Glossary
RADV audit
A CMS review of whether the diagnoses a Medicare Advantage plan reported are supported by medical records, used to recover payments that were too high.
Also called: Risk Adjustment Data Validation, Medicare Advantage risk adjustment audit, RADV
Medicare pays a Medicare Advantage plan a monthly amount for each member, adjusted upward for members who are sicker. The adjustment relies on diagnosis codes the plan submits, a process called risk adjustment. A Risk Adjustment Data Validation audit checks, in a sample of members, whether the medical records actually support those diagnoses.
If a diagnosis is not supported, the plan was overpaid for that member. Say a plan is paid an extra $2,000 for a member's reported condition that the record does not support (a labeled hypothetical): CMS can seek that amount back, and may extrapolate the sample's error rate across the whole contract. The extrapolation approach has been the subject of dispute and court challenges by plans.
RADV audits are the main tool CMS uses to police risk-adjustment accuracy, and results feed the broader count of improper payment in Medicare Advantage. Findings of intentional upcoding can also lead to cases under the False Claims Act.
For members, benefits do not change because of an audit. It is a dispute between CMS and the plan about payments. The common mistake is treating an audit finding as proof of fraud; most are about documentation.
Last reviewed November 16, 2026