A recent report from the HHS Assistant Secretary for Planning and Evaluation (ASPE) found that social risk factors and indicators such as disability status, low income and lack of social supports were strong signals of poor health outcomes and negatively affected performance on Medicare quality measures and value-based reimbursement programs. Dual-eligible status was the greatest predictor of poor health outcomes in the quality measures studied. The report made various recommendations for addressing inequities in performance assessment across organization types that industry experts say bode well for high-dual health plans and safety net providers that want to see additional enhancements to the quality ratings system.

The 374-page report, released Dec. 21, examined nine Medicare payment programs, including the Hospital Readmissions Reduction Program, the Physician Value-Based Payment Modifier Program (soon to be replaced by the Merit-based Incentive Payment System) and the Medicare Advantage Quality Star Rating Program. Beneficiaries with “social risk factors” (replacing the term low socioeconomic status as per a recommendation from the National Academies of Sciences, Engineering and Medicine) demonstrated worse outcomes on many quality measures, regardless of the providers they saw.

Outcomes in many instances were found to be particularly poor for those dually eligible for Medicare and Medicaid. For example, dually enrolled beneficiaries had significantly greater odds of hospital readmission than non-dually enrolled patients even within the same hospitals. And after adjusting for clinical risk factors, their odds of readmission remained higher for duals vs. non-duals.

Using data from performance year 2014, researchers observed that dually enrolled or low-income subsidy (LIS) beneficiaries, black and rural beneficiaries, beneficiaries living in low-income neighborhoods, and beneficiaries with disabilities experienced worse outcomes compared to other beneficiaries on “many to most” of the star ratings metrics.

To evaluate the impact of social risk factors on stars performance, ASPE used an analytic sample containing 15,282,565 MA beneficiaries (or about 97% of total MA enrollees in 2014), of which 18.2% were partially or fully dual-enrolled, 14.6% had disabilities and 3.6% qualified for LIS. ASPE considered these a single group for the purposes of its analysis. After removing contracts that didn’t have 2015 star ratings (e.g., those that were too new or too small to be rated), the sample consisted of 505 contracts, which included HMOs, PPOs and private fee-for-service plans.

Researchers considered total and within-contract differences for 19 of the 47 MA star ratings measures, all of which were beneficiary-level measures, and found that for 16 of the measures, performance was worse for dual/LIS beneficiaries within a contract, although to what degree they differed varied. For example, dual/LIS members had 32% lower odds of having their blood sugar controlled than non-dual/non-LIS beneficiaries but only 7% lower odds of having their kidneys appropriately monitored.

ASPE concluded that its analyses showed “evidence of a significant within-contract impact of a number of social risk factors on performance on individual measures in the Star Ratings.” However, ASPE observed that since most of these measures are not adjusted for clinical risk, the research does not show whether the differences are due to the social risk factor itself or to the fact that beneficiaries with social risk factors also have a higher risk for certain medical conditions.

Although some contracts serving a high proportion of dual/LIS members earned high star ratings, such contracts generally fared worse on overall quality scores, added ASPE. As a result, plans with a high proportion of duals members are much less likely to receive quality bonus payments. “The impact is really significant for these plans, and I think this report confirmed prior research and strengthened our current thinking, which is that we need to use better data,” observes Christie Teigland, Ph.D., vice president of advanced analytics with Avalere Health.

Teigland currently sits on the standing disparities committee of the National Quality Forum (NQF), which is nearing the end of a two-year pilot to assess the impact of social risk factors on quality outcomes. But the various measures NQF’s developers have tested thus far found little impact of social risk factors, reports Teigland, who argues that more robust data is needed to “accurately measure which of these social risk factors are having the biggest impact on quality outcomes.”

After separate studies from Avalere and CMS supported the theory that SES can have an adverse effect on star ratings, CMS last year attempted to address the concerns of plans with large percentages of dually eligible beneficiaries by adding a Categorical Adjustment Index (CAI) to the star ratings for contracts that serve a high proportion of duals/LIS members and/or patients with disabilities (MAN 12/3/15, p. 1). But only 15 out of 364 contracts saw their 2017 summary rating go up by half a star, estimated Avalere’s parent company, Inovalon, Inc. (MAN 11/3/16, p. 1).

The ASPE report outlined several MA-related strategies that fell into three buckets: (1) measure and report quality for beneficiaries with social risk factors, which included enhancing data collection and developing statistical techniques to allow such actions on key measures; (2) set high, fair standards for all beneficiaries; and (3) reward and support better outcomes for beneficiaries with social risk factors, such as by providing targeted star adjustments to reward contracts that demonstrate high quality or significant improvements in quality for dual eligibles.

“I think they’re moving in the right direction,” observes Teigland of the ASPE recommendations. “If you look at what the [CAI] adjustment did, less than 3% of plans getting any kind of impact [shows] it’s not going far enough. The ASPE report acknowledges that we need to go further.”

Report Validates SNP Concerns

The SNP Alliance also agrees with ASPE’s recommendations in terms of implementing changes to the quality measurement and reporting system under Medicare. This would include improving measures, risk adjusting for social determinants of health and dual status, enhancing data collection, grouping like plans, and moving toward more meaningful population groups for measurement and reporting. The alliance tells AIS Health that the ASPE report “validates some of the points the Alliance has been making for the last few years about the effects of a high proportion of duals on observed quality outcomes.” The alliance strongly supports, for example, comparing like plans to like plans (e.g., Fully Integrated Dual Eligible Special Needs Plans to other FIDE-SNPs), considering specific exclusions or exceptions, and adjusting more of the star ratings to take into account dual status.

“Our sense is that CMS was really waiting for this ASPE report before they went ahead and developed or proposed a long-term solution to the star ratings methodology, so the report could inform CMS’s thinking in terms of how they propose to really adjust the star ratings for social risk factors,” adds Christine Lynch, senior vice president of policy and external affairs at The SNP Alliance.

That said, while the alliance is “very concerned about advancing fair and appropriate performance measurement, we also don’t want to sweep poor performance under the rug,” adds Rich Bringewatt, co-founder and CEO of the National Health Policy Group, of which the alliance is an initiative. “We want to report it where it exists and want to do everything we can to improve clinical and care outcomes as well as address imperfections in performance ratings.”

In addition, researchers looked at the impact of Medicare’s current value-based payment programs on providers serving socially at-risk beneficiaries and found that those who served a disproportionate number of disadvantaged beneficiaries also tended to perform worse on quality measures.

“I think what’s important about this report is it not only substantiates the importance of social risk factors in caring for poor beneficiaries, it really proves the pervasive extent to which performance rating methods have really failed to account for all of this across the spectrum of plans and providers,” weighs in Bringewatt.

View the report at https://aspe.hhs.gov/reports.