On January 26, 2026, the Centers for Medicare & Medicaid Services (CMS) released the Calendar Year (CY) 2027 Advance Notice of Methodological Changes for Medicare Advantage (MA) Capitation Rates and Part C and Part D Payment Policies, which makes proposals to update program policies for MA and Part D beginning in 2027. A press release and a fact sheet were also released.
CMS proposes the following:
- Updates to the CMS-Hierarchical Conditions Categories (HCC) risk adjustment model, including exclusions for diagnoses from audio-only encounters and unlinked Chart Review Records (CRRs),
- Continued transition to the CMS-HCC risk adjustment model for Program of All-Inclusive Care for the Elderly (PACE) organizations,
- Continued use of the End-Stage Renal Disease (ESRD) risk adjustment model,
- Updates to the frailty adjustment for Fully Integrated Dually Eligible (FIDE) Special Needs Plans (SNPs),
- Updates to Star Ratings, including requests for feedback on new quality concepts, and
- Updates to the Part D risk adjustment (RxHCC) model reflecting Inflation Reduction Act (IRA)-related changes, separate MA-PD and Prescription Drug Plan (PDP) segments, and exclusions of audio-only diagnoses and unlinked CRRs.
Overall, if these proposals are finalized, CMS anticipates a 0.09 percent, or over $700 million, increase in MA plan payments from 2026 to 2027.
Public comments on the Advance Notice must be submitted by 11:59 PM Eastern Time on Wednesday, February 25, 2026. The CY 2027 Rate Announcement will be published no later than April 6, 2026.
Effective Growth Rate and MA Plan Payments For 2027
See CMS’s Fact Sheet
The Effective Growth Rate reflects the current estimates of the growth rates of benchmarks utilized in determining payments for MA plans. Growth in Medicare FFS per capita costs, as assessed by the CMS Office of the Actuary, have been the main driver for these growth rates.
For 2027, CMS proposes an Effective Growth Rate of 4.97 percent, with a net impact of $22.86 billion. Overall, if these proposals are finalized, CMS anticipates a 0.09 percent, or over $700 million, increase in MA plan payments from 2026 to 2027.
The proposed 0.09 percent increase represents a very small change compared with last year’s 5.06 percent increase, meaning overall plan payments are essentially flat year over year. This minimal growth may have limited impact on premiums or benefits, though effects could vary by plan depending on their beneficiary mix and cost structure.
Updated CMS-HCC Model
Pages 50-53 and 63-66
Though largely similar to the 2024 CMS-HCC risk adjustment model used for payment for CY 2024 through 2026, CMS proposes to implement an updated CMS-HCC risk adjustment model for CY 2027, which reflects more recent data and additional exclusions. Several notable proposals are as follows:
- Exclusion of Diagnoses from Audio-Only Services: CMS proposes excluding diagnoses from audio-only encounters. While telehealth diagnoses are eligible for risk adjustment if provided face-to-face via interactive audio-video systems, services for audio-only telehealth identified with the “93” or “FQ” modifiers would be excluded unless another claim line meets risk adjustment criteria.
- Exclusion of Unlinked CRRs: Given the maturity of encounter data and program integrity concerns raised by stakeholders, including the Office of Inspector General (OIG)[1] and the Medicare Payment Advisory Commission (MedPAC),[2] CMS proposes excluding diagnosis information from unlinked CRRs from risk score calculations. CMS believes that requiring diagnoses to be linked to actual service records will support more accurate payments, and that excluding unlinked CRRs may help reduce payment differences resulting from differential coding across MA organizations. This proposal would not prohibit MA organizations from submitting unlinked CRRs to the encounter data system (EDS).
In addition, CMS proposes recalibrating the model using 2023 diagnoses and 2024 fee-for-service (FFS) expenditures. The denominator year would be updated to 2024 to reflect more current costs and to calculate relative factors for demographics and HCCs. CMS also proposes removing the specified constraint on certain HCCs for chronic kidney disease, allowing each HCC to reflect its true relative cost based on updated data and sufficient sample sizes.
The proposed 2027 CMS-HCC updates are designed to improve payment accuracy. The exclusion of unlinked CRRs is particularly significant, as it ensures that only diagnoses linked to actual service records are used in risk score calculations, reducing the potential for differential coding and inconsistent payments across MA organizations. Combined with other updates, these changes aim to better align payments with true beneficiary risk.
CMS-HCC Risk Adjustment Models for PACE Organizations
Pages 53-54
Per prior Advance Notices and Rate Announcements, CMS has signaled its intention to transition PACE organizations to encounter data submissions and alignment with the CMS-HCC model used for other MA organizations. For CY 2027, CMS proposes continuing this transition by calculating PACE risk scores as a 50/50 blend of 50 percent from the 2017 CMS-HCC model and 50 percent from the proposed 2027 CMS-HCC model. CMS views the updated CMS-HCC model as an improvement in payment accuracy because it reflects more recent cost and diagnosis data and uses ICD-10 coding. The Advance Notice also provides additional details on sources of diagnoses for PACE risk score calculation in Attachment II, Section L1.[3]
For CY 2027, the proposed blended risk score approach for PACE organizations continues the transition to using encounter data and the CMS-HCC model applied to other MA organizations, incorporating both recent and historical data. This change may affect risk scores and payment amounts by reflecting updated diagnoses and coding patterns.
