Revenue Optimization Without Defensibility Is a Risky Strategy
Risk adjustment has been one of the most reliable growth levers in Medicare Advantage (MA) for the better part of a decade. The logic was straightforward: higher risk scores resulted in higher capitated payments from the Centers for Medicare & Medicaid Services (CMS), and health plans invested heavily in retrospective chart reviews, in-home health risk assessments, and coding optimization programs to capture the full acuity of their member populations. The scale of MA enrollment is now massive, with 51% of the eligible Medicare population — a record 35.4 million people — now enrolled.¹
Multiple market trends and regulatory shifts have now converged to create a new state of operations. From the V28 risk model transition, flat payment updates, tightened enforcement, and the proposed exclusion of unlinked chart review records (CRRs) from risk scoring, the convergence of these factors represents a structural shift for health plan revenue. While the V28 risk adjustment model has begun to narrow the gap between MA payments and fee-for-service (FFS) spending every year since 2024, there is an immense amount of work still to do , with 2026 projections estimating $76 billion in excess payments, and MA payments reaching 114% of FFS spending.²
Enforcement Reality Is a Government Priority
Medicare Advantage is a priority enforcement area for the Department of Justice (DOJ) and the Department of Health and Human Services (HHS), spanning administrations. CMS Administrator Dr. Mehmet Oz stated during his Senate confirmation hearing that he would target upcoding, emphasizing bipartisan support for the effort.³
“[Upcoding and Medicare Fraud] is something that is addressable…it’s relatively enjoyable to go after, because I think we have bipartisan support.”
– Mehmet Oz, MD, Senate Finance Committee Hearing – March 14, 2025
The message was reinforced when DOJ and HHS relaunched the False Claims Act (FCA) Working Group on July 2, 2025, designating MA fraud as a priority area.⁴
Recent DOJ enforcement actions and OIG audit findings illustrate the scope and magnitude of exposure health plans face. In December 2024, Independent Health agreed to pay up to $98 million for allegedly submitting unsupported diagnosis codes that inflated risk scores.⁵ In March 2025, Seoul Medical Group and related parties settled for $62.85 million.⁶ Most recently, in March 2026, Aetna agreed to pay $17.7 million to resolve False Claims Act allegations specifically targeting its add-only chart review program — the exact type of retrospective coding practice this regulatory environment is systematically devaluing.⁷ For FY 2025 overall, the DOJ reported over $6.8 billion in FCA settlements and judgments, over 83% from the healthcare industry.⁸
OIG audits are compounding the enforcement picture. A March 2026 audit of Blue Cross Blue Shield of Alabama found a 91% error rate on sampled enrollee-years, with $7.06 million in estimated overpayments for plan years 2018–2019. The most common error: history-of conditions coded as active diagnoses, with acute stroke and myocardial infarction showing 100% error rates.⁹ A separate OIG audit of Gateway Health Plan, also published in March 2026, found an 81% error rate across 286 sampled enrollee-years and $4.3 million in estimated overpayments for the same period.¹⁰ These audits reveal systemic coding failures — and a growing audit apparatus prepared to find them.
The takeaway for health plans is clear: managing enforcement exposure is a proactive mandate, not a reactive one. Now is the time to invest in defensibility infrastructure — before, not after, an audit.
Redefining the CFO Mandate
What does this convergence of market forces mean for CFOs and their technology investment strategies?
Health plan margins across the MA sector are tight, and these pressures are compounding, not easing. The V28 risk adjustment model reached full phase-in for 2026, with CMS now calculating 100% of risk scores using only the updated CMS-HCC model.¹¹ Looking forward, CMS’s CY2027 Advance Notice projects an effectively flat payment update for MA at just 0.09% year-over-year, with CMS stating its goals of “strengthening payment accuracy” and “modernizing risk adjustment.”¹² The notice also proposes excluding diagnoses reported through unlinked chart review records from risk score calculations — a direct signal that documentation intensity without encounter linkage is over. This further intensifies near-term pressure on health plans.
Defensible revenue, not maximum revenue, is the metric that protects financial performance moving forward. The CFO mandate in this environment is accuracy and audit readiness — and investing in the technology infrastructure to enable it.
Building Defensibility Through Technology
MA financial defensibility in this market state requires a fundamentally different technology approach to risk adjustment. AI is the second-largest technology area, behind only cloud infrastructure, that healthcare organizations are investing in to support meeting their business objectives.¹³ According to Frost & Sullivan’s AI Maturity Index, AI is no longer exploratory in the healthcare vertical¹⁴; it is operational. But the architecture of the AI matters as much as the decision to adopt it. The market is ready.
An emerging technology innovation category, Neuro-symbolic AI is purpose-built to meet the defensibility requirements that health plans face. Neuro-symbolic AI combines the intuitive comprehension power of neural networks, which excel at reading and interpreting unstructured clinical text, with the structured reasoning of symbolic AI systems that encode medical knowledge, coding guidelines, and regulatory rules into traceable logic. This hybrid approach replicates a more human-like thought process, with both abstract reasoning and rule-governed outputs, along with an auditable chain of reasoning.¹⁵
According to Frost & Sullivan research,RAAPID, an AI-powered risk adjustment company focused on risk adjustment, exemplifies this approach. The company’s Neuro-symbolic Clinical AI platform provides the “why” and “where” behind every code, creating an audit trail that proves a diagnosis is real and defensible. Built on a clinical knowledge graph encompassing more than 4 million clinical entities and over 50 million relationships, the platform grounds every HCC recommendation in MEAT (Monitoring, Evaluation, Assessment, Treatment)-based clinical evidence and produces audit-ready documentation. This is the technology approach the current market demands: AI designed not to inflate risk scores, but to defend revenue by proving it.
