The RADV Odyssey: Extrapolation Vacated

Last week, on September 25, 2025, Judge O’Connor in the U.S District Court for the Northern District of Texas ruled in favor of Humana (the Humana Decision) and vacated the Centers for Medicare & Medicaid Services (CMS) Medicare Advantage 2023 risk adjustment data validation (RADV) rule that allowed CMS to extrapolate its audit findings (RADV Extrapolation Rule). Vacating the RADV Extrapolation Rule raises significant questions and likely has material impacts for all Medicare Advantage organizations (MAOs), CMS, and possibly the government’s budget. This decision feels like it has been a very long time coming, and at the same time, the start of another long road.

How Did We Get Here?

• 2007 – CMS fully transitioned to using the Medicare Advantage risk adjustment payment model
• 2009 – CMS and the U.S. Department of Health & Human Services Office of Inspector General (OIG) both conducted risk adjustment audits of select MAOs for 2007 and published findings years later
• • CMS audited a total of 37 contracts, did not extrapolate, and identified $13.7 million in alleged overpayments/recoveries
  • OIG’s audits alleged that audited plans were overpaid by hundreds of millions of dollars based on extrapolation (see here for one example)
• 2010 – after OIG pushed to extrapolate in its risk adjustment audits, CMS proposed to extrapolate for all RADVs
• 2012 – CMS announced its RADV contract level audit error calculation methodology that affirmatively recognized the need for and application of a fee-for-service (FFS) adjuster
• 2014 – CMS’s proposed adjustments to risk adjustment regulations in its plan year 2015 proposed rules for Medicare Advantage and Part D, but did not address the FFS adjuster
• 2018 – CMS published its plan year 2020 proposed rules for Medicare Advantage and Part D and announced that it did not believe a FFS adjuster was necessary, contradicting its long-held position
• 2019 – CMS did not finalize the updated RADV rule but continued to consider it
• 2021 – CMS published a notice of a 1-year extension, giving it more time to analyze the rule
• 2022 – CMS again extended the timeline for review of the rule, pushing the deadline to February 1, 2023
• 2023 – CMS adopted the RADV Extrapolation Rule with no FFS adjuster and justified it with reasons that differ from those relied on in the 2018 rule
• 2025 – RADV Extrapolation Rule vacated

What’s Next?

CMS’s RADV Extrapolation Rule impacted audits of PY 2018 forward, which included both CMS- and OIG-initiated audits. CMS began auditing PY 2018 in November 2024. OIG initiated PY 2018 audits earlier and has finalized some audits for both PY 2018 and 2019 that applied extrapolation. Based on the Humana Decision, MAOs that were subject to such audits and took steps to repay extrapolated amounts did so based on a now vacated rule. Both CMS and OIG are in the middle of conducting audits: CMS-initiated audits for PY 2019 earlier this year, and OIG is conducting many targeted audits of more recent years.

CMS is faced with difficult options; it must first decide whether to appeal the Humana Decision and how to handle pending and future audits. If CMS decides to appeal, it might move forward with the pending audits and seek to recover the identified errors in the sample. But it has to consider whether conducting such audits is an efficient use of government resources if there is a strong chance that the recoveries will not include extrapolation.

Further, as we reported earlier this year, CMS announced drastic changes to how it conducts its RADV audits when it announced that beginning for plan year 2020, it would audit every Medicare Advantage contract every year, rather than just auditing the relatively small sample of contracts that it has historically audited. CMS set an aggressive timeline for itself when it announced that it would complete audits for PY 2020-2024 by early 2026. Prior to the Humana Decision, we expected CMS to announce this week that it was updating the RADV timeline for PY 2020-2024 because the agency has been delaying various PY 2019 RADV dates. With the Humana Decision, like with the pending audits, CMS now must contemplate whether conducting large scale audits of every single Medicare Advantage contract with no ability to extrapolate its findings is worth it.

CMS was expected to provide some insight into how the agency intends to react to its previously scheduled September 30 call with MAOs, but CMS canceled the call.

Humana’s current victory is a very clear reminder of the importance of the Administrative Procedure Act and the implications for agencies in a post-Loper world. If CMS does not appeal the decision, the agency will have to initiate a new rule-making and reintroduce and defend its position on how RADV audits should be conducted and how extrapolation can be applied. As evidenced by the greater than 12 years between CMS’s initial extrapolation proposal (December 2010) and CMS finalizing the RADV Extrapolation Rule (February 2023), extrapolation and actuarial equivalence are incredibly complex issues to try to appropriately address. We anticipate that CMS, MAOs, and highly qualified actuaries will have a lot of work to do in the coming years.

Subscribe To Viewpoints