Health Law Daily Wrap Up, ADMINISTRATION OF MEDICARE/MEDICAID PROGRAMS—S.D. Ga.: Medicare Advantage Star Ratings are confined to HEDIS, HOS, and CAHPS data, (Jun 3, 2026)

By Justin Marcus Smith, J.D.

CMS did not have the discretion to draw from other data sources, but it could vary the survey questions asked within the statutorily permitted data collection systems, subject to notice-and-comment rulemaking.

The plain text of 42 U.S.C. § 1395w-23(o)(4)(A), when read in Medicare Act context, supported the conclusion that CMS may only base its Medicare Advantage Plan “Star Rating” quality rating calculations on measures using data collected under § 1395w-22(e), held the federal district court in Brunswick, Georgia. A plan provider that challenged its unexpectedly low Star Rating interpreted § 1395w-23(o)(4)(A) in line with limits on agency power, allowing CMS to consider the specific data Congress gave it the power to consider, without implying additional discretion. To comply with the statutory guardrails set by 42 U.S.C. § 1395w-22(e)(3)(B)(i), CMS systems must remain unchanged from those which were utilized as of November 1, 2003. The systems comprise only HEDIS (Healthcare Effectiveness Data and Information Set), HOS (Health Outcomes Survey), and CAHPS (Consumer Assessment of Healthcare Providers & Systems). The court held the overarching categories of questions within those systems must stay the same, but the exact survey questions asked from year to year can change within these confines. However, CMS failed to engage in required notice-and-comment rulemaking when it decided to include ten disputed measures in the 2026 Star Rating calculation. The manner in which CMS calculated the 2026 Star Rating for the Medicare Advantage plan provider was therefore procedurally invalid. The court accordingly granted in part and denied in part the provider’s motion for summary judgment, and it denied HHS’s cross-motion for summary judgment (Clover Ins. Co. v. HHS, No. 2:25-CV-142 (S.D. Ga. May 27, 2026)).

Background. Health insurance plan provider Clover Insurance Company sued HHS, HHS Secretary Robert F. Kennedy, Jr., and CMS Administrator Mehmet Oz (collectively, HHS) after CMS gave Clover an unexpectedly low 2026 Star Rating of 3.5.

CMS rates Medicare Advantage (Part C) health plans on a one-to-five-star scale purportedly meant to reflect plan quality. Clover is a Medicare Advantage Organization (MAO). MAOs like Clover contract with CMS and agree to offer coverage for a price lower than the CMS “benchmark” rate, defined as the maximum amount CMS is willing to pay to cover an average beneficiary in each county. The CMS Star Ratings system is a complicated weighted grading system for Medicare Advantage and Part D plans purportedly based on four kinds of data: data collected by health and drug plans; surveys of plan enrollees; data collected by CMS contractors; and administrative data collected by CMS. CMS displays the final star ratings in its online and print resources available to Medicare beneficiaries, including the “Plan Finder” tool. Higher-rated plans may attract more customers. The ratings also carry financial implications insofar as plans at a four-star rating or higher may have increased rebates.

Before initiating the instant suit, Clover objected to some of the measures CMS considered for the Star Rating calculation. During the “Second Plan Preview” period, CMS released preliminary Star Ratings and preliminary ratings for various individual measures. Clover contended that CMS erred in including a variety of measures, identified certain “unlawful” measures, and requested recalculation, but CMS rejected the request as “outside the scope of the plan review” process.

Clover filed the instant complaint on November 7, 2025. Clover contended CMS may only consider certain measures when it calculates Star Ratings and that it failed to follow the law in changing the measures. Clover alleged the unlawful use of erroneous measures caused Clover to receive a rating of only 3.5 Stars instead of 4 Stars, thereby causing it to suffer reputational damage and the loss of $120 million in “statutorily mandated quality bonus and related payments” which Clover believed it would have otherwise received. The six-count complaint alleged five counts for violations of the Administrative Procedure Act (APA) and one count for violation of the private non-delegation doctrine under Article II, Section 1 of the United States Constitution, and the Fifth Amendment.

HHS moved to dismiss or transfer the action under 28 U.S.C. § 1406 on the basis that the Southern District of Georgia was an improper venue, but the court denied transfer. Clover moved for expedited summary judgment and HHS filed a cross-motion for summary judgment.

Under the APA, a district court does not apply the Fed. R. Civ. P. 56 standard for summary judgment. It sits instead as an appellate tribunal and may hold unlawful and set aside agency action if it is arbitrary, capricious, an abuse of discretion, or otherwise not in accord with the law or statutory jurisdiction.

Clover contended CMS erroneously considered twenty different measures when calculating Clover’s 2026 Star Rating, including ten measures that purportedly did not derive from plan quality improvement program data. Measures challenged on that basis included things like medication adherence, appeals decisions, medication therapy management completion, and pharmacy statin use. Clover challenged all twenty measures for lack of compliance with APA notice and comment procedure.

Discretion distinguished. The court held that a plain reading of 42 U.S.C. § 1395w-23(o)(4)(A) favored Clover’s interpretation that HHS did not have the discretion to use data other than that specified for the Star Ratings.

