Risk Adjustment Data Validation (RADV) audits have long had a reputation for creating worry and anxiety among Medicare Advantage organizations. The moment a plan learns it has been selected, fear quickly follows—fear of the unknown, fear of financial exposure, and fear of decisions made years ago that can no longer be changed. Even without finalized repayment demands, RADV audits introduce uncertainty around coding determinations, submission decisions, and how the Centers for Medicare & Medicaid Services (CMS) will ultimately calculate financial impact.
That anxiety has intensified in recent years as CMS moved toward formalizing extrapolation. While extrapolation has historically been discussed and modeled, it has not yet been enforced in completed RADV audits. The RADV Final Rule issued in 2023 was intended to define how extrapolation would be applied and was expected to be used for recent audit cycles.
However, that Final Rule was vacated by a federal court in September of 2025, and CMS has appealed the ruling. As a result, Medicare Advantage organizations have not yet been required to make extrapolated repayments. Still, audit findings remain, repayment liability has not disappeared, and many in the industry believe extrapolation, or some form of it, will return.
Against this backdrop of regulatory uncertainty, plans must focus on what can be controlled during a RADV audit. The original coding is complete. Past documentation decisions cannot be undone. What remains within a plan’s control is its submission strategy, understanding the universe of diagnoses that can be impacted, optimizing what is submitted, and using visibility into the medical record to protect revenue. In this context, audit preparation becomes a critical revenue protection function.
How extrapolation has historically worked
Under the vacated Final Rule, CMS relied on a statistically valid sample rather than reviewing the entire population. CMS calculated the average risk adjustment factor (RAF) change after medical record review, applied a confidence interval, and extrapolated that result across the full sampling frame.
In simple terms:
- A sample is audited
- The average RAF change is calculated
- A confidence level is applied
- The adjusted result is applied across the population
Although this methodology is currently in legal limbo, audit results are still calculated, and plans will ultimately be required to reconcile those findings under whatever framework is upheld.
A hypothetical—but sealistic—RADV Scenario
Consider a medium-sized Medicare Advantage plan preparing for a RADV audit. Prior to the audit, the average RAF score across the audited population is 2.78.
Scenario 1: No optimized submission strategy
In the first scenario, the plan approaches the audit using non-optimized workflows. The focus is primarily on defending what was originally submitted, without a clear strategy to optimize what can still be influenced. There is limited visibility into the full universe of diagnoses available for submission and no structured process to:
- Identify additional HCCs that were not being audited but could be submitted, or
- Surface supported higher-severity HCCs that could replace lower ones
- Prioritize submissions based on confidence or impact
After CMS medical record review, the average RAF score drops from 2.78 to 2.28—an 18 percent decrease, or 0.50 RAF points.
If extrapolation were applied using historical methodology and a 90 percent confidence interval, that level of RAF erosion could translate into an estimated repayment exposure exceeding $30 million. While extrapolation remains uncertain, this figure illustrates the magnitude of risk embedded in audit findings when submission strategy is not optimized.
Scenario 2: Optimized submission strategy with full visibility
Now consider the same plan, same audit, same records, but with a different approach.
Before final submission, the plan uses fit-for-purpose software to gain visibility into the diagnoses that can still be impacted. With a clear strategy, the plan can:
- Identify supported additional HCCs not being audited
- Surface valid higher-severity diagnoses
- Prioritize the strongest, most defensible records for submission
These additional and higher HCCs do not create new payment during RADV audits. Instead, they offset removals, reducing the net RAF loss.
In this scenario, the post-audit RAF drops from 2.78 to 2.53—a nine percent decrease, or 0.25 RAF points. Under historical extrapolation methods, that reduced RAF impact would result in an estimated repayment exposure of approximately $15 million, cutting potential liability roughly in half.
Same audit findings. Very different financial implications.
The difference between these two scenarios is not regulatory interpretation or audit timing. It is whether the plan treated RADV as an uncontrollable compliance event, or as a strategic exercise grounded in visibility, prioritization, and control over what could still be influenced.
Even in today’s uncertain environment, this distinction matters. When repayment methodologies are finalized, whether through reinstated extrapolation, revised rules, or settlement frameworks, the underlying audit results will still drive financial outcomes.
Why this matters now
With extrapolation vacated but widely expected to return in some form, plans face a familiar challenge: preparing for financial exposure without knowing the exact rules that will apply.
What is clear, however, is that:
- RAF erosion still creates liability
- Missed opportunities to identify and submit additional and higher HCCs compound risk
- Visibility and optimized submission strategy reduce exposure
Plans that invest now in understanding the universe of what they can control during a RADV audit will be better positioned regardless of how the regulatory landscape ultimately settles.
The bottom line
RADV audits are not just about what CMS invalidates; they are about what plans fail to review, prioritize, and submit in defense of their revenue.
Even as extrapolation remains uncertain, the math behind it remains instructive. When a few tenths of a RAF point can represent millions of dollars, optimizing submission strategy through visibility into additional and higher HCCs is not optional, it is foundational to audit readiness today and in whatever RADV framework comes next.
We have the tools to give you peace of mind, and we’d love to connect with you to share more about the Wolters Kluwer Risk Adjustment Solution, the Health Language Coder Workbench, and the capabilities of our Regulatory Audit Module.
Also, make sure to register to attend our upcoming RISE Webinar on March 10th titled, RADV Audit Playbook: Slam Dunk Strategies to Avoid Costly Turnovers. It’s going to be a great session!
About the authors
Melissa James
Melissa James, CPC, CPMA, CRC, risk adjustment SME, senior consultant, Wolters Kluwer, Health Language, supports the company's Health Language solutions with content maintenance. She has more than 20 years of health care experience in coding, risk adjustment, billing, physician and coder education, accounts receivable management, regulatory and compliance, and consulting. She received her associate degree from Pueblo Community College.
Kimberly Rykaczewski
Kimberly Rykaczewski, R.N., BSN, CPC, CRC, clinical content management–medical terminology & risk adjustment, Wolters Kluwer, Health Language, supports the company’s Health Language medical terminology solutions focusing on the standardization of medical terminologies to expedite data normalization to enhance health care system interoperability. Rykaczewski manages a team focused on providing data quality solutions by monitoring regulatory coding content and providing diagnostic and procedural mapping sets.
She has over 20 years of health care experience in the areas of nursing, case management, utilization review, overpayment recovery, vendor management, coding and billing, risk adjustment, and regulatory compliance. One of the biggest rewards of her profession is bringing providers and payors together to achieve regulatory and coding compliance. Rykaczewski holds a Bachelors of Science in Nursing and is an AAPC-certified professional and risk adjustment coder.
