Comments on the 2027 Medicare Advantage and Part D proposed rule

The Centers for Medicare and Medicaid Services (CMS) recently issued a proposed rule that would make changes to Medicare Advantage (MA) and Medicare Part D for the 2027 contract year.¹ Matthew Fiedler submitted a comment letter on CMS’ proposals.

The letter first addresses CMS’ proposal to broaden its authority to release MA encounter data, making three main points:

• Broadening CMS’ authority to release encounter data has the potential to facilitate research that would improve understanding of MA, at least if this authority is used effectively.
• Because CMS would continue to bar the release of dollar amounts reported on encounter records, it would remain difficult or impossible to use encounter data to answer a range of important questions about MA, including how plans pay providers, what plans spend on different types of care, and how much cost-sharing enrollees bear.
• CMS’ stated rationale for barring release of these data is that it wishes to keep the prices that MA plans negotiate with providers secret. But it is doubtful that disclosure would harm market outcomes, and, in any event, this approach runs directly contrary to the Trump administration’s stated policy of expanding price transparency.

Fiedler then examines several potential changes to risk adjustment policy that CMS raises in its request for information on strategies to improve risk adjustment, making the following points:

• Adopting an “inferred risk” model, under which enrollee health risk would be measured based solely on their health care utilization, could create incentives that would seriously distort care delivery. However, incorporating carefully selected utilization data alongside the diagnosis data currently used has clear potential to improve the risk adjustment system.
• Excluding diagnoses obtained from certain sources, such as health risk assessments and chart reviews, from risk adjustment calculations might reduce MA coding intensity and put plans on a more level playing field. However, plans would likely seek to capture these diagnoses in other ways, which could at least partly offset the reduction in coding intensity and create new inefficiencies, including wasted provider time and low-value utilization.
• Contrary to what CMS has suggested in the past, calibrating the CMS-HCC model using encounter data may not help address coding intensity and selection problems, but it might create legal obstacles to applying an appropriate coding intensity adjustment.
• While structural improvements to the risk adjustment model are important to explore, improving the processes that CMS uses to set coding intensity adjustments and creating a similar adjustment aimed at addressing favorable selection could also play an important role in improving the accuracy of the risk adjustment system.

Read the full comment letter here