On September 25, a federal district court vacated the Medicare Advantage risk adjustment and data validation (“RADV”) final rule that would have enabled the Centers for Medicare & Medicaid Services (“CMS”) to extrapolate RADV audit results across a contract’s entire population, which would have allowed CMS to recover substantial amounts. Humana Inc and Humana Benefit Plan of Texas, Inc. v. Becerra, No. 4:24-cv-00909-O (N.D. Tex. Sept. 25, 2025). For example, under the final rule, if CMS paid $1 billion under a Medicare Advantage contract, and CMS’s RADV audit validates 95% of hierarchical condition categories (“HCCs”) in a sample and applies the 5% error rates across the entire contract, CMS would have a $50,000,000 overpayment recovery. The court’s ruling is a significant blow to CMS’s ambitious plan to audit all Medicare Advantage contracts for payment years 2018 through 2024, which already faced challenges given CMS’s inability to hire sufficient medical coders to support the audits.
Medicare Advantage Risk Adjustment
Medicare Advantage Organizations (“MAOs”) must cover the same benefits that are available under Medicare fee-for-service. Unlike fee-for-service, however, which pays providers directly after they deliver services, CMS prospectively pays MAOs a monthly amount for each beneficiary enrolled. To account for this difference in payment, the Medicare statute requires that CMS adjusts payments to MAOs to account for the health status of enrolled beneficiaries via a risk adjustment process so that the payments to MAOs are actuarially equivalent to payments under fee-for-service Medicare. Risk adjustment modifies payments to MAOs using estimated healthcare costs based on risk factors, such as health status, age, and gender. MAOs report enrollees’ diagnosis codes that map to HCC and receive adjusted payments. Prior to the most recent RADV rule, CMS would then apply a fee-for-service adjuster (“FFS Adjuster”) to calculate a permissible payment error rate to account for the presence and impact of diagnosis errors in FFS claims data and limit RADV audit recoveries to errors above that level (and ensure that MAOs are not unfairly penalized for CMS’s more stringent analysis of their diagnosis data).
The Department of Health and Human Services Office of the Inspector General (“HHS-OIG”) and CMS audit MAOs’ risk adjustment data to confirm the accuracy and validity of the diagnosis codes submitted based on a sampling of enrollees and their medical records. Historically, these RADV audits targeted a subset of contracts with a limited set of records. CMS then would recoup an “overpayment” from an MAO based on payments made for diagnosis codes in the sample set that CMS determined are not supported in the medical record.
CMS Finalizes the RADV Final Rule, Litigation Ensues
In 2018, CMS proposed the RADV rule to eliminate the FFS Adjuster and allow extrapolation of RADV audit findings. More than four years later, in February 2023, CMS finalized the rule. Humana sued in September 2023, arguing that:
- The final rule is arbitrary and capricious because CMS abandoned the FFS Adjuster without adequate explanation;
- CMS abused its discretion to retroactively apply the final rule beginning in payment year 2018 because it misinterpreted the Medicare statute; and
- CMS promulgated the final rule without following appropriate procedures.
The court granted summary judgment for Humana and vacated the rule and remanded it back to CMS, finding that the final rule is procedurally invalid because it is not a logical outgrowth of the proposed rule. That is, the proposed rule did not enable MAOs to anticipate CMS’s final rule and could not address it during the notice-and-comment period. The proposed rule offered two justifications to eliminate the FFS Adjuster in extrapolation: (1) an empirical study; and (2) concern over potential inequities between audited and unaudited plans. But, the final rule justified itself using two different bases: (1) the FFS Adjuster is not required or appropriate in RADV audits because the Medicare statute’s actuarial equivalence requirement does not apply as a matter of law; and (2) the coding-intensity adjustment forecloses use of the FFA Adjuster.
The court rejected the coding-intensity adjustment rationale because CMS did “not and cannot assert that [it] logically flowed from the Proposed Rule’s justifications.” Instead, the court focused its analysis on the first rationale. Noting that the proposed rule’s concern about potential inequities does not appear anywhere in the final rule, the court explained that CMS’s “broad and affirmative reference to sub-sections of a statute did not satisfy their burden to notify the public with any reasonable specificity that they were considering finding actuarial equivalence inapplicable.” Because there was no meaningful notice of CMS’s determination that actuarial equivalence does not apply to RADV audits, there was no discussion of the costs and benefits of this unanticipated change. The court explained that the lack of discussion is an independent ground for vacatur and remand that is exacerbated by the rule’s retroactive application to 2018.
Application to Current RADV Audits and Next Steps
CMS started the PY2018 RADV audits in November 2024 targeting 60 2018 MA contracts for audit without an FFS Adjuster and subject to an extrapolation methodology consistent with the RADV final rule. CMS noted, however, that “in rare circumstances . . . [it] may use its discretion to not utilize extrapolation . . . .” As of June 2025, CMS anticipated “issuing PY 2018 audit findings in mid-calendar year 2026, including instructions on how the overpayments will be collected as part of the audit.” CMS may need to revise PY2018 audits that were finalized based on the RADV final rule to comply with the vacatur of the RADV final rule—though CMS may opt not to do so pending an appeal.
Likewise, CMS’s PY2019 RADV audits, which apply to approximately 550 MA contracts, already are well underway. CMS must ensure those audits comply with the prior version of the rule without extrapolation and with the use of the FFS Adjuster. This may slow CMS’s progress on the ongoing RADV audits and will affect how CMS proceeds with future audits. CMS also may decide to postpone some audits or modify its RADV audit strategy depending on whether CMS elects to appeal the ruling. CMS postponed a previously scheduled September 30 webinar with stakeholders to discuss the RADV audits to enable CMS to assess and respond to the court’s ruling.
CMS may appeal the ruling. Although remands to an agency typically are not appealable orders, the Fifth Circuit does allow an exception when the agency would not be able to later appeal the issue that is the subject of the remand order, such as when “all that is left for remand is a ministerial accounting . . . .” Adkins v. Silverman, 899 F.3d 395, 401 (5th Cir. 2018). Similarly, courts have concluded that remands are reviewable when the issues would be unreviewable as a practical matter if immediate appeal was not permitted. New Mexico Health Connections v. United States Department of Health & Human Services, 946 F.3d 1138, 1157-1158 & n.17 (10th Cir. 2019). Here, the Fifth Circuit could decide that remanding for a different rule (e.g., not allowing extrapolation) is effectively preventing CMS from appealing that issue. Alternatively, CMS also could engage in a new notice and comment rulemaking that responds to the deficiencies cited by the court.
Vacatur of the RADV final rule gives MAOs significant relief, but risk adjustment remains a significant focus by CMS, HHS-OIG, the DOJ, and potential False Claims Act relators. Regardless of whether CMS appeals or proceeds with a new rulemaking, MAOs should remain focused on ensuring that each diagnosis code is supported in the medical record, partner with providers to obtain complete and accurate records, and review their compliance and oversight programs.
Copyright © 2025 Groom Law Group, Chartered. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. With permission, excerpts and links may be used, provided that full and clear credit is given to Groom Law Group, Chartered and www.groom.com with appropriate and specific direction to the original content. For assistance, you may contact us.
