Last week, the Centers for Medicare and Medicaid Services (CMS) published the new star ratings for Medicare Advantage (MA) and Part D plans for plan year 2026. Due to the government shutdown, CMS has not done its customary rollout of the 2026 data but instead, simply released the ratings to the public in advance of the upcoming open enrollment window. Because star ratings are an influential consumer shopping tool and incentivize plan competition and service improvements, this Weekly Check-up highlights the ups and downs of star ratings before open enrollment begins. 

The Medicare star rating system is CMS’s 1-to-5-star scorecard for MA and Part D prescription-drug contracts. It rolls up dozens of quality measures into an overall score meant to help beneficiaries compare plans and to help CMS oversee plan performance. About 64 percent of MA enrollees are in 4–5-star contracts, broadly stable year over year. The average contract-level star rating is estimated to be 3.98 (up from 3.96 in 2025). Methodologically, 2026 reduces the weight of patient-experience/access measures from 4 to 2 and adds new measures (kidney evaluation, and physical and mental health status).

Notably, star ratings do more than inform consumers; they also drive money. Under the Affordable Care Act’s Quality Bonus Program, contracts with 4 stars or higher receive payment bonuses that plans can use to enhance benefits or lower premiums. Those dollars can be sizable. KFF estimated MA quality bonus payments will total at least $12.7 billion in 2025, with roughly three-quarters of those with MA enrolled in plans receiving a bonus.

Although a regular part of the quality evaluation program since 2015, the star ratings system is not universally beloved. Supporters argue these incentives promote competition on quality, not just price. Critics counter that the program is complex and not clearly linked to better outcomes.

There are many things that the system does well. For beneficiaries, a simple 1-to-5 label is easier to digest than raw measure tables. Ratings are easily accessible and refreshed each fall in alignment with the annual enrollment window so people can act on the information. Patients are often at the center of these assessments. CMS includes and focuses on patient experience measures about getting care, provider communication, and service. This recognizes how care is delivered matters alongside clinical process measures. This manifests in plan decision-making, as well. Because stars affect both reputation and revenue, plans invest in adherence programs, care management, and service redesign, all of which can support beneficiary care and access.

But not everything about the program is viewed positively. Ratings methodologies are still a contentious point for both plans and CMS. Star ratings are calculated at the contract level, which can span multiple, even non-contiguous markets and encompass very different plan designs and networks. Thus, a single contract score may not reflect the quality beneficiaries experience in their particular care setting. MedPAC, Urban Institute, and others have flagged this as a fundamental flaw. The program also includes many measures, and the performance thresholds (“cut points”) can move materially year to year. Methodology changes contributed to volatility and were the subject of litigation and policy reversals. Analysts note guardrails designed to limit big swings were not always applied as initially described for 2024, fueling disputes.

In short, star ratings remain an influential tool that blends consumer information with powerful financial incentives. Star ratings succeed at making quality visible and salient, rewarding plans that perform well on widely accepted clinical and experience metrics. The system’s emphasis on patient experience is directionally right: Enrollees care deeply about access and service, and those signals should influence plan competition. At the same time, the architecture of contract-level scoring and a sprawling measure set can blur what matters most to a specific beneficiary in a specific place. Volatile thresholds and periodic methodology pivots – however well-intended – introduce noise, and equity adjustments are still evolving.