PROVIDENCE, R.I. [Brown University] — New research from Brown University suggests that federal rankings of Medicare Advantage plans may unfairly penalize those that enroll a disproportionate number of non-white, poor and rural Americans.

The study, published in Health Affairs, used data collected by the Centers for Medicare and Medicaid Services (CMS) to measure the quality of care provided in Medicare Advantage plans, and adjusted performance rankings for race, neighborhood poverty level and other social risk factors. After the adjustments, plans serving the highest proportions of disadvantaged populations improved considerably in the rankings.

The findings show that existing Medicare Advantage plan rankings may not accurately reflect the quality of care a given plan’s enrollees receive, said Amal Trivedi, an associate professor of health services, policy and practice at Brown and the study’s senior author.

“Policymakers have focused a lot of attention on measuring quality and rewarding better performance among health plans and providers,” Trivedi said. “But in order for these quality assessments to be accurate, they need to take into account the characteristics of the populations that are served.”

Medicare Advantage is a newly popular option among Americans who qualify for Medicare, according to statistics from CMS. Before the 21st century, almost everyone opted for traditional Medicare, which allowed beneficiaries to visit any medical professional they wanted. But today, almost a third of those who qualify for Medicare choose the more affordable Medicare Advantage option. While patients who use Medicare Advantage are restricted to specific networks of doctors, they’re also able to compare dozens of plans and select the best one for their needs based on rankings, cost and other factors.

For the last decade, Trivedi said, CMS rankings have measured a plan’s quality by examining how well its health care providers perform in about 30 categories, including customer service, efficiency in processing claims and appeals, disease screening rates and patients’ body mass indexes. The Brown researchers adjusted for socioeconomic disadvantage in just three of those categories — blood pressure control, cholesterol control and diabetes control — and found that many lower-ranked plans suddenly moved substantially higher in the rankings.

Shayla Durfey, the study’s lead author and a third-year medical student in the primary care-population medicine program at Brown’s Warren Alpert Medical School, said she and her colleagues chose to adjust the data in those three categories because previous literature has shown that disadvantaged populations disproportionately suffer from uncontrolled high blood pressure, high cholesterol levels and diabetes.

“To control diabetes, for example, you need things like good health literacy, access to healthy foods, and access to money that buys healthy foods,” Durfey said. “If you live somewhere rural and have a low-paying job, you have fewer healthy choices near you, and they’re often too expensive to consider.”

Currently, CMS rankings account for just two risk factors: dual eligibility — which indicates that someone qualifies for both Medicare and Medicaid — and disability. Durfey said that while health scholars have long debated which CMS categories should be adjusted to account for social risk, many experts agree CMS should do more.

“The adjustments CMS uses do not fully account for true measures of socioeconomic status, such as income level, education and employment,” Durfey said. “These factors have been shown to play a huge role in a person’s lifetime health.”

Accurate quality rankings are important, Trivedi said, because CMS gives plans an incentive to compete against each other. A plan that receives a five-star rating is rewarded with a sizeable payment bump. A plan that gets a one-star rating, on the other hand, is penalized: All of its enrollees receive letters encouraging them to switch to better plans.

Trivedi said that if plans notice a connection between their low rankings and their socioeconomically disadvantaged enrollees, they’ll have little incentive to continue serving the underserved.

“Medicare plans can’t deny coverage to anyone with a pre-existing condition, but they can operate in areas that are more affluent or have healthier, less disadvantaged populations, leaving poor and rural populations with fewer and fewer options,” he said.

While Trivedi, Durfey and their co-authors say it’s still unclear what precise set of adjustments will lead to the most equitable CMS rankings, they hope the agency soon takes action one way or another. If one thing is clear to them, it’s that determining whether and how to adjust Medicare Advantage plan quality measures for sociodemographic factors is critically important to accuracy and equitable payment.

“There needs to be a lot more research on the topic, especially as we move toward a value-based payment system where dollars are attached to clinical performance,” Trivedi said. “We need to be sensitive to the effects of these policies on disadvantaged populations and the providers that serve them. That’s really the take-home message.”

In addition to Trivedi and Durfey, other study authors included Amy J. H. Kind, Roee Gutman, Kristina Monteiro, William R. Buckingham and Eva H. DuGoff. The work was supported by the National Institutes of Health (P01AG027296, R01AG044374, R01MD010243).