OBJECTIVES: This study determined incremental Medicaid expenditures associated with dementia among Indiana Medicaid recipients in 2004. METHODS: Analysis of Indiana Medicaid claims and enrollment files was conducted. Sample inclusion criteria were being at least 40 years old and having Indiana Medicaid eligibility in 2004. Twenty-six ICD-9 diagnosis codes determined in a prior study to be specific for dementia were used to identify dementia. Regression models were developed to estimate incremental expenditures in U.S. dollars. Assessment of model fit was used to select from OLS regression, OLS on log of response variables, generalized linear negative binominal, and zero-inflated negative binomial models. Based on fit and Voung statistics, zero-inflated negative binomial regression was used to fit separate models for total expenditures and expenditure subcategories. Models included covariates for age, gender, race, marital status, geographic region, months of Medicaid eligibility, months of spend-down in effect, months of Medicare coverage, mortality, and Charlson comorbidity index score. RESULTS: A total of 145,684 individuals meeting inclusion criteria were identified, 18,950 (13%) had dementia. Individuals with dementia had more comorbidity, were older, and were more likely to be female, white, unmarried, and to die during the study period. After adjusting for covariates, Indiana Medicaid spent, mean (standard error), $9,829 (211) p<0.001, more per recipient with dementia in incremental expenditures associated with the disease. Much of the incremental expenditures were due to nursing home expenditures $9,154 (141) p<0.001. Drug expenditures accounted for the second largest component of the incremental expenditures $970 (37) p<0.001, followed by inpatient expenditures $211 (23) p<0.001 and other medical expenditures $128 (57) p<0.05. Dementia was associated with lower outpatient expenditures $-78 (2.50) p<0.001. CONCLUSION: Incremental Medicaid expenditures associated with dementia were substantial even after adjusting for other risk factors and accounted for $186 million or 8% of 2004 Indiana Medicaid expenditures.