CMS Releases Data Briefs That Provide Key Medicaid Demographic Data for the First Time

The Centers for Medicare & Medicaid Services (CMS) considers the full range of Medicaid and CHIP enrollees in its decision-making and in its efforts to measure disparities in access to care. CMS is committed to making focused investments to improve health equity. [1] Building on this commitment, CMS released a series of data briefs using the first-ever national estimates of the demographic composition of Medicaid and CHIP, including analyses stratified by race, ethnicity, primary language, geography, and eligibility on the basis of disability. In addition to representing a major step forward in data transparency, these data briefs more accurately describe the demographic makeup of program enrollees and provide a richer picture of the individuals served by Medicaid and CHIP, which together constitute some of the nation’s largest and most vital health coverage programs.

Thanks to this demographic data enhancement, for the first time in agency history, CMS analysis has yielded the following important insights. These results are not only a significant step forward in the data capacities of CMS, they also underscore the critical importance of Medicaid and CHIP in addressing health equity:

• Race and ethnicity: Medicaid and CHIP provided coverage for nearly 55 million people from racial and ethnic minority backgrounds in 2020. Further, the programs’ enrollees were more racially and ethnically diverse than the broader U.S. population. These findings are particularly pronounced for children, with 61 percent of child enrollees in 2020 being from racial and ethnic minority backgrounds.
• Residence in Rural Areas - Medicaid and CHIP enrollees are slightly more likely to reside in rural areas than the total U.S. population. Enrollees in rural areas are more likely to be non-Hispanic White and non-Hispanic American Indian and Alaska Native (AIAN) than enrollees in non-rural areas.
• Primary Spoken Language - Over 10 percent of Medicaid and CHIP enrollees have a primary language other than English, which is slightly lower than the overall U.S. population. The primary language brief excludes 16 states - a few of which are populous states with many non-English-speaking residents - due to incomplete or unreliable primary language data in the 2020 Medicaid and CHIP data. [2] Among remaining states with sufficient data quality, enrollees with a non-English primary language are more likely to be Hispanic or non-Hispanic Asian/Pacific Islander (API) compared to enrollees whose primary language is English. They are also more likely to be over the age of 65, and to have qualified for benefits through an eligibility category open only to older adults.
• Disability - Around 10 million Medicaid enrollees (or 11 percent of the Medicaid population) qualified for benefits based on disability in 2020. Most of these enrollees obtained Medicaid coverage because they receive Supplemental Security Income (SSI), indicating they have limited financial means and a long-lasting disabling condition. Medicaid enrollees who qualify for benefits based on disability are more likely to be non-Hispanic White or non-Hispanic Black than those who qualify for Medicaid through other eligibility categories.

To appreciate the significance of these new analyses, it is important to understand the context around demographic data collection in Medicaid and CHIP. Although survey results indicate that Medicaid and CHIP cover large, diverse populations, data quality issues have historically made it impossible for CMS to use administrative data to analyze the demographic composition of the programs. These issues are particularly challenging for demographic data fields that enrollees are not required to report, such as race and ethnicity and language, because those factors are not considered during eligibility determination. [3] As such, many enrollees do not report these characteristics, and states are therefore unable to report this information to CMS. CMS has invested significant resources in working with states to improve data quality relating to these elements in the face of the collection barriers, [4] yet the resulting lack of information left a large gap in federal and state agencies’ collective understanding of Medicaid and CHIP. Now, instead of relying on survey data, CMS has developed a well-validated, evidenced-based method for combining high-quality, self-reported race and ethnicity data with indirect estimates known as imputation, which underlies this new data analysis. [5] Alongside imputation, CMS invested in geocoding of enrollee address information that enables this new data analysis by geography. The insights from these new analyses will enable CMS to better understand and more equitably serve the diverse population of Medicaid and CHIP.

Moving forward, CMS will build upon the historic nature of these products to include analyses of demographic data in additional data products. Updated counts that reflect data from later years will be publicly available for download on data.medicaid.gov as the data become available for each year.