CMS Kept V28 for 2027 — and That’s Not the Relief It Sounds Like

In the CY2027 Medicare Advantage Rate Announcement, CMS made a decision that drew less attention than the payment headline but matters more for your risk-adjustment strategy: it declined to move to a newer risk model and kept the 2024 model in place for another year.

That is a reprieve. It is not a reset — and treating it like one would be a mistake.

  1. What CMS actually decided

In April 2026, CMS finalized that for CY2027 it will continue using the 2024 CMS-HCC model — the V28 model fully phased in for 2026 — rather than the updated model (recalibrated on 2023 diagnoses and 2024 expenditures) it had floated in the January Advance Notice.

The stated rationale: give the MA market more time to adjust to the V28 phase-in it just finished absorbing. So the model you spent three years adapting to is not changing again next year.

  1. Why this is a reprieve, not relief

The model stayed the same. The scrutiny did not.

The same Rate Announcement tightened the rules on how diagnoses get counted. Starting in CY2027, CMS is excluding diagnoses from unlinked chart-review records — those not tied to a specific clinical encounter — from risk-score calculation, with a narrow exception for beneficiaries switching between MA plans. It also continues excluding audio-only-encounter diagnoses.

If your risk scores have leaned on either source, the model staying put will not save you. Those diagnosis pathways are narrowing regardless.

  1. The payment context that’s easy to misread

CMS finalized a net payment increase of about 2.48% — over $13 billion — for 2027, up sharply from the 0.09% it proposed in January.

That sounds generous, but most of the swing came from updated cost data and the decision not to adopt the full proposed risk-model change — not from any loosening of program-integrity policy. The direction of travel, toward payment accuracy and away from diagnosis-capture-as-strategy, has not changed.

  1. What to do with the extra year

Use stable model rules to fix documentation — not to coast on coding.

A year of model stability is exactly the window for the unglamorous work: tighten clinical documentation so conditions are captured at the encounter where they are actually addressed; wean any dependence on unlinked chart reviews; and make sure your highest-value HCCs are supported by encounter-based evidence that survives an audit.

The organizations that use 2027 to build durable documentation accuracy will be the ones still standing when the next model revision — and it is coming — finally lands.

  1. The audit backdrop nobody should ignore

This stability sits inside an enforcement environment that is tightening, not loosening, with CMS having signaled aggressive RADV audit activity across prior payment years.

A risk score built on thin or unlinked documentation is a liability whether or not the model changes. The reprieve buys you time to make your risk adjustment defensible — it does not make undefensible coding any safer.

Final Thoughts

It is natural to exhale when CMS leaves a model alone. But the smart read of the 2027 announcement is that CMS bought the market a year of model stability while continuing to close the loopholes that made risk scores easy to inflate.

As someone trained as a physician who later spent years in coding and audit, I would treat this year as preparation, not pause. The model is not moving in 2027. Everything around it is.

If your 2027 plan is to coast because the risk model didn’t change, you’re reading the announcement backwards. At HealtheNomics I help MA plans and risk-bearing groups use this year of stability to build documentation accuracy that survives both the next model and the next audit.

Explore the services:  https://healthenomics.com/services-2/

Request a strategy conversation:  https://healthenomics.com/contact-us/

Connect on LinkedIn:  https://www.linkedin.com/in/muhammad-ayoub-ashraf/

Website:  https://www.drayoubashraf.com

Watch on YouTube:  https://www.youtube.com/@HealtheNomics

Scroll to Top