OBJECTIVES: To estimate the prevalence of dementia in the Indiana Medicaid population and to evaluate how varying the dementia patient selection criteria used in previous studies affects prevalence estimates. METHODS: Indiana Medicaid claims data for the period July 1, 2001 through December 31, 2003 were analyzed. An expert panel survey also was conducted to assess specificity of ICD codes used in previous studies to define dementia. Prevalence estimates were calculated with varying levels of each of the selection criteria, i.e., ICD code set, interval of data examined, and number of occurrences of dementia-related claims. RESULTS: Prevalence varied greatly across age strata. The impact of each selection criteria on prevalence estimates was similar. Depending on selection criteria used, the prevalence of dementia for individuals 40 years old or older varied from 6% to 13%, while prevalence for individuals 60 years old or older varied from 11% to 22%. Using as criteria the occurrence of two or more dementia-related claims, the expert panel ICD set, and 30 months of data for defining dementia, the prevalence of dementia in the Indiana Medicaid population was 7.9% for individuals 40 years old or older, 11% for individuals 50 years old or older, and 14% for individuals 60 years old or older. CONCLUSIONS: Selection criteria used to identify patients with dementia from medical claims can affect estimates of prevalence by nearly two-fold. Prevalence of dementia among Indiana Medicaid beneficiaries age 50 years old or older and among beneficiaries age 60 years old or older was 2.5 to 3.5 times higher than the reported prevalence from a decade ago in Medicaid populations of other states, even when the same selection criteria were used. This analysis provides an ICD code set that can be used in claims-based work, to identify patients with dementia.