Trends in adult antipsychotic polypharmacy: progress and challenges in Florida's Medicaid program.

We studied trends in antipsychotic polypharmacy over a 4 year period in order to see if a change occurred when a statewide quality improvement program aimed at reducing the practice was implemented. Antipsychotic polypharmacy prevalence rates were calculated for eight 6-month periods for enrollees with schizophrenia and schizoaffective disorder and for those with all other diagnoses. Prevalence increased from 1/2003 to 12/2004 and then declined for 4 successive 6 month periods beginning in the 1/2005-6/05 period when the program began. Piecewise linear regression results for both diagnostic groups confirmed that the change in the likelihood of antipsychotic polypharmacy during the four 6 month periods before program implementation were significantly different than during the four 6 month periods following implementation. While it is impossible to control for the effects of all variables in evaluating the impact of any system wide intervention the data suggest that the program did help to reduce the use of antipsychotic polypharmacy.