Bottom line: CMS's CRUSH proposed rule (CMS-6098-P) is a proposal, and nothing in it is in effect. The RADV provision is the one with the most direct financial reach for Medicare Advantage organizations. It shifts the risk adjustment question from "did we capture the diagnosis?" toward "can we defend it across an entire contract?"
What CMS proposed
CMS released CRUSH (Comprehensive Regulations to Uncover Suspicious Healthcare) on November 16. It is a discretionary program integrity rule, separate from the annual payment rules, and CMS has said it plans similar rulemaking each year. CMS estimates the full rule would save nearly $90 billion over 10 years across Medicare, Medicaid, CHIP and the Marketplace. The 545-page proposal covers provider enrollment, MA and Part D oversight, Medicaid and CHIP eligibility, DMEPOS suppliers, and Marketplace agents and brokers.
This note focuses on the two provisions with the clearest reach into MA operations.
RADV extrapolation returns to rulemaking
For RADV, CMS proposes to:
- extrapolate audit findings, starting with payment year 2018 audits;
- codify that no fee-for-service (FFS) adjuster is applied; and
- determine the sampling methodology itself.
CMS estimates this would collect an average of about $9 billion per payment year, and ties the RADV reforms to an estimated $21 billion per year in MA overpayments from inflated risk scores. Both are CMS estimates, not outcomes.
The policy is not new. CMS finalized extrapolation and the absence of an FFS adjuster in its 2023 RADV rule. In September 2025, the U.S. District Court for the Northern District of Texas vacated that rule in Humana v. Becerra on Administrative Procedure Act grounds, without ruling on extrapolation or the adjuster themselves. The government appealed in November 2025. CRUSH sends the same policy back through notice and comment, which gives plans, actuaries and trade groups a formal opportunity to weigh in on the FFS adjuster question.
The audit program is also already operating at scale. CMS staff said in October that the PY 2024 audit covers 590 contracts, with PY 2023 and PY 2022 next.
Why this changes the documentation question
Under extrapolation, audit findings from a sample are applied across a full contract and back across several payment years. That makes contract-level error rates, not individual chart results, the number that matters. Three implications follow:
- Defensibility, not just capture. A diagnosis that is coded but not clearly supported in the medical record carries financial exposure far beyond the single chart.
- Provider contracts. Many risk-bearing provider contracts pass RADV adjustments downstream. The final provision would reach delegated groups, IPAs and MSOs, not only the plan.
- Hardship exceptions. CMS would write RADV hardship exceptions into regulation for the first time, defining the extraordinary circumstances that qualify a plan for relief from medical record submission. Today this lives only in subregulatory guidance.
Provider enrollment in Original Medicare
Proposed § 422.222(a) would require every eligible provider and supplier billing MA to be enrolled in Original Medicare in an approved status, with PACE excepted. That includes the full screening process: identity verification, licensure review, site visits and fingerprint-based background checks. The proposed start is the first day of the plan year beginning two years after the final rule is published. A final rule in mid-2027, for example, would put the start at January 1, 2030.
Providers already enrolled in an approved status see no change. The exposure sits with clinicians who see only MA members, or whose enrollment has lapsed or is pending. Network management and credentialing teams would add a federal enrollment check alongside existing credentialing.
Questions worth asking now
- What is our documentation support rate for high-impact HCCs, measured at the contract level rather than by chart?
- How would an extrapolated finding across several payment years affect our financial position?
- Where do RADV adjustments flow in our risk-bearing provider contracts, and do those contracts reflect it?
- How many contracted clinicians are not enrolled in Original Medicare in an approved status?
- Do we have a position on the FFS adjuster worth submitting as a comment?
Timeline
The 60-day comment period began with Federal Register publication, and CMS lists comments as due January 19, 2027. CMS will then review comments and issue a final rule on an unannounced schedule. The RADV, fraud referral and Medicaid reporting provisions would apply on finalization; the MA enrollment requirement starts later, as described above.
The comments filed over the next two months will show where the industry wants CMS to revisit its approach. The final rule will show how much of the proposal survives.
A note on approach
IQRP works with MA plans' quality, medical management and data teams on risk adjustment accuracy on a V28 basis. If it would be useful to talk through how these proposals map to your plan's documentation and contract structure, we are glad to set up an exploratory conversation at no cost and with no obligation.
This article is general information, not legal or compliance advice.
Sources
- CMS, Comprehensive Regulations to Uncover Suspicious Healthcare (CRUSH) Proposed Rule (CMS-6098-P) fact sheet (Nov. 16, 2026).
- CMS, CRUSH proposed rule, Federal Register public inspection document 2026-23461.
- HHS, CMS CRUSH proposed rule aims to reap nearly $90 billion for taxpayers in savings (Nov. 16, 2026).
- Crowell & Moring, Court vacates CMS's 2023 final rule on RADV audits (Sept. 26, 2025).
- Crowell & Moring, CMS appeals Humana v. Becerra (Nov. 2025).