OBJECTIVES: The objective of this study was to determine if an association existed between race/ethnicity and exposure to Alzheimer’s disease (AD) pharmacotherapy across four state Medicaid populations. METHODS: Data from the Centers for Medicare and Medicaid services was used in this retrospective study.  Individuals had to have an AD diagnosis (ICD-9 code 331.0) and be enrolled in a California, Florida, New Jersey, or New York Medicaid program during 2004.  Outcomes of interest were exposure to a cholinesterase inhibitor (ChEI) or memantine.  Multivariate logistic regression was used to test for the association between race/ethnicity and the exposure to a ChEI or memantine.  Variables of interest included demographic characteristics and resource utilization factors.  The Oaxaca-Blinder decomposition method to test for disparities was used to determine if exposure to a ChEI or memantine was influenced by race. RESULTS: Approximately 158,974 individuals qualified for this study.  Race, age, long-term care admittance, inpatient care admittance, Charlson co-morbidity score, state of residence, and gender all were significant predictors of AD medication use (p <0.0001 for all variables).  Racial/ethnic disparities were observed with respect to exposure to a ChEI or memantine between Non-Hispanic Whites and Hispanics (in favor of Hispanics) in California (p = 0.002) and Florida (p <0.0001), between Non-Hispanic Whites and Non-Hispanic Others (in favor of Non-Hispanics Others) across all states, between Non-Hispanic Blacks and Non-Hispanic Others (in favor of Non-Hispanic Others) in California (p <0.0001) and New York (p <0.0001), and between Hispanics and Non-Hispanic Others (in favor of Non-Hispanic Others) in California (p = 0.001), Florida (p = 0.013), and New York (p <0.0001).   CONCLUSIONS: Disparities in AD medication use among minority-to-minority populations are just as prevalent, if not higher, than minority-white disparities.  Furthermore, geographical location was important predictor of health disparities.