An analysis of three existing payment models found certain screening tools and other designs led to some Medicare beneficiaries being excluded and prompted the creation of a new guide on how to root out bias in models.
Researchers with the Center for Medicare and Medicaid Innovation (CMMI) outlined in a Health Affairs article troubling instances of implicit bias in three models: Kidney Care Choices, Comprehensive Care for Joint Replacement and Million Hearts Cardiovascular Disease Risk Reduction. The findings come as CMMI is undergoing a strategic refresh aimed at boosting health equity in value-based care.
“These findings are troubling not only because of the limited access to the benefits of Innovation Center models but also because diverse model participation is critical for robust evaluation and confidence in generalizing results to all of the populations served through [Centers for Medicare & Medicaid Services (CMS)] programs,” the article said.
The center’s review started looking at health disparities in each of the model’s target populations. Researchers then explored the screening tools, provider tools, design of payments and risk adjustment as well as model and evaluation design.
CMMI took initial actions to address the sources of bias for the models, but the authors said the findings underscore the need for more systemic actions. The center has started developing a step-by-step guide for how to screen and mitigate bias within models.
This is the latest action taken by CMMI to address health equity, part of a larger effort by the Biden administration. CMMI called for participants in the ACO REACH model to create a health equity plan outlining the participants’ actions on equity, and officials have said the requirement could be extended to other models.
Here are the key findings for each of the models:
Kidney Care Choices Model
CMMI found that while Black Americans were more than three times more likely to have end-stage renal disease, the model’s medical eligibility criteria for beneficiaries may have erroneously excluded Black Americans.
The model, however, aligns beneficiaries based in part on their level of kidney function rate. But experts have warned that the race adjustment applied to the beneficiary screening tool “artificially elevates kidney function in Blacks, leading to delayed referrals to specialists and transplant listing, and potentially worse health outcomes for Black patients,” the article said.
A taskforce from the National Kidney Foundation and American Society of Nephrology recommended in September 2021 that there should be a revised screening tool that doesn’t include race. The model started its performance period on Jan. 1, 2022, and providers may have assessed beneficiaries based on the race-adjusted tool before the task force’s guidance was released.
Using the race-adjusted tool may have “erroneously elevated black beneficiaries’ kidney function, and, as a result, they may have been incorrectly assessed as not meeting the medical eligibility criteria for the model,” the article said.
Comprehensive Care for Joint Replacement Model
The payment model offers participating hospitals a target price for their episodes of care from the hospitalization to 90 days after the facility discharge. That price is based on the blend of the facility’s historical spending and regional averages, but not on social risk factors.
An evaluation of the model found that beneficiaries “receiving joint replacements at participating hospitals while the model was in effect were less medically complex than those receiving joint replacements at those same hospitals before model implementation began,” the article said.
Beneficiaries in the model were also less likely to be dual-eligible. CMS, in turn, revised its risk adjustment formula for the target price to include dual eligible status when the model was extended starting this year.
Million Hearts Model
The model was developed to predict cardiovascular risk in patients over 10 years. Any beneficiary that had an elevated risk is eligible for the model and gets targeted interventions for heart attack and stroke, the Health Affairs said.
Researchers found that the risk calculator used to predict risk scores were developed specifically for Black and White populations, the article said. It in turn appears to have underestimated risk in patients in other racial and ethnic groups as well as those with lower socioeconomic status.