Black, Hispanic patients underrepresented in value-based care models

CMS's more than 50 value-based care models implemented over the past decade mostly benefit certain demographics, raising questions about health equity and whether data gleaned from these pilots can truly be broadly applicable, researchers wrote in JAMA Network Open.

Most Medicare beneficiaries exposed to the agency's value-based care models have been White, higher-income people living in urban or suburban settings.

This is the result of many of these models' voluntary participation, the researchers wrote. Individual providers and organizations opt into some of these models, and many choose to participate because they anticipate some financial gain when they hit certain quality targets.

Providers treating a more clinically complex population or patients with more social determinants of health might've opted out of many of CMS's value-based care programs over the years. With a higher-risk patient population, these providers may have wanted to avoid potential financial penalties.

Conversely, providers that treat a healthier patient population or folks with fewer SDOH might be compelled to participate in value-based care models, because it might be easier to meet quality measures with a low-risk population.

Ultimately, this could leave some patients to experience value-based care while others don't, the researchers hypothesized. Those who experience value-based care might not reflect Medicare's broader population.

This all bears out in the numbers.

Looking at a random 20% sample of Medicare fee-for-service beneficiaries from 2013 to 2022, the researchers identified different demographics that were and were not attributed to a value-based care model.

Notably, there were serious racial disparities in value-based care participation, with White people being vastly overrepresented in these models. A sizeable 83.3% of those in a value-based care model were White, compared with 72.2% of those not in a value-based care model. This means White people were overrepresented in value-based care by 11.1%.

For Black and Hispanic patients, those proportions were flipped. Only 7.4% of the people in value-based care models were Black, while 4.3% were Hispanic. This compares to 11.2% of those not in value-based care models being Black and 10% being Hispanic.

Said otherwise, Black people were underrepresented in value-based care by around 34%. For Hispanic people, that figure was 57%.

Dual-eligible beneficiaries were also underrepresented in value-based care by around 25%. Meanwhile, folks living in disadvantaged areas were underrepresented by 23%, while those living in rural areas were underrepresented by around 57%.

Disparities in participation in value-based care threaten quality improvement, the researchers argued. Many of these models have been experimental, meaning their results are intended to help create better value-based care models moving forward. With them tested only on certain patient populations, researchers are unable to design models that benefit the Medicare population at large.

But it's not just about program design, the researchers said. These findings raise questions about equity and justice in value-based care.

"In addition to posing a threat to generalizability of model results, our findings of systematic differences between participants and nonparticipants also challenge the internal validity of such experiments," they concluded. "Sociodemographic differences between beneficiaries treated by participants vs nonparticipants may confound attempts at accurate causal comparisons, which policymakers should anticipate in the design of future models."

Sara Heath is an executive editor at Xtelligent Healthcare Media, where she covers patient engagement, healthcare policy and health IT.