The CMS-HCC Model, Risk Scores, & Risk-Adjusted Payments
Every year, the federal Centers for Medicare & Medicaid Services (CMS) is required by law to update how much Medicare Advantage (MA) plans are paid to cover their enrollees. Annual payment rates are mostly driven by how CMS updates 1) benchmarks and 2) their risk adjustment model.[1] Benchmarks are covered in our recent blog “Benchmarks for Payments to Medicare Advantage Plans.” This blog will cover the CMS-HCC risk-adjustment model, risk scores, and risk-adjusted payments.
Medicare payments to MA plans are adjusted to account for differences in enrollees’ expected medical spending. Risk adjustment ensures fair payment for sicker enrollees and reduces the incentive for plans to only select the healthiest, lowest-cost retirees for coverage.[2] CMS uses a risk-adjustment model known as the CMS Hierarchical Condition Categories (CMS-HCC) model. The model is prospective: it uses a profile of major medical conditions in the base year, along with demographic information, to predict Medicare expenditures in the next year.[3] In an example, an enrollee’s medical conditions documented in 2025, along with their current demographic data, would predict Medical expenditures in 2026.
Demographic and Disease Variables in the CMS-HCC Model
The CMS-HCC model uses current demographics (age, sex, disability status, etc.) and documented diagnosis codes as variables. An “Age-in”, or a retiree who recently became Medicare-eligible and transitioned to a Medicare-based plan, will only have demographic variables to input. However, after accumulating experience data from provider visits, they will have diagnosis variables to factor in as well.[4]
Diagnosis codes are reported by providers from patient encounters.[1] In 2026, version 28 (V28) of the CMS-HCC, also referred to as the 2024 CMS-HCC model, will use the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes.[5] The HCC classification system assigns hierarchies or families of ICD-10-CM diagnosis codes to a single HCC code*.[6] Some closely related HCCs are grouped into hierarchies based on condition severity[7] so that each enrollee’s risk score, and payments to plans, are driven only by their most serious and costly conditions.[1]
Variable Relative Factors and Raw Risk Scores
The CMS-HCC model provides relative factors associated with each variable, which are numbers that represent their expected costs.[1] The relative factors, also known as coefficients[8], of an enrollee are added together to calculate their raw risk score.[9]
Relative factors for Demographics + Diagnoses = Raw risk score
Raw Risk Score Adjustments
The raw risk score is adjusted for normalization and the Medicare Advantage coding pattern difference. Normalization establishes a 1.0 risk score for an enrollee with expected spending equal to average Original Medicare spending. The Medicare Advantage coding pattern adjustment accounts for differential coding patterns between MA plans and Fee-For-Service providers. The result of these two adjustments is the payment risk score[8], or simply risk score. Healthier enrollees will have risk scores below 1.0, while sicker enrollees will have risk scores above 1.0.[1]
Applying Risk Scores to Project Risk-Adjusted Payments
Applying an enrollee’s risk score to their benchmark determines how much Medicare will pay an MA plan for that enrollee.[4] This is referred to as a risk-adjusted payment. Building on our recent blog about benchmarks, we will refer to CMS’s 2026 MA Rate Book in the Local EGWP tab and pull county-level benchmarks for retirees in a group Medicare Advantage plan with a 4.0 Star Rating.
| County | Benchmark PMPM | Risk Score | Projected Payment PMPM | |
| Retiree A | Burlington (NJ) | $1,234.08 | 1.25 | $1,542.60 |
| Retiree B | Mercer (NJ) | $1,211.09 | 0.83 | $1,005.20 |
| Retiree C | Camden (NJ) | $1,232.53 | 1.16 | $1,429.73 |
| Group Average | $1,225.90 | 1.08 | $1,323.97 |
In our example, we multiply the MA group’s average benchmark of $1,225.90 per member per month (PMPM) by the group’s average risk score of 1.08 to project an average $1,323.97 PMPM in risk-adjusted payments from Medicare to pay for the group’s covered medical expenses.
*Not all ICD-10-CM diagnoses are linked to an HCC that affects risk scores and, therefore, not all health conditions are risk adjustable: 90.5% of ICD-10-CM codes are not included in the final payment model. Further, even for risk adjustment-eligible diagnoses, not all performed services qualify for inclusion in the risk adjustment models.[6]
___
Sign up for our quarterly newsletter, the LaborFirst Review, to receive the latest industry insights and education as we gear up for Medicare changes in 2026. Scroll down to our footer and enter your email to stay in the loop.
[1] The Commonwealth Fund, How the Government Updates Payment Rates for Medicare Advantage Plans, March 4, 2024.
https://www.commonwealthfund.org/publications/explainer/2024/mar/how-government-updates-payment-rates-medicare-advantage-plans
[2] CMS, Risk Adjustment
https://www.cms.gov/priorities/innovation/key-concepts/risk-adjustment
[3] CMS, Report to Congress: Risk Adjustment in Medicare Advantage, December 2024.
[4] Humana, Group Medicare Pricing 101 presentation, November 2024.
[5] RetireeFirst and BluePeak, “The Evolution of Employer Retiree Plans: Group Medicare Advantage & Part D in 2025” July 31, 2024.
http://laborfirst.com/wp-content/uploads/RF-White-Paper-The-Evolution-of-Employer-Retiree-Plans.pdf
[6] Milliman, “Risk Adjustment: Methodologies for Identifying Uncaptured Conditions” by Meng Sun, Dillon Afenir, and Austin Barrington, Jan 10, 2025.
https://www.milliman.com/en/insight/risk-adjustment-methodologies-uncaptured-conditions
[7] MedPAC, Comments to CMS Administrator, March 1, 2024.
https://www.medpac.gov/wp-content/uploads/2024/03/03012024_MA_PartD_CY2025_AdvanceNotice_MedPAC_COMMENT_SEC.pdf
[8] CMS, Training Module 1: Risk Adjustment Introduction and Overview
[9] CMS, Training Session Two: Risk Score Calculation, Nov 28, 2018.
