OBJECTIVES: To estimate the prevalence and trends of antipsychotic polypharmacy, categorize polypharmacy according to the type of antipsychotic and duration of use, and contrast polypharmacy usage patterns with published treatment guidelines. METHODS: Medicaid recipients at least 16 years of age with at least one primary diagnosis of schizophrenia (ICD-9-CM=295) between 1998-2000 were identified from the Georgia and California (20% random sample) Medicaid claims databases. Polypharmacy cohorts were built in a hierarchical fashion based on the antipsychotic use profile of the schizophrenia patients and in accordance with published treatment guidelines i.e. any polypharmacy, clozapine (clozapine + atypical; clozapine + conventional), non-clozapine (atypical+atypical; conventional+conventional; and atypical+conventional) and long-term i.e. duration of use > 2 months. Total 3-year prevalence of polypharmacy, mean/median duration of episodes and year wise trends in usage were estimated. All results were reported by state and in aggregate. RESULTS: Out of a total of 31,435 persons with schizophrenia, the overall prevalence of antipsychotic polypharmacy was 40% (n=12,549, mean age: 43 years, caucasian: 47%, female: 48%) over 1998-2000 and was 46% in California compared to 35% Georgia (p<0.0001). Long-term polypharmacy (exposure > 2 months) had a prevalence rate of 23% (n=7,222) with a long-term episode lasting a median of 197 days. Among the long-term groups, non-clozapine polypharmacy was around 8 times more prevalent than clozapine polypharmacy (p<0.0001) and long term non-clozapine atypical + conventional had the highest prevalence of 16%. Except for clozapine+conventional (no change) and conventional + conventional (decreased) polypharmacy, all polypharmacy prevalences increased from 1998 through 2000 (Cochran-Armitage test; p<0.0001). CONCLUSIONS: The overall polypharmacy prevalence rate of 40% is similar to that observed in other studies. Average long-term polypharmacy with a median duration of over 6 months and significantly higher prevalence of non-clozapine vs. clozapine polypharmacy contradicts treatment guidelines and may be a cause for concern.