> For the complete documentation index, see [llms.txt](https://wiki.medicaiddatalearningnetwork.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://wiki.medicaiddatalearningnetwork.org/basics/race-and-ethnicity.md). # Race and Ethnicity ## **Overview** Missing race and ethnicity data is a significant and widespread issue across Medicaid T-MSIS Analytic Files (TAF). In calendar year 2020, the share of missing race and ethnicity records varied dramatically by state, ranging from 0% in Delaware and South Dakota to 100% in Rhode Island and Tennessee. Many large states also showed substantial gaps, with Kansas at 83%, Alabama at nearly 72%, and Massachusetts approaching 47%. This variability undermines the ability to conduct reliable, equity-focused analyses at the national level. Race and ethnicity data in Medicaid are self-reported by enrollees at the time of application or renewal. States are required to report enrollee demographic data, including race and ethnicity, through the Transformed Medicaid Statistical Information System (T-MSIS). Two separate fields are used: a race code and an ethnicity code, following national reporting standards. States convert their own data formats into a CMS-approved format before submitting to a relational database. CMS then packages the submitted data into T-MSIS Analytic Files (TAF), which are the research-ready outputs. ## **Data Structure** There are 17 valid race code values and 7 valid ethnicity code values in the TAF. These are combined into an expanded race and ethnicity code with 20 valid values, and a condensed version with 7 valid values. A record is considered missing if both fields are absent or unknown, or if the race code is missing and the ethnicity code is missing or indicates non-Hispanic. ## **Why Data May Be Missing** Gaps in race and ethnicity data can occur for two main reasons. First, enrollees are not required to provide this information since it is not a factor for eligibility. Some individuals may decline due to concerns about discrimination or uncertainty about how the data will be used. Second, states may not always submit complete data to CMS due to technical difficulties. CMS has identified improved T-MSIS race and ethnicity data as a priority area in its guidance to states. Other data fields such as self-reported preferred language and English proficiency are also optional. Data quality is variable by state and is influenced by type of application (online vs. paper) as well as the training of the application assistor. ## Recommendations for Researchers Researchers working with TAF race and ethnicity data are encouraged to treat the DQ Atlas as a starting point rather than a definitive guide. Practical steps include prioritizing states with low to medium concern ratings, conducting independent assessments of subgroups of interest (such as MAGI versus non-MAGI populations), and examining data by region, which is especially important in county-run Medicaid programs. For analyses involving dual-eligible individuals, linking in Medicare race and ethnicity data may be valuable. State websites and household surveys such as the Behavioral Risk Factor Surveillance System may provide additional benchmarking opportunities, though care should be taken to compare populations within similar income ranges. Index of Dissimilarity: Compare the consistency of distribution of race/ethnicity variables on a month-to-month basis. Mathematica uses this method as one of their data quality checks. Leverage Panel or Longitudinal Data Structures: For comparisons that span across time and/or states, it may be helpful to use race records in other years/states to impute a missing value using a bene\_id. **Resources:** * * * * --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://wiki.medicaiddatalearningnetwork.org/basics/race-and-ethnicity.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.