Initially, the court found the administrative record supported Clover’s contention that each of the ten measures subject to the disputed data challenge derived from data beyond what is collected pursuant to statutory authority. HHS did not dispute (1) that the sources for 42 U.S.C. § 1395w-22(e) “quality improvement program” data are HEDIS, HOS, or CAHPS; and (2) that it determined Clover’s 2026 Star Rating by relying on some measures that used other data, for example, prescription drug event (PDE) data and “call center” monitoring data collected by CMS.

Turning to the primary dispute, the proper interpretation of 42 U.S.C. § 1395w-23(o)(4)(A), the court looked to its plain meaning. The words “based on” could have multiple meanings, depending on context. The parties heavily disputed the contextual impact of surrounding statutory interpretations.

The court concluded Congress made data exclusivity explicit in the context of the statutory provisions and additional modifiers. Section 1395w-23(o)(4)(A) created an HHS “obligation to consider certain data sets” for the Star Ratings, data collected under Section 1395w-22(e). A second conclusion, based on the statutory context, was that surrounding provisions used the words “based on” to communicate specific information which should factor into a particular calculation, as well as when Congress might allow discretion to consider other data. With the different uses of “based on” in mind, it was wrong to assume it gave CMS authority to consider other data deemed appropriate where there was no explicit grant of discretion. Congressional silence on additional discretion was controlling.

Remaining data challenges. As to remaining measures, like improving mental health, Clover challenged improper reliance on post-2003 data in the quality rating calculation and alleged CMS failed to engage in the required APA notice-and-comment process before considering them. The court held CMS was not precluded from considering certain measures in Clover’s 2026 Star Rating Calculation, but it did need to follow required notice-and-comment rulemaking to the extent the changes affected a standard governing payments to plan providers.

Clover essentially argued CMS did not have broad statutory discretion to add or subtract “whatever data collections it wishes” from the HEDIS, HOS, and CAHPS system-specific data collections administered as of November 1, 2003, under § 1395w-22(e). While CMS contended it had discretion to “add, delete, or modify” measures within those three systems, it did not agree with Clover’s contention that it could “jam” any data it wanted into those systems, because other entities have control over them and provide checks on CMS.

Starting again with the text, Congress did not define “types of data” to clarify what qualified for HHS collection as of November 1, 2003. The statute was ambiguous. The plain language definition could support either side’s interpretation, but the Medicare Modernization Act curtailed HHS data-collection flexibility with enactment of § 1395w-22(e)(3). Clover’s reading was more compelling. If the HHS Secretary wants to alter the data reporting responsibilities as the HEDIS, HOS, or CAHPS systems evolve or as other systems develop, it may have to report to Congress and consult with MAOs and private accreditation bodies before doing so.

CMS acknowledged the checks placed on its authority under the new statutory scheme; however, the court interpreted the limitation on the Secretary’s data collection power in a manner that provided more discretion than the interpretation posed by Clover, which would have limited the HHS Secretary to the 2003 survey questions. Limiting CMS to the HEDIS, HOS, and CAHPS systems was in line with the agency’s contemporary interpretation of Section 1395w-22(e)(3)(B) in 2004 and its interpretations in the two decades since then. Limiting CMS to particular categories of data within its existing systems would mitigate concerns about the burden of data collection on MAOs, as Clover highlighted with reference to the legislative history.

Although Congress confined CMS to using only those systems which were used as of November 1, 2003, comprising only HEDIS, HOS, and CAHPS, and overarching categories of questions within those systems must stay the same, the court held CMS could change the exact survey questions asked from year to year.

The limitation in § 1395w- 22(e)(3)(B)(i) did not preclude CMS from considering measures C15 (Reducing Falling) and C27 (Care Coordination) in Clover’s 2026 Star Rating calculation. C15 drew its data from HEDIS-HOS and tracked questions about falling problems. Essentially the same analysis applied to C27.

As for Clover’s notice-and-comment challenge, the CMS decision to include the ten remaining disputed measures in the 2026 Star Rating calculation triggered the notice-and-comment requirements of § 1395hh(a)(2). HHS did not dispute that the 2026 Star Rating calculations met the “minimum” set by the D.C. Circuit to qualify as a “substantive legal standard,” a standard that ultimately affects the scope of MAO legal rights to rebates by impacting the percent of per capita savings a plan can receive. Because CMS made a change that impacted MAO rebates and did not undertake the required notice-and-comment rulemaking therefore required, the manner in which it calculated the Clover 2026 Star Rating was procedurally invalid.

The case is No. 2:25-CV-142.

Judge: Wood, L.

Attorneys: Andrew D. Prins (Latham & Watkins LLP) for Clover Insurance Co. Otto Woelke Leithart, U.S. Attorney's Office, for U.S. Department of Health and Human Services.

Companies: Clover Insurance Co.

MainStory: TopStory CaseDecisions CMSNews GCNNews PartCNews PartDNews ProgramIntegrityNews GeorgiaNews