More than four years after it was proposed, on February 1, 2023 the Centers for Medicare & Medicaid Services (“CMS”) published the long-awaited risk adjustment data validation (“RADV”) Final Rule (“Final Rule”) that will affect CMS’s payments to Medicare Advantage organizations (“MAOs”) and likely provoke litigation challenging the Final Rule. The Final Rule is effective on April 3, 2023 and makes two primary changes:

  1. CMS will not apply a fee-for-service (“FFS”) adjuster to RADV audit findings. Announced back in the 2012 RADV Payment Error Calculation Methodology, the FFS Adjuster was intended to account for FFS payment errors and create a baseline level of payment errors for MAOs. RADV audit recoveries would apply to errors above that baseline. The linchpin for this rationale is the statutory requirement that payments to MAOs must be “actuarially equivalent” to payments made under Medicare FFS for beneficiaries with similar demographic and health status. Thus, because Medicare FFS payments have some level of errors, MAOs should not be held to a stricter accuracy standard than Medicare FFS. Recent case law interpreting the actuarial equivalence provision, however, held that it does not apply to an MAO’s obligations to report and return improper payments, such as those identified in a RADV audit. CMS, therefore, determined that “it would not be reasonable” to require a statutorily-set minimum reduction in payments to MAOs “while at the same time prohibiting CMS from enforcing longstanding documentation requirements by requiring an offset to the recovery amounts calculated for CMS audits.”
  2. CMS may extrapolate audit findings to all RADV-eligible beneficiaries in the MAO’s contract, beginning with payment year (“PY”) 2018. The Final Rule did not state that extrapolation is mandatory or would occur automatically. Instead, the rulemaking finalized § 422.311(a)(2) to state “CMS may [emphasis added] apply extrapolation to audits for payment year 2018 and subsequent payment years . . . .” CMS explained that extrapolation will be the standard practice, but CMS may use its discretion not to extrapolate in certain circumstances, such as if medical records have been lost because of a natural disaster or if CMS performs a probe sample of RADV reviews without a statistically valid sample. Likewise, CMS noted that OIG may decide not to extrapolate for its own reasons. Notably, the Final Rule effectively waives extrapolation of RADV audit amounts for PY 2011 – 2017, so MAOs may be able to release any liabilities for those amounts.

CMS estimates that RADV audits will recover approximately $4.7 billion from MAOs between 2023 – 2032. CMS will specify how MAOs must remit improper payments identified as a result of a RADV audit to CMS, which may include offsets to the MAO’s monthly capitation payment (as noted in the proposed rule) or lump-sum payments.

Background on MA Risk Adjustment and RADV Audits

CMS’s payments to MAOs vary based on their enrollees’ demographics and health status. The MA risk adjustment process is prospective, so that the demographics and conditions of beneficiaries in year 1 are used to risk adjust the payments to plans for year 2. Diagnosis codes must be submitted annually, even for chronic conditions that are ongoing. The risk score calculation is an additive process; as hierarchical condition category (“HCC”) factors—and, when applicable, disease interaction factors—accumulate, an enrollee’s risk score increases, and the monthly risk-adjusted payment to the MAO also increases.

MAOs must ensure that the information used to calculate a beneficiary’s risk score are accurate. If the medical records do not substantiate the diagnosis codes and an MAO submits them to CMS, the HCCs are unvalidated and, therefore, the enrollee’s risk score is inflated. This results in an overpayment to the MAO. Conversely, if the medical records support diagnosis codes that the MAO did not submit to CMS, the MAO would receive an underpayment because the validated HCCs would not have been considered in the enrollee’s risk score.

CMS’s RADV audits review medical records for up to 201 beneficiaries per contract. The audits examine whether the diagnosis codes submitted by the MAO to CMS for those beneficiaries are supported by the medical records. If the medical records do not support the diagnosis codes, the MAO must repay funds to CMS.

The FFS Adjuster and Actuarial Equivalence

CMS’s 2012 RADV Payment Error Calculation Methodology announced that CMS would apply an FFS Adjuster to account “for the fact that the documentation standard used in RADV audits to determine a contract’s payment error (medical records) is different from the documentation standard used to develop the Part C risk-adjustment model (FFS claims).” The FFS Adjuster was intended to ensure actuarial equivalence between FFS and MA payments by establishing a baseline payment error rate for all of Medicare. MAOs would be responsible for errors above that level.

Subsequently, CMS studied “the presence and impact of diagnosis error in FFS claim data” and determined that those errors “do not have any systematic effect” on the risk scores calculated in the risk adjustment model and, therefore, no effect on payments to MAOs. The proposed rule issued in 2018 relied on this analysis to justify removal of the FFS adjuster. It also invited comment on the study. CMS received comments critical of the study and the potential elimination of the FFS adjuster.

