In the most recent technical changes made to Part C and Part D plans for 2019, CMS codified the star ratings methodology in regulations. Now, CMS is proposing changes to these regulations, such as new definitions to clarify the meaning of terminology used in describing the star ratings methodology. In addition, CMS is proposing several changes to improve program quality and accessibility of the Medicare Advantage (MA) and Part D Prescription Drug Program (PDP) Plan Quality Rating for measures other than Consumer Assessment of Healthcare Providers and Systems (CAHPS).

Cut Point Predictability and Stability

CMS proposes two enhancements to the current hierarchical clustering methodology that is used to set “cut points” for non-CAHPS measures. Cut points are used to separate a distribution of measure scores into distinct groups or star categories, such that each grouping accurately reflects true performance. These two enhancements would increase year-over-year stability and predictability for a plan’s cut points, but may also slow the programs’ ability to keep pace with changes in performance across the industry.

First, CMS proposes to modify §§ 422.166(a)(2)(i) and 423.186(a)(2)(i) to add mean resampling to the current clustering algorithm. Mean resampling would reduce the sensitivity of the clustering algorithm to outliers. In short, it would reduce the random variation that contributes to fluctuations in cut points and, therefore, improve the stability of the cut points over time. The second proposed enhancement is a 5% “guardrail” for all measures that have been in the Part C and D Star Ratings program for more than 3 years, whether scored on a 0 to 100 scale or an alternative scale. The guardrail would be a cap that imposes a 5% maximum allowable movement from the previous year’s cut points for each measure threshold. Thus, it would allow a degree of predictability.

Measure Updates and Other Adjustments

CMS also proposes changes to measures in the Star Ratings program for performance periods beginning in 2020 and 2021, including:

  • MA plans controlling high blood pressure, to apply new hypertension treatment guidelines from the American College of Cardiology and American Heart Association by moving the blood pressure target to <140/90 mmHg;
  • Medicare plan finder (MPF) pricing accuracy for PDP plans, to better measure the reliability of sponsors’ prices advertised on MPF;
  • MA plan all-case readmissions, to assess the percentage of hospital discharges resulting in unplanned readmissions within 30 days of discharge; and
  • MA and PDP plan improvement measures, to exclude any measure that has a reduction due to data integrity concerns.

Among other things, CMS proposed to add a provision that would assign a 1-star rating to the applicable appeals measure(s) if an MA organization or PDP plan sponsor failed to submit Timeliness Monitoring Project (TMP) data for CMS’s review to ensure the completeness of their independent review entity (IRE) data.

CMS is also adopting detailed modifications for the Star Ratings measures for MA and PDP plans in areas that have “experienced extreme and uncontrollable circumstances” like natural disasters. These modifications would apply at the contract level to mitigate certain negative Star Ratings effects on any contract within an “emergency area” during an “emergency period” as defined by federal law.

In sum, CMS recognizes the public’s “overall support for the use of the hierarchical clustering algorithm,” and now seeks to further refine the Star Ratings program to ensure that the methodology is stable, predictable, and free from undue influence of outliers. This rulemaking presents a significant opportunity for stakeholders to shape the rating methodology for quality and performance of MA and Part D programs. Comments on the proposed rule are due to CMS no later than December 31, 2018.