Medicare has a data problem

The race and ethnicity data Medicare collects on its beneficiaries is less accurate for such groups as American Indians, Asian-Pacific Islanders and Hispanics, limiting the program's ability to assess health disparities, according to a federal audit.

Why it matters: Inaccuracies can undercut efforts to address inequities and measure results.

• The pandemic underscored sharp disparities among Medicare beneficiaries, with Black, Hispanic and American Indians accounting for disproportionately high rates of cases and hospitalizations.

Go deeper: Medicare's primary source for race and ethnicity information is Social Security data, even though it's limited to only a few categories and missing for 3.3 million beneficiaries, the HHS inspector general's office reported in the audit.

• Medicare administrators applied an algorithm to improve visibility that relies on a list of names frequently associated with Asian-Pacific Islanders and Hispanic individuals.
• But even after applying the algorithm, 1.3 million beneficiaries still have "Unknown" or missing information, while another 600,000 are categorized as "Other," the inspector general found.

What's ahead: The program is taking steps like incorporating race and ethnicity questions that are consistent with the HHS standards and is expanding use of an alternative algorithm.

• The inspector general suggested Medicare also develop a way to collect race and ethnicity information from beneficiaries when they enroll and to incorporate the existing data from the nursing home and other post-acute care assessments that have more race and ethnicity categories.