Reprinted with AIS Health permission from the February 20, 2020, issue of RADAR on Medicare Advantage.
In addition to codifying previous policy changes and extending some new flexibilities to Medicare Advantage organizations, a recent 900-page proposed rule (85 Fed. Reg. 9002, Feb. 18, 2020) from CMS contains some significant changes to the star quality ratings. And while CMS is giving plans plenty of lead time to prepare for proposed modifications that would take effect with the 2021 measurement year, it may be hard for plans to keep up with the sheer volume of planned changes, suggests Melissa Smith, senior vice president of stars & strategy at Gorman Health Group, LLC.
“I think the good news is that CMS is delivering on its promise for transparency and stability and plenty of warning,” says Smith, referring to a contract year 2019 rule that codified key aspects of the Parts C and D star ratings rather than allowing CMS to continually make retroactive changes to measures.
In addition to making “routine measure updates and technical clarifications,” CMS in the rule released Feb. 5 proposed increasing the weight of patient experience/complaints and access measures from 2 to 4 beginning with the 2021 measurement year. “That is significant because it really puts plans to the test of whether their efforts are satisfying and pleasing their customers, and most plans very much struggle to improve performance on these measures,” Smith tells AIS Health. And while CMS suggested that plans on average would not see their star ratings drop as a result, the change could have a range of impacts, she points out.
“Plans that already have strong [Consumer Assessment of Healthcare Providers and Systems] ratings will benefit from this change, and plans that struggle in CAHPS and currently compensate for that and have strong ratings on HEDIS [Healthcare Effectiveness Data and Information Set] and administrative measures are really facing a moment of clarity while they prepare for this change,” says Smith.
CMS also proposed the removal of outliers before calculating star measure cut points using the Tukey outlier method, which Smith describes as a “complicated statistical model that removes the impact of small volumes of outliers in the clustering methodology.” Removing the impact of low-performing, low-volume outliers from the determination of cut points will have “the impact of raising the lower cut points, so there will be more plans who wind up with 1, 2 or 3-star ratings than there were before” on the non-CAHPS measures, she observes.
CMS suggested that use of the Tukey outlier methodology would “further increase the predictability and stability of the Star Ratings system” and result in a 10-year net savings to the Medicare Trust Fund of approximately $4.4 billion, largely from CMS making fewer Quality-Based Payments (QBPs) because of a reduction in star ratings.
Smith says the rule contained other significant proposals, but one that particularly stood out to Gorman Health Group is the potential addition of two Part C HEDIS measures: (1) Transitions of Care and (2) Follow-up after Emergency Department Visit (ED) for Patients with Multiple Chronic Conditions.
While the former was expected, Smith says the proposal to assess follow-up care provided after ED visits was “new news” and likely represents a step toward recommendations made by the Medicare Payment Advisory Commission last year to replace the current QBP-based program with a “value incentive program.” That program would rely on a small set of population-based outcome and patient experience measures as well as clear, prospectively determined performance standards to tie rewards (or penalties) to plan quality.
And in a discussion of potential new measure concepts related to the star ratings, CMS noted that the NCQA is working on a new kidney health evaluation measure that may in the future replace the Diabetes Care — Kidney Disease Monitoring measure in the Part C star ratings. CMS said it is seeking comments on end-stage renal disease measures.
Contact Smith at firstname.lastname@example.org.