The Last Word
Revenue resilience equals defensible revenue. The MA financial model has changed. Flat payment updates, tightened risk adjustment models, aggressive enforcement, and evolving regulatory proposals are converging on a single reality. Health plans require the right enabling technology and tools to drive, defend, and future-proof their business in this environment.
Those who don’t invest in defensibility are increasing their risk of exposure. Those who do are building resilience.
In the next installment of this blog series, we examine why generative AI needs symbolic guardrails for defensible risk adjustment.
Endnotes
[1] HealthScape Advisors A Chartis Company (February 2026). 2026 Medicare Advantage Enrollment Report. Last accessed March 2026. https://www.healthscape.com/insights/medicare-advantage-enrollment-depicts-industry-crossroads
[2] MedPAC (2026). “The Medicare Advantage Program Status Report.” January 16, 2026. https://www.medpac.gov/wp-content/uploads/2026/01/Tab-N-MA_Status-Jan-2026.pdf
[3] Office of Senator Elizabeth Warren, D. Mass. (2025). “Dr. Oz Agrees with Sen. Warren: Cracking Down on Private Health Insurers in Medicare Advantage Will “Improve the Health Care of the American People.” March 14, 2025. https://www.warren.senate.gov/newsroom/press-releases/dr-oz-agrees-with-sen-warren-cracking-down-on-private-health-insurers-in-medicare-advantage-will-improve-the-health-care-of-the-american-people
[4] US Department of Health and Human Services (2025). DOJ-HHS False Claims Act Working Group Press Release, July 2, 2025. https://www.hhs.gov/press-room/hhs-doj-false-claims-act-working-group.html
[5] US Department of Justice (2024). “Medicare Advantage provider Independent Health to pay up to $98M to settle False Claims Act suit.” December 20, 2024. https://www.justice.gov/usao-wdny/pr/medicare-advantage-provider-independent-health-pay-98m-settle-false-claims-act-suit
[6] US Department of Justice (2025). “Medicare Advantage Provider Seoul Medical Group and Related Parties to Pay over $62M to Settle False Claims Act Suit.” March 26, 2025. https://www.justice.gov/opa/pr/medicare-advantage-provider-seoul-medical-group-and-related-parties-pay-over-62m-settle
[7] US Department of Justice (2026). “Aetna Agrees to Pay $17.7 Million to Resolve False Claims Act Allegations.” March 11, 2026. https://www.justice.gov/opa/pr/aetna-agrees-pay-1177-million-resolve-false-claims-act-allegations
[8] US Department of Justice (2026). “False Claims Act Settlements and Judgments Exceed $6.8B in Fiscal Year 2025.” January 16, 2026. https://www.justice.gov/opa/pr/false-claims-act-settlements-and-judgments-exceed-68b-fiscal-year-2025
[9] HHS OIG (2026). Audit Report A-07-22-01207: “Medicare Advantage Compliance Audit of Specific Diagnosis Codes That Blue Cross and Blue Shield of Alabama (Contract H0104) Submitted to CMS.” March 12, 2026. https://oig.hhs.gov/reports/all/2026/medicare-advantage-compliance-audit-of-specific-diagnosis-codes-that-blue-cross-and-blue-shield-of-alabama-contract-h0104-submitted-to-cms/
[10] HHS OIG (2026). Audit Report A-03-22-00004: “Medicare Advantage Compliance Audit of Specific Diagnosis Codes That Gateway Health Plan, Inc., (Contract H5932) Submitted to CMS.” March 16, 2026. https://oig.hhs.gov/reports/all/2026/medicare-advantage-compliance-audit-of-specific-diagnosis-codes-that-gateway-health-plan-inc-contract-h5932-submitted-to-cms/
[11] CMS (2025). “2026 Medicare Advantage and Part D Rate Announcement.” Fact Sheet, April 7, 2025. https://www.cms.gov/newsroom/fact-sheets/2026-medicare-advantage-and-part-d-rate-announcement
[12] CMS (2026). “CY 2027 Advance Notice of Methodological Changes for Medicare Advantage Capitation Rates and Part C and Part D Payment Policies.” January 2026. As reported by AJMC, at https://www.ajmc.com/view/cms-proposes-nearly-flat-2027-medicare-advantage-payment-rates
[13] Frost & Sullivan (2025). Voice of Customer — Healthcare Cloud User Survey. Report KBEB. Frost & Sullivan. San Antonio, TX.
[14] Frost & Sullivan (2024). Global AI Maturity Survey. Report PFFE-69. Frost & Sullivan. San Antonio, TX.
[15] Prenosil, G.A., et al. (2025). “Neuro-symbolic AI for Auditable Cognitive Information Extraction from Medical Reports.” Communications Medicine 5, 491. November 21, 2025. https://www.nature.com/articles/s43856-025-01194-x