Prior to publication of the Final Rule, the D.C. Circuit ruled in UnitedHealthcare Ins. Co. v. Becerra that the actuarial equivalence and “same methodology” provisions in the Social Security Act “do not apply to the obligation to return improper payments for MAO diagnosis codes that are unsupported by medical records.” 16 F.4th 867 (D.C. Cir. Aug. 13, 2021, reissued Nov. 1, 2021), cert. denied, 142 S. Ct. 2851 (U.S. June 21, 2022). Although that case concerned CMS’s overpayment rule at 42 C.F.R. § 422.326, CMS cites that ruling as part of its rationale not to apply an FFS adjuster. CMS explained that the actuarial equivalence requirement only applies to how CMS risk adjusts payments to MAOs, not an MAO’s obligation to return improper payments for diagnoses lacking medical record support. CMS said that the D.C. Circuit’s holding supports this conclusion: “The role of the actuarial-equivalence provision is to require CMS to model a demographically and medically analogous beneficiary population in traditional Medicare to determine the prospective lump-sum payments to [MAOs].” Id. at 870.

GROOM INSIGHT: MAOs should evaluate their potential exposure for erroneous risk score submissions given that there will be no baseline amount of erroneous submissions permitted under the Final Rule.  The impact of eliminating the FFS adjuster becomes more pronounced in the context of an extrapolation by CMS for PY2018 and later which will now be permissive under the Final Rule, where inaccuracies in the sample can be amplified in a way that does not reflect the larger data submission by the MAO to CMS.  Particularly for earlier payment years covered by the RADV audit process, the difficulty associated with adequately documenting risk scores with medical records may limit the ability of MAOs to justify their submitted risk scores, and thus the lack of an FFS adjuster means that the impact could be larger the further back in time that CMS audits the risk scores.

Extrapolation and Retroactive Rulemaking

Under the Final Rule, CMS may extrapolate RADV audit findings to determine the amount of overpayments recoverable from an MAO for PY 2018 and later. The proposed rule indicated that CMS would extrapolate audit findings from PY 2011 onward. Instead, under the Final Rule, CMS only will collect the non-extrapolated overpayments identified in the RADV audits and HHS-OIG audits for PY 2011 – 2017.

Perhaps anticipating potential legal challenges, the Final Rule explained that the rulemaking does not impose any new requirements on MAOs—the same documentation standards apply:

We are not imposing additional liabilities, penalties or retroactive application of new requirements or policy. We only seek to recover improper payments received by MAOs for HCCs that are not substantiated by enrollees’ medical records.

[. . .]

We emphasize that nothing in this rule changes the longstanding principle that a diagnosis code that is not documented in a patient’s medical record is not a valid basis for CMS risk adjustment payments to an MAO. Nor does this rule change the longstanding obligation of an insurer to refund payments to CMS if it learns that a diagnosis lacks support in the beneficiary’s medical record.

And, even if the Final Rule does constitute retroactive rulemaking (a point CMS does not concede), CMS has express statutory authorization to apply regulations retroactively when it is necessary to do so to comply with statutory requirements and it is in the public interest.

Despite this statutory authority, CMS decided not to pursue extrapolation prior to PY 2018 based on practical considerations. Namely, by not extrapolating prior to PY 2018, CMS expects to limit the number of active appeals submitted in the first few years following the Final Rule. This should reduce the burden on both CMS and MAOs. And it will enable CMS to finalize PY 2011 – 2017 audits relatively quickly, as CMS pursues enrollee-level improper payments from CMS and OIG audits. CMS will notify MAOs of these enrollee-level audit findings and recovering payments after the Final Rule’s April 3, 2023 effective date.

GROOM INSIGHT: Consistent with a general trend evidenced in litigation, the extrapolation rule places added pressure on MAOs to fully and accurately document medical record support for HCC submissions.  Failure to do so can expose the MAO to disproportionately harsh audit results if the sample extrapolated across the contract does not represent the accuracy levels of the contract as a whole.  While these issues are likely most prevalent in earlier years of the extrapolation period, they will continue to have regulatory compliance and financial implications for MAOs in the future.

Audit Methodologies

CMS declined to adopt any specific sampling or extrapolation audit methodology. Instead, the Final Rule announced that CMS “will rely on any statistically valid method for sampling and extrapolation that is determined to be well-suited to a particular audit.” CMS will disclose its extrapolation methodology via HPMS memos and other guidance.

GROOM INSIGHT: While CMS did not adopt a rigid rule here, MAOs should evaluate the actuarial soundness of any sampling and extrapolation methodology published by CMS. This review will ensure that CMS does not miss any unique actuarial characteristics of the underlying MAO’s business that could render the chosen methodology unreasonable. This will require enhanced engagement with the RADV audit process by different levels of the internal business experts at the MAO.

Copyright © 2023 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 with appropriate and specific direction to the original content. For assistance, you may contact us.


Latest Resources


April 20, 2023

Groom Event


March 23, 2023

George Washington University Law School