ESRD Risk Adjustment Models
Pages 54-55
CMS proposes to maintain its current approach to ESRD risk adjustment for MA organizations, while continuing a transitional blended policy for PACE organizations. For MA plans other than PACE, CMS would continue using the 2023 ESRD CMS-HCC models[4] to calculate risk scores for beneficiaries in dialysis, transplant, and post-graft status, with no substantive methodological changes and continued reliance on diagnoses from encounter data and Medicare FFS claims.
For PACE organizations, CMS proposes to continue a 50/50 blended ESRD risk adjustment for CY 2027, combining risk scores calculated using the 2019 ESRD CMS-HCC models (with diagnoses from RAPS, encounter data, and fee-for-service claims) and the 2023 ESRD CMS-HCC models (with diagnoses from encounter data and FFS claims only). CMS characterizes this approach as transitional and reiterates its intent to fully align PACE ESRD risk adjustment with the MA framework in future years.
Together, these proposals signal near-term stability for MA ESRD payments and a deliberate, phased transition for PACE organizations, with a clear policy direction toward standardized models and, over time, exclusive reliance on encounter and FFS data.
Frailty Adjustment Methodology for FIDE SNPs
Pages 55-57
For CY 2027, CMS proposes updating the frailty adjustment for FIDE SNPs to align with the proposed 2027 CMS-HCC risk adjustment model. The frailty adjustment improves payment accuracy for community-dwelling beneficiaries with functional impairments not fully captured by diagnosis-based risk adjustment and is applied by adding a frailty score to a beneficiary’s risk score when a FIDE SNP’s average frailty is comparable to that of the PACE population.[5] Because frailty factors are calibrated by regressing unexplained CMS-HCC model costs on beneficiaries’ limitations in activities of daily living (ADLs), and these residual costs vary by model version, the factors must be updated when the CMS-HCC model changes.
For 2027, CMS proposes recalibrating frailty factors using the 2027 CMS-HCC model, based on 2023 diagnoses and 2024 Medicare FFS expenditures, and ADL data from the 2023 Medicare FFS Consumer Assessment of Health Providers & Systems (CAHPS) survey, replacing the prior 2018 CAHPS-based calibration. The proposed factors address negative or nonmonotonic results by setting Partial Medicaid frailty factors equal to those of the Non-Medicaid group and constraining certain ADL group factors for the Full-Medicaid group.
CMS reiterates that, effective January 1, 2025, FIDE SNPs must limit enrollment to full-benefit dually eligible individuals and expects all beneficiaries used in the 2026 survey data for CY 2027 frailty scores to meet this requirement.
CMS plans to estimate the PACE minimum frailty score used as the threshold to establish whether a FIDE SNP qualifies to receive a frailty adjustment in CY 2027 in the same way it proposed to calculate FIDE SNP frailty scores.
These proposals highlight CMS’s focus on technical alignment, updated data, and payment stability rather than expanding access to frailty adjustments. The recalibrated frailty factors and applied constraints for Partial and Full Medicaid groups aim to ensure that payments more accurately reflect beneficiary functional status and promote consistent, equitable payments for eligible FIDE SNPs
Adjustment to FFS Per Capita Costs in Puerto Rico
Pages 36-37
According to CMS, a significantly higher proportion of Medicare Part A and B FFS beneficiaries in Puerto Rico had zero claims in 2019 through 2023 (13.8 percent) than nationwide (6 percent). To maintain stability for the FFS program in Puerto Rico, CMS has historically increased per-capita FFS costs, including by 4.1 percent in CY 2026. For CY 2027, CMS will conduct a similar analysis using updated data, and requests feedback on implementing such an adjustment. CMS also reaffirmed their methodology for setting MA benchmark rates for the territory remains unchanged.
Star Ratings
Star Ratings are a quality rating system for MA and PDPs. These ratings are released annually and consist of a one-to-five-point scale (with five indicating excellent performance). Measures used to calculate 2027 Star Ratings are included in Table IV-1 of the Advance Notice.[6]
Proposed Changes to Existing Star Ratings Measures
Pages 109-114
Key proposed updates include changes to measure specifications for five measures and six display measures, and the retirement of two display measures (Disenrollment Reasons Survey measures (Part C and D) and Antipsychotic Use in Persons with Dementia for Long-Term Nursing Home Residents (APD-LTNH) (Part D)).
Feedback Requested to Simplify and Refocus the Measure Set
Page 111
As the Part C and Part D Star Rating programs continue to evolve, CMS requests feedback on new measures or measure concepts to incentivize plans to reduce unnecessary, inappropriate, or low-value care. The agency is also particularly interested in measures that address medical errors or misdiagnoses.
Future Potential Methodological Changes
Page 114
CMS states that, as part of ongoing efforts to simplify the Star Ratings program, it is considering methodological changes to simplify calculations for the agency and consumer. The agency suggests this could be implemented through simplifying the methodology for determining measure thresholds, including using percentile distribution cut-offs to determine stars rather than the current methodology.
