Overview
Risk Adjustment Data Validation (RADV) is the CMS-sponsored audit program that validates the diagnoses Medicare Advantage plans submit to CMS for risk-adjusted payment. Contractors sample member charts, request source documentation, and evaluate whether each diagnosis reported against a CMS-HCC category maps to MEAT-compliant clinical evidence. Unsupported diagnoses generate recoupment of the portion of the per-member payment attributable to those HCCs.
RADV has existed since Medicare Advantage's predecessor Medicare+Choice program, but the audits were historically limited in scope and did not include extrapolation. The 2023 CMS RADV Final Rule fundamentally changed the financial calculus. For payment years 2018 and forward, CMS now extrapolates sample-level error rates across the entire audited contract. An audited contract with a 5% HCC error rate in its sampled charts faces recoupment on approximately 5% of its CMS-HCC risk-adjusted payments for that year — a dollar amount typically in the tens to hundreds of millions for a mid-sized plan.
The sampling methodology pulls 200 or more members per contract, stratified by risk-score deciles. For each sampled member, the contractor requests charts covering every encounter from which an HCC was reported. Plans have limited windows to produce documentation — typically 6 weeks — after which missing charts count as errors. Audit adjudication includes an informal reconsideration, formal hearing, and potential Administrator Review; appeals can extend the final determination timeline by 2+ years.
Plans invest in RADV preparedness through several programs: pre-submission MEAT validation (CRC review before claim submission), retrospective chart review that purges unsupported HCCs before annual submission deadline, provider documentation education (especially for chronic conditions frequently under-supported like CKD staging, morbid obesity, and major depression), and RADV-simulation exercises that sample the plan's own roster and test documentation sufficiency before CMS audits.
OIG audits are a parallel oversight mechanism with similar mechanics but different sampling and appeal structures. OIG audits often target specific condition categories (cancer, heart failure, vascular disease) because OIG analytics flag them as commonly-unsupported. OIG findings are advisory to CMS, which can then initiate formal RADV recoupment.
The False Claims Act exposure layer is material. Submitting diagnoses a plan knows to be unsupported can trigger FCA liability with treble damages. Several high-profile DOJ settlements — UnitedHealth's $3.7B proposed FCA case, Kaiser's $90M+ settlement, multiple smaller plan settlements — reference RADV-style documentation gaps. Provider groups in delegated risk arrangements carry parallel exposure.
The 2023 Final Rule also eliminated the longstanding "FFS adjuster" that previously offset RADV errors against the baseline FFS coding error rate. This change materially raised recoupment severity and is the subject of ongoing litigation by industry plaintiffs.
Industry benchmark
CMS estimates aggregate RADV recoupment of ~$4.7B annually once 2018-forward audits are complete. Plan-level exposure varies widely; plans with MEAT-disciplined coding typically face <3% error rates, while plans relying heavily on retrospective chart review without MEAT validation face 8–15% error rates in audit samples.
Worked example
A Medicare Advantage contract reports $750M in CMS-HCC risk-adjusted payments for payment year 2022. A RADV audit of 250 sampled members finds a 7% HCC error rate. Under extrapolation, recoupment is approximately 0.07 × $750M = $52.5M, minus any successful appeal reductions. The plan contests 40% of errors at reconsideration; final recoupment settles at about $32M over the 24-month adjudication cycle.
Frequently asked questions — Risk Adjustment Data Validation (RADV) Audit
What is extrapolation in a RADV audit?
Extrapolation applies the sample error rate to the contract's full risk-adjusted payment pool. A 5% sample error rate means ~5% of the entire payment is recouped. Pre-2018 audits did not extrapolate; the 2023 Final Rule made extrapolation the default for 2018-forward audits.
How often are RADV audits conducted?
CMS targets approximately 30 MA contracts per payment year for formal RADV; additional contracts may be targeted via OIG referral. Every MA contract faces non-zero probability of selection; large contracts and contracts with anomalous coding patterns are at higher risk.
Can providers in risk-sharing arrangements be liable for RADV errors?
Yes, in delegated or downstream risk arrangements the provider organization typically bears a contractually specified share of recoupment. Provider-side CRC programs have become a core risk-management control for groups in MA-PD shared-risk contracts.
What is the 'FFS adjuster' and why does it matter?
The FFS adjuster was a longstanding offset that reduced RADV recoupment by the estimated fee-for-service Medicare baseline coding error rate. CMS eliminated the FFS adjuster in the 2023 Final Rule, materially increasing plan recoupment. Industry litigation on the change is ongoing.
Disclaimer
This glossary entry is operational reference for revenue-cycle and medical-billing professionals. It is not legal, clinical, or contractual advice. Industry benchmarks cite named public sources where available; always verify against the current guidance from the authority body before relying on a number in a contract, policy, or compliance filing.