Payment year 2026 is the year the CMS-HCC transition ends. In its CY 2026 Rate Announcement on April 7, 2025, CMS confirmed that Medicare Advantage risk scores for non-PACE organizations will be calculated using the 2024 CMS-HCC model (commonly called V28) at 100 percent, completing the three-year phase-in that began in 2024. Because 2026 payments are based largely on 2025 dates of service, the documentation being produced in exam rooms right now is what will drive next year's revenue. For organizations that have been leaning on the partial V24 blend to soften the impact, the cushion is gone.

How the Phase-In Worked

CMS finalized the new model in its 2024 Rate Announcement and chose to phase it in rather than apply it all at once:

  • Payment year 2024: 67 percent V24, 33 percent V28.
  • Payment year 2025: 33 percent V24, 67 percent V28.
  • Payment year 2026: 100 percent V28.

CMS projected an average MA revenue increase of 5.06 percent for 2026, but that headline figure includes the combined effect of the risk model revision and fee-for-service normalization, which CMS estimated at negative 3.01 percent. The aggregate number masks wide variation. A plan or provider group whose population leans heavily on conditions that V28 dropped or devalued will see a very different result from the average.

What Changed in the Model

V28 is not a recalibration of V24. It is a restructured model built on ICD-10 rather than legacy groupings, with 115 payment HCCs compared with 86 in V24. More than 2,000 ICD-10 codes that mapped to payment HCCs under V24 no longer do, while a smaller number of codes were added. CMS described the removals as targeting conditions where coding was more discretionary and where MA coding intensity outpaced fee-for-service.

Conditions that lost payment weight

Several common diagnoses no longer map to a payment HCC, including:

  • Protein-calorie malnutrition.
  • Angina pectoris, with unstable angina remaining in the model.
  • Atherosclerosis of the extremities without more severe manifestations such as ulceration or gangrene.
  • Acute kidney failure.
  • Major depressive disorder documented as mild or unspecified. Moderate and severe major depression remain.

Conditions where specificity now matters more

Some categories were split to reflect clinical severity more precisely. Heart failure is the clearest example: V28 separates heart failure into several categories based on acuity and severity, so documentation of type and status carries more weight than it did. Chronic kidney disease, vascular disease and several other families were also reorganized. At the same time, diabetes categories remain constrained to a common coefficient, so documenting complications is essential for clinical accuracy but does not change the payment weight among those categories.

Conditions that were added

V28 added categories for several conditions, including severe persistent asthma, certain liver and biliary conditions, and eating disorders such as anorexia and bulimia nervosa. For most populations these additions do not offset the removals, but they are worth capturing accurately where they apply.

Where RAF Moves

The impact varies by population, but a few patterns are consistent:

  • Populations with significant vascular disease often see the largest declines, because peripheral atherosclerosis and stable angina are common and no longer risk adjust.
  • Behavioral health populations are affected where depression has historically been documented without severity.
  • Heart failure populations can hold or improve scores where documentation is specific, and lose ground where it is generic.
  • Frail and complex members may move in either direction depending on how fully their conditions are documented.

The only reliable way to know your exposure is to score your own population under V28 at the member level. Contract-level estimates hide the variation that matters for provider-level performance and for downstream risk arrangements.

How to Respond Operationally

Organizations that have handled the transition well share a few practices.

  1. Rescore at the member level. Run every attributed member through V28 and compare with blended scores. Identify which providers, panels and conditions drive the change.
  2. Retire outdated suspect logic. Pre-visit prompts built for V24 can steer clinicians toward conditions that no longer carry payment weight while missing the specificity V28 rewards. Rebuild suspect lists around V28 categories.
  3. Focus education on specificity. Clinicians do not need a lecture on HCCs. They need a few concrete reminders: heart failure type and acuity, depression severity and episode, CKD stage, the manifestations that make vascular disease clinically significant.
  4. Do not chase dropped conditions. Removed conditions still matter clinically and should be documented when present. They simply should not be the focus of revenue-driven outreach.
  5. Stress-test contracts. For MSOs and ACOs in shared-risk arrangements, model how full V28 affects benchmarks, surplus and deficits, and whether contract terms anticipated it.
  6. Pair accuracy with defensibility. CMS's May 2025 decision to audit every eligible MA contract means V28 scores will be validated more often and on larger samples. Specific documentation that is unsupported by evidence of management is no better than vague documentation.

Keep the Clinical Lens

V28's emphasis on specificity is, on balance, aligned with good medicine. Knowing whether a patient has heart failure with reduced or preserved ejection fraction, or whether depression is moderate or severe, changes treatment. Organizations that frame the transition as a clinical documentation improvement effort rather than a coding exercise get better buy-in from clinicians and better notes for care management.

Practical Support

Curitics scores CMS-HCC V28 RAF at the patient level so teams can see exactly where scores move and why, and its AI chart classification separates documented from inferred HCCs for coder review. Combined with prospective and concurrent coding review that routes specificity gaps to clinicians as provider queries, that lets organizations focus effort where V28 actually pays and where documentation can be defended.

The Bottom Line

The transition years gave organizations time to adapt. For dates of service in 2025 that time is now being used up, and in 2026 the V24 safety net disappears entirely. Plans and provider groups that have rescored their populations, rebuilt their suspect logic and trained for specificity will see a predictable result. Those that have not will discover the impact in their first full V28 reconciliation.