These changes reflect the Trump administration’s focus on streamlining and reshaping quality programs in line with the MAHA agenda and broader regulatory reform initiatives.
Part D Risk Adjustment
Pages 83-98
CMS uses the RxHCC risk adjustment model to improve payment accuracy for Part D plans based on the expected drug costs of their enrolled populations.
To account for IRA-related changes to the Part D benefit, which significantly increased gross plan liability, CMS proposes updates to this model, including:
- Increasing manufacturer discounts for specified manufacturers and specified small manufacturers, as aligned with the phase-in schedule under the IRA,
- Update the model’s list of adult vaccines and covered insulin products subject to revised IRA cost-sharing, using the most current applicable National Drug Codes (NDCs), and
- Adjusting gross drug costs for selected drugs that have a maximum fair price (MFP) in effect for 2026. 2027 MFPs will be incorporated into the model for future years, as 2027 MFPs were not available in time for the calibration of the 2027 RxHCC Model.[7]
CMS also proposes continuing to use adjusted annual out-of-pocket (OOP) thresholds for data years prior to the IRA to reflect what the thresholds would have been had the IRA been in place at the time.
Other proposed adjustments to the model include:
- Calibrating the model using more recent years (diagnoses from 2023, costs from 2024),
- Updating the denominator year to from 2023 to 2024,
- Excluding diagnoses from audio-only services and from unlinked CRRs, aligned with changes proposed for the CMS-HCC model, and
- Calibrating separate model segments for MA-PD and PDP beneficiaries, to address CMS’s findings that MA-PD plan costs tend to be overpredicted, while PDP costs tend to be
Aligned with previous years, CMS also proposes to continue to calculate separate normalization factors for MA-PD and PDP beneficiaries.
These proposals are aligned with CMS’s priorities of improving payment accuracy and promoting stability for beneficiaries in prescription drug plans. As a result of calibrating separate model segments for MA-PD and PDP beneficiaries, some disease states, particularly in plans with high concentrations of certain conditions, may see notable changes in Part D reimbursement. Overall, the net impact will vary by plan.
Part D Benefit Parameters for 2027 Plan Year
Pages 75-83
CMS updates the Medicare Part D benefit parameters for the defined standard drug benefit on an annual basis. Given the changes enacted by the IRA, the defined standard benefit deductible, annual OOP threshold, and low-income subsidy (LIS) benefit parameters have been updated by the methodology provided under the Social Security Act.
| Standard Benefit | 2026 | 2027 (Draft) |
| Deductibles
Beneficiary is responsible for 100 percent of drug costs. |
$615 | $700 |
| Out-of-Pocket Threshold
Beneficiary does not have cost-sharing after out-of-pocket spending reaches $2,400. The Coverage Gap Discount Program was sunset effective January 1, 2025, and was replaced with the Manufacturer Discount Program. For applicable drugs, plans will be responsible for 60 percent of drug costs, Medicare will be responsible for 20 percent, and manufacturers will be responsible for 20 percent. For non-applicable drugs, plans will be responsible for 60 percent of drug costs and Medicare will be responsible for 40 percent. An applicable drug is a drug approved under a New Drug Application or Biologics License Application. |
$2,100Statutorily set under the IRA. | $2,400Statutorily set under the IRA. |
| Maximum Copayments for Non-Institutionalized Dual Eligibles | ||
Full Subsidy-Full Benefit Dual Eligible (FBDE) Beneficiaries Up to 100 percent of federal poverty level (FPL)
|
$1.60
$4.90 |
$1.65
$5.00 |
Full Subsidy-FBDE Beneficiaries Between 100 percent and 150 percent of FPL
|
$5.10
$12.65 |
$5.80
$14.40 |
Full Subsidy-Non-FBDE Beneficiaries Applied or eligible for QMB/SLMB/QI or SSI, income at or below 150 percent FPL for 2026 and resources ≤ $15,720 (individuals, 2026) or ≤ $31,360 (couples, 2026)
|
$5.10
$12.65 |
$5.80
$14.40 |
********
This Applied Policy® Summary was prepared by Caitlyn Bernard with support from the Applied Policy team of health policy experts. If you have any questions or need more information, please contact her at CBernard@appliedpolicy.com or at (202) 558-5272.
Download a copy of this summary here.
[1] https://oig.hhs.gov/documents/evaluation/2792/OEI-03-17-00470-Complete%20Report.pdf
[2] https://www.medpac.gov/wp-content/uploads/2025/03/Mar25_Ch11_MedPAC_Report_To_Congress_SEC.pdf
[3] See page 66-67 of the Advance Notice.
[4] See section H of the CY 2023 Advance Notice: https://www.cms.gov/files/document/2023-advance-notice.pdf
[5] Section 1853(a)(1)(B)(iv) of the Act
[6] See pages 102-106 of the Advance Notice.
[7] The IRA introduced the Medicare Drug Price Negotiation Program, under which CMS negotiates an MFP for drugs selected for negotiation.
