Hospital readmission rates among patients with schizophrenia treated with long-acting injectables or oral antipsychotics

Abstract

Objective: This study analyzed hospital readmission rates of patients with schizophrenia who were treated with longacting injectable antipsychotics (LAIs) or with oral antipsychotics after being discharged from a hospitalization. Methods: Medical claims of patients with schizophrenia who were ages 18-64 and had a first hospitalization for a serious mental illness (index hospitalization, October 2007 through September 2012) and at least one prescription for a first- or second-generation antipsychotic were analyzed from the Truven Health MarketScan Multi- State Medicaid Database. Analyses were conducted for patients with a sole diagnosis of schizophrenia (N=1,450) and for all patients with schizophrenia (N=15,556), which added patients with a codiagnosis of bipolar disorder or major depressive disorder. Probability of rehospitalization for any cause at 30 and 60 days after the initial hospitalization was assessed with multivariate logistic regression and propensity score matching (PSM) methods. The PSM model matched age, preindex use of LAIs or short-acting injectables, and select comorbidities between the LAI and the oral antipsychotics groups. Results: LAIs were associated with significantly lower probability of rehospitalization compared with oral antipsychotics at 60 days for schizophrenia-only patients (adjusted odds ratio [AOR]=.60, 95% confidence interval [CI]=.41-.90) and for all patients (AOR=.70, CI=.52-.95). The absolute difference in probability of rehospitalization for all patients was significantly lower by 5.0% at 60 days in the LAI group compared with the oral antipsychotics group. Conclusions: Compared with use of oral antipsychotics, use of LAIs was associated with fewer readmissions of Medicaid patients with schizophrenia within 60 days after an index hospitalization.