or full-time hospitalization care. Schizophrenia represents the first diagnosis that psychiatric sectors take in charge. The costs associated with schizophrenia, mainly hospital costs, are important and were estimated at 2% of the total medical costs in France. In the French social welfare system, the social costs (pensions, allowances, managements of custody or guardianship by social workers) are also to be taken into account: it amounts to a third of the global direct cost. Schizophrenia also generates indirect costs (losses of productivity and premature deaths) which would be at least equal, or even more important, than direct medical costs. The non-compliance to the antipsychotic treatment is a major problem with people suffering from schizophrenia. Indeed the lack of compliance to the treatment, estimated at 20 to 40%, is a major handicap for schizophrenic patient stabilization. The poor level of compliance is due to many various causes: adverse effects that are considered unbearable, medicine viewed as persecutory, negation of the disease, nostalgia for the productive phases of the disease, lack of social support, complexity of the prescription, relapse itself. Compliance is thus influenced by the patient's clinical features, local provision of health care and the specific nature of the drug (adverse effects, pharmaceutical formulation). The atypical antipsychotics present fewer extrapyramidal side effects and reduce the cognitive deficits associated with the disease, which results in improved compliance. Long-acting injectable antipsychotics allow a better therapeutic compliance and thus better efficacy of the treatment. Several studies have shown a significant improvement in compliance related to the pharmaceutical formulation of antipsychotics. Hospitalization and relapse risks are lower in compliant than in non-compliant patients.

OBJECTIVES: The main objective of this pharmacoeconomic analysis is to evaluate the impact in terms of medical benefits and costs of the following strategies: 1. Risperidone long-acting injection: first long-acting injectable atypical antipsychotic; 2. Haloperidol depot: long-acting injectable conventional neuroleptic; 3. Olanzapine: atypical antipsychotic available commercially in oral formulation.

