OBJECTIVES: To identify characteristics associated with antipsychotic polypharmacy (APP) in the Texas Medicaid population. METHODS: Adults newly initiated on antipsychotics between July 1, 2006 and December 31, 2010 were followed for 365 days after the index antipsychotic claim (index date). APP was defined as the concomitant use of two or more antipsychotics for at least 60 days without a gap in polypharmacy greater than 31 days. Monotherapy (MT) was defined as exposure to no more than one antipsychotic at a time during the 1-year post-index period. A logistic regression was conducted   to identify characteristics associated with APP; presence of APP (yes/no) was the dependent variable, and demographic, clinical, physician, and prior utilization characteristics were independent variables. RESULTS: Of the 23,232 eligible patients, 5.4% were on APP and 94.6% on MT during the study period. Older patients (Odds ratio [OR]=1.01) and males (OR=1.14) were more likely to have APP. Patients with bipolar disorder, depression, other mental health diagnoses, multiple mental health diagnoses, and no mental health diagnoses were 44%, 59%, 43%, 53%, and 45%, respectively, less likely to have APP compared to those with schizophrenia/schizoaffective disorder. Those with current substance abuse were 22% less likely to have APP. A 1-unit increase in number of unique mental illnesses increased the likelihood of APP 1.13 times, while a 1-point increase in pre-index Chronic Disease Score decreased the likelihood of APP by 6%. Use of psychotropic and anticholinergic drugs increased the likelihood of APP 1.40 and 2.76 times, respectively. CONCLUSIONS: Identifying predictors of APP could help providers and payers identify patients likely to be prescribed APP early on during the course of their illness. These patients can then be carefully monitored to determine if they are in fact appropriate candidates for APP and managed to ensure they do not experience negative health outcomes.