Effectiveness of Antipsychotic Therapy in a Naturalistic Setting: A Comparison Between Risperidone, Perphenazine, and Haloperidol

Background: Therapeutic ineffectiveness and noncompliance with antipsychotic agents are major contributors to rehospitalization in patients with psychotic disorders. It is unknown whether risperidone's favorable side effect profile compared with that of the conventional antipsychotics results in improved compliance and reduced hospitalizations in a naturalistic setting. The purpose of this study was to test the hypothesis that treatment with risperidone reduces readmission rates and associated costs when compared with treatment with perphenazine or haloperidol.

Method: Inpatients prescribed either risperidone, perphenazine, or haloperidol between January 1, 1995, and December 31, 1995, as a single oral antipsychotic at discharge were retrospectively identified. Data were collected for that index hospitalization and for a 1-year follow-up period. Primary outcome measures included readmission rates, changes in antipsychotic therapy, anticholinergic drug use, and costs.

Results: There were 202 evaluable patients (81 treated with risperidone, 78 with perphenazine, and 43 with haloperidol). Baseline demographics were similar between groups except that more patients in the risperidone group had a primary diagnosis of psychotic disorder or had been hospitalized in the year prior to study. The percentage of patients readmitted during the 1-year follow-up period was similar among drug groups (41% risperidone, 26% perphenazine, and 35% haloperidol) when controlled for baseline differences in diagnosis and hospitalization history (p = .32). Anticholinergic drug use was more common in the haloperidol group (p = .004). Mean yearly cost (drug + hospitalization) in the risperidone group was $20,317, nearly double that in the other treatment groups (p < .001).

Conclusion: The results from this naturalistic study indicate that the high cost of risperidone is not offset by a reduction in readmission rates when compared with conventional antipsychotics.

J Clin Psychiatry 1999;60(12):850-856