METHODS: The target population defined for the study are young schizophrenic patients treated for at least 1 year and whose disorder has not been diagnosed for longer than 5 years. The time horizon is 2 years. A cost-effectiveness analysis is performed. The perspective adopted is the French Health System. The main hypothesis of the model is that an increase in compliance linked to the use of long-acting injectable formulation could lead to an increased efficacy and a modification of the cost-effectiveness ratio. A decision tree was built. Six periods of follow-up are identified with a duration of 4-months per period. The tree contains 3 principal arms, each one corresponding to a specific treatment: risperidone LA injection, haloperidol decanoate and olanzapine. For each arm, at the chance node, two health states are identified: either the patient responds favourably to the treatment or does not respond favourably and requires a switch to another drug treatment. After a period of response, the patient can either remain in the same state or experiences a clinical deterioration. If the patient presents a clinical deterioration, he can either go back to a positive response state after a period of intensive follow-up or remain in an insufficient response state; in this case, a change of antipsychotic treatment is necessary. In the model, a patient should receive four different treatments before a long-term hospitalization takes put in place. According to the market authorization labelling, clozapine is proposed only as a 2nd or 3rd line therapeutic option, so when at least one or two successive neuroleptics have failed. The efficacy data used in the model are provided by clinical research recently published. These studies estimate the efficacy of oral risperidone, LA risperidone, olanzapine, and treatment by haloperidol. When available data in the literature were insufficient, the opinion of experts was sought. The effectiveness criteria is the rate of patients treated successfully: patients responding to the initial treatment with the possibility of experiencing one or two episodes of clinical deterioration but without requiring a switch to another drug during 2 years of follow-up. The base case is as follows: efficacy for oral risperidone is used for the LA risperidone strategy, increased by 10% within the first 4 months of follow-up; efficacy for oral haloperidol is used for haloperidol depot, increased by 5% within the first 4 months of follow-up; for olanzapine, observed data in clinical trials were applied. The hypotheses for long acting forms are rather conservative because the increase of efficacy which can be expected for the long-acting injectable formulations varies between 5% to more than 30% according to the literature data. The analysis of sensibility includes three scenarios: scenario 1: for LA risperidone, 5% of patients treated successfully improvement in regard to oral risperidone instead of 10% in the base case; scenario 2: for haloperidol depot, 10% of patients treated successfully improvement in regard of oral haloperidol instead of 5% in the base case; scenario 3: the results of an open trial conducted within the framework of the LA risperidone license are used, leading to an increase of up to 13,3% of the rate of successfully treated patients, compared to oral risperidone literature data. As for the side effects, only extrapyramidal symptoms were considered. Other side effects are described in the literature such as the obesity or the occurrence of a diabetes; these effects were not taken into account in the model, their impact on the cove-rage of schizophrenic patients and on resources utilisation being poorly known. Only direct medical costs were considered in the pharmaco-economic analysis. Two types of costs were identified: hospital costs and community care costs. The stays in overnight hospitalisation and day hospitalisation were derived from the Disease Related Groups (DRG) and valued from the data of the National Cost Study (Etude Nationale de Coûts; 1999). The DRGs corresponding to the diagnosis of schizophrenia are the DRG 627 (complete hospitalization) and DRG 819 (day hospitalisation). Ambulatory care: procedures and visits, were valued in euros in reference with the tariffs for reimbursement issued in the Naming General of the Professional Acts (NGAP) and published by the French National Health Insurance (Year 2001). Medication consumption was quantified by using the daily dosage specified in the the MAA and the French prescription database IMS-Dorema. The cost of medicines was valued from tariffs 2001 (SEMPEX). LA risperidone price being not fixed to date, the reserved hypothesis is a 141,62 Euro retail price. As schizophrenia is listed among the diseases reimbursed at a 100% rate by the Health insurance, out of pocket expenses by patient are not considered in the analysis. The cost for the extrapyramidal effects was attributed to all the strategies. This cost was calculated according to the rates of extrapyramidal effects occurrence collected in the literature. Globally, in the published studies, the incidence of the side effects for the patients treated by olanzapine or risperidone is similar. It was thus decided by the experts to use the same rate of occurrence for extrapyramidal effects for olanzapine and risperidone (20%). This rate is 40% for haloperidol decanoate, 10% for oral clozapine. For the cost estimation, the expenses for treating a schizophrenic patient for two years were taken into account.

RESULTS: The results show that in two years, LA risperidone is more effective than the two other antipsychotics. After 2 years, the rate of patients treated successfully is 82,7% for LA risperidone, 74,8% for olanzapine and 57,3% for haloperidol depot. The 2 year-cost per patient treated by LA risperidone is 14,055 Euro. This cost is 14,351 Euro and 17,203 Euro respectively for the strategies olanzapine and haloperidol depot. The cost-efficacy ratios per patient successfully treated are 16,995 Euro for the strategy LA risperidone, 19,186 Euro for olanzapine and 30,023 Euro for haloperidol depot. LA risperidone is a dominant strategy compared with both olanzapine and haloperidol depot. Scenario 1 shows that LA risperidone strategy remains the most effective. Indeed, this strategy allows a response increase of 3,5% regarding olanzapine strategy and of 21% regarding haloperidol depot strategy. Under the hypothesis tested in scenario 1, LA risperidone is a partial dominant strategy against olanzapine and a total dominant strategy against haloperidol depot. In scenario 2, as efficacy is improved for haloperidol decanoate (61,10%), a decrease of 1,763 Euro in the cost per patient treated is observed for this strategy. Cost per patient treated successfully and efficacy for LA risperidone and olanzapine are the same than in the base case. LA risperidone is a total dominant strategy against olanzapine and haloperidol decanoate. In scenario 3, the rate of patients treated successfully at 2 years is 88,6% for LA risperidone with a cost per patient of 12,347 Euro. LA risperidone is dominant against olanzapine and haloperidol depot.

DISCUSSION AND CONCLUSION: The schizophrenia is a relatively frequent disease. (ABSTRACT TRUNCATED)