CMS Releases PY 2020 RADV Audit Methods and Instructions: Key Takeaways for Health Plans

On March 20, 2026, the Centers for Medicare and Medicaid Services (CMS) released new guidance outlining the agency’s audit methods and instructions for Medicare Advantage (MA) plans subject to upcoming risk adjustment data validation (RADV) audits for payment year (PY) 2020. In addition to providing necessary context for MA plans selected for auditing, this resource clarifies CMS’s methodological and procedural expectations. While the high-level takeaways are recapped below for convenience, we strongly recommend that MA organizations selected for PY 2020 audits closely review the guidance to understand what may be involved — or required — during the agency’s review.

Understanding CMS’s Audit Approach for PY 2020

For PY 2020, CMS plans to focus its audit activity on MA enrollees considered most likely to have generated overpayments. To be eligible for selection, an enrollee must rank in the top quartile (as opposed to top decile, as for PY 2018 and PY 2019) of all program-wide RADV-eligible enrollees as scored by one or both of the predictive models used by CMS’s Center for Program Integrity (CPI) to identify enrollees most likely to see the greatest reduction in their risk score upon audit. The number of enrollees ultimately sampled from any given contract (notably, now 35, 50, 100, or 200) will be determined by the size of that contract’s sampling frame size.

Acknowledging that a 2023 Final Rule concerning the use of statistical sampling and extrapolation in MA contract-specific RADV audits was vacated by a federal district court in 2025 (a decision that CMS has appealed), CMS states that it has not decided whether results from each PY 2020 audit will be extrapolated. As explained in the guidance, the agency has “designed its PY 2020 MA contract-specific RADV audits to support the collection of extrapolated recoveries,” but has “not decided whether it will make extrapolated recoveries in its PY 2020 MA contract-specific RADV audits, or only recover the specific overpayments associated with the sampled enrollees” — a determination CMS indicates it will make at a later date. The lack of a decision is significant, as plans have been waiting to see whether extrapolation will potentially multiply audit recoveries from a sample-based figure to a potentially far larger contract-wide liability.

Deadlines and Next Steps for Medicare Organizations

MA organizations selected for audit should consider taking prompt action on two fronts:

1. 1. The Enrollee Data List (EDL) will become available in the Centralized Data Attestation Tool (CDAT) on April 3, 2026. Plans selected for audit should strongly consider retrieving this data as soon as possible to ensure they have ample time to gather and submit necessary medical records. The submission window opens on April 13, 2026, and closes on August 28, 2026.
   2. MA organizations should consider filing for hardship exceptions, if applicable, by the September 11, 2026, deadline, while recognizing that submission does not guarantee approval. This PY 2020 guidance expands on the hardship exception process, offering more specificity than in prior years, in Appendix B to the guidance. Appendix B includes a table identifying qualifying circumstances — such as a major system or technology failure, a natural disaster, or a declared emergency — along with the specific documentation required for each.

Our lawyers are available to assist MA organizations with questions about the PY 2020 RADV audits, including those regarding compliance counseling, medical record submission review, hardship exception requests, and appeal strategy. If you have any specific concerns, please reach out to any author of this alert or your preferred Crowell & Moring lawyer.