Medicaid data are used frequently to investigate racial and ethnic disparities in health. However, there is considerable variation in the completeness of race/ethnicity information in Medicaid data across the United States (U.S.). To address these gaps, the U.S. Census Bureau’s Enhancing Health Data (EHealth) Program assessed the feasibility, benefit, and effectiveness of linking Medicaid enrollment data (T-MSIS) with Census Bureau microdata (i.e., Decennial Census, American Community Survey (ACS), Census Numident) to enable researchers to document and track racial/ethnic disparities in health more effectively. Additionally, this research evaluated whether and to what degree bias was introduced into mortality estimates when Medicaid beneficiaries with missing race/ethnicity information were omitted from analysis. Although Decennial/ACS data reduced missing race/ethnicity information (from 19.47% to 7.11%), this varied considerably across states. Findings suggest that the exclusion of beneficiaries with missing race/ethnicity did not pose a problem for estimating race-specific mortality for most states, but researchers should be cautious about using Medicaid enrollment data for states with a high degree of bias. Results from this research demonstrate significant potential for using Census Bureau data to complement existing health records that commonly lack important demographic characteristics, such as race/ethnicity. Overall, we find that enhancing race/ethnicity information in Medicaid data with restricted Census Bureau microdata is feasible and can advance our understanding of population health.