Antipsychotic polypharmacy : an examination of pharmacoepidemiology and potential predictors of utilization

Access full-text files




DeLeon, Anthony

Journal Title

Journal ISSN

Volume Title



Despite frequent utilization, little data are currently available documenting outcomes associated with antipsychotic polypharmacy. The purpose of this study was to evaluate antipsychotic polypharmacy pharmacoepidemiology and identify potential predictors of utilization. The CQI+™ database (Mental Health Outcomes Inc.) provides outcomes data for patient and treatment-focused variables for all age/diagnostic groups in multiple psychiatric settings. From this database, adults (18-65 years) diagnosed with schizophrenia-related or schizoaffective disorder prescribed one or more antipsychotics upon discharge between January 1998 and May 2004 were eligible for inclusion. Demographics and other characteristics of patients receiving antipsychotic monotherapy or polypharmacy were compared using the Pearson chi-square test and student’s t-test. All tests were two-tailed with significance defined as an alpha < 0.05. Binary logistic regression was performed to identify significant predictors of patients receiving antipsychotic polypharmacy at discharge with complete covariate data. A stepwise forward conditional method was utilized with a covariate entry criterion of an alpha < 0.05. A total of 3,003 patients received antipsychotic monotherapy (n=2185) or polypharmacy (n=818) upon discharge. Significant associations were observed between antipsychotic polypharmacy use and race/ethnicity (p=0.001), comorbid substance abuse history (p<0.001), mental health facility regional location (p<0.001), and payment method (p<0.001). Those receiving polypharmacy had higher mean number of hospitalization days (39.44±72.55 vs. 21.63±50.15, p<0.001), higher mean admission Behavior and Symptom Identification Scale (BASIS-32) Psychosis subscale scores (1.10±0.04 vs. 1.05±0.03, p=0.040), and lower mean admission BASIS-32 Depression/Anxiety subscale scores (1.79±1.15 vs. 1.89±1.11, p=0.041). Two, three, and four antipsychotic regimens were prescribed in 24.1%, 2.9%, and 0.27% of patients receiving an antipsychotic, respectively. A typical antipsychotic and olanzapine or risperidone were most frequently prescribed antipsychotic combinations. Significant predictors of antipsychotic polypharmacy included: charity/contract funding (p=0.001), northeastern or western United States location (p<0.001), positive substance abuse history (p=0.037), a higher number of previous hospitalizations (p=0.004) and higher admission BASIS-32 Psychosis subscale scores (p<0.001). Despite the lack of evidence supporting its use, antipsychotic polypharmacy is frequently prescribed. Potential predictors of antipsychotic polypharmacy include substance abuse history, payment method, service region, and illness severity. These data provide important information regarding the phenomenology of antipsychotic polypharmacy


LCSH Subject Headings