Risk of rehospitalization among bipolar disorder patients who are nonadherent to antipsychotic therapy after hospital discharge

Purpose. The relationship between non-adherence to antipsychotic medication after hospital discharge and risk of rehospitalization in patients who were previously hospitalized for treatment of bipolar disorder was studied. Methods. Administrative claims data from 2000 through 2006 were obtained from commercial insurance plans. Patients age 18-64 years who were discharged from a hospital with a diagnosis of bipolar disorder and given a prescription for an antipsychotic 0-14 days after discharge comprised the study sample. Adherence to antipsychotic medication was determined by measuring the number of unique days during which medication was supplied during the treatment period, a calculation known as the medication possession ratio (MPR). Rehospitalization was considered to be an indicator of relapse. A multivariate, stepwise logistic regression, which controlled for patient characteristics, type of bipolar disorder, general health status, and comorbid conditions, was used to assess the relationship between medication non-adherence and rehospitalization. Results. A total of 1973 individuals were included in the analyses. The mean +/- S.D. MPR for this patient population was 0.46 +/- 0.32. Patients whose MPR was 0.75 or greater had a lower risk of all-cause rehospitalization (odds ratio [OR], 0.730; 95% confidence interval [CI], 0.580-0.919) and a lower risk of a mental-health-related rehospitalization (OR, 0.759; 95% CI, 0.603-0.955). As medication adherence increased above the MPR of 0.75, the risk of rehospitalization significantly decreased. Conclusion. Among patients who were previously hospitalized for treatment of bipolar disorder, those who were adherent to their antipsychotic medication at least 75% of the time had lower risks of all-cause rehospitalization and mental-health-related rehospitalization.