Medicare needs better data on race and ethnicity

The Office of Inspector General said the inaccurate and incomplete data makes it harder to improve health equity, a major goal of the Biden administration

The Medicare program has failed to get accurate data for race and ethnicity of some groups, a federal watchdog says.

The Office of Inspector General of the U.S. Department of Health and Human Services found that Medicare’s data is lacking in some cases, particularly among Hispanic, American Indian/Alaska Native, and Asian/Pacific Islander beneficiaries.

The incomplete or inaccurate data poses a big problem in both identifying and eliminating disparities in healthcare among minority groups, the inspector general’s office says in the report, which was released June 15. More than 66 million Americans are enrolled in Medicare.

“Inaccuracies in race and ethnicity data have far-reaching consequences, affecting understanding of disparities in the prevalence, severity, and outcomes of diseases and conditions—including COVID-19—and in health care quality and access,” the report stated.

“As such, these inaccuracies limit the ability to effectively design and target efforts to address disparities in these areas and to measure the results of such efforts."

Chiquita Brooks-LaSure, the administrator of the Centers for Medicare and Medicaid Services, has said repeatedly that addressing health equity is atop the agency’s priorities.

“How are we promoting health equity is the first question we are asking ourselves, not the last one,” she said at the HIMSS 2022 conference in March.

Nonetheless, Medicare needs to get accurate data to make progress, the inspector general’s office said, and too often, it’s not happening.

In some cases, Medicare is identifying patients with a race or ethnicity that nursing home residents don’t use to identify themselves.

The report said 28% of the beneficiaries identified as Hispanic in the Medicare enrollment data do not identify themselves as Hispanic on their nursing home assessments. Nearly half (46%) of the beneficiaries in Medicare’s data who are identified as American Indian or Alaska Native don’t identify themselves that way.

At the same time, Medicare in some instances isn’t using the race or ethnicity that beneficiaries do use in their own descriptions.

The report said 35% of those beneficiaries who identified as American Indian/Alaska Native beneficiaries are not listed that way in Medicare’s data. Roughly a quarter (24%) of those identifying as Asian or Pacific Islander are not classified that way in Medicare’s records.

In addition, the inspector general said Medicare is utilizing the Social Security Administration as its primary source for information on race and ethnicity. It’s problematic because Social Security stopped widely collecting data on race and ethnicity in 1989, so Medicare’s records are lacking.

Plus, up until 1980, Social Security beneficiaries could only identify themselves as white, Black or other, so that information is limited to three categories for most Medicare beneficiaries.

For 3.3 million Medicare beneficiaries, the data for race is either listed as “other” or “unknown,” the inspector general’s report stated.

Medicare’s data collection on race doesn’t comply with federal standards, according to the report. The Census Bureau, for example, is gathering much more complete information on race and ethnicity.

The inspector general is recommending that Medicare develop its own source of data for race and ethnicity, while using self-reported data for better information on current beneficiaries. In addition, Medicare should develop a process to ensure data is as standardized as possible, and educate beneficiaries about CMS’ efforts to improve race and ethnicity data, the report said.

CMS agreed with most recommendations but did not explicitly concur with the suggestion that Medicare should develop its own source of data, the inspector general’s office said.