Skip to content
Back to Blog
March 18, 2026·8 min read

RAF Score Calculator: How to Calculate Risk Adjustment Factor for Medicare Advantage

Step-by-step guide to using a RAF score calculator — inputs required, a worked example, gap analysis workflow, and how HCC Buddy's built-in risk adjustment factor calculator works.

RAF Score CalculatorRisk Adjustment FactorMedicare AdvantageHCC Coding

By Daniel Plasencia — Certified Risk Coder (CRC), Certified Professional Coder (CPC)
Reviewed: March 18, 2026

RAF Score Calculator: How to Calculate Risk Adjustment Factor for Medicare Advantage

If you are verifying a submitted RAF score, estimating the impact of a documentation gap, or validating coding accuracy before a submission deadline, a RAF score calculator is the fastest way to get an answer. The coefficients every calculator relies on come from the CMS 2026 risk-adjustment model software and ICD-10 mappings release. This post covers exactly how to use a risk adjustment factor calculator: what inputs it needs, how the math works step by step, and how HCC Buddy's built-in RAF score calculator fits into a coder's daily workflow. For a deeper look at what RAF scores are and why they exist, see how RAF scores work in detail.

What Inputs Does a RAF Score Calculator Need?

A RAF score calculator is only as accurate as the data fed into it. Before running any calculation, you need four categories of inputs:

Demographic inputs determine the patient's baseline coefficient. These include:

  • Age band (CMS uses 5-year age bands, e.g., 70-74)
  • Sex (male or female)
  • Dual-eligible status (Medicaid-Medicare dual, or Medicare-only)
  • Originally disabled status (did the patient become eligible for Medicare due to disability before age 65?)
  • Institutional status (long-term care facility resident vs. community-dwelling)
  • Diagnosis inputs are the list of Hierarchical Condition Categories that remain after the hierarchy is applied — not raw ICD-10-CM codes. The calculator works at the HCC level, not the code level. If two codes map to HCCs in the same disease hierarchy, only the highest-severity HCC is counted. Entering codes without first applying the hierarchy is the most common source of inflated RAF estimates.

    Model selection determines which coefficient table is used. In 2026, the CMS-HCC model is blended: 33% V24 coefficients plus 67% V28 coefficients. A well-designed RAF score calculator displays both V24 and V28 results individually and computes the blended 2026 score automatically.

    Interaction terms are additional coefficients that CMS adds when specific pairs of conditions appear together. Examples include heart failure combined with Chronic Obstructive Pulmonary Disease, and diabetes combined with chronic kidney disease. Missing interaction terms understates the RAF score; double-counting them overstates it.

    Step-by-Step RAF Score Calculation

    Here is a worked example using a community-dwelling Medicare Advantage member to show exactly how the numbers come together.

    Patient profile: Female, age 72, community-dwelling, non-dual, not originally disabled.

    Step 1 — Look up the demographic baseline coefficient. In the V28 model, the community non-dual female age 70-74 row carries a coefficient of approximately 0.381. This is the starting point before any diagnosis is added.

    Step 2 — Identify the coded HCCs after hierarchy application. The patient has three documented conditions:

  • Type 2 Diabetes Mellitus with Chronic Kidney Disease — ICD-10-CM code E11.22 maps to HCC 37 (Diabetes with Chronic Complications, V28). Coefficient: approximately 0.302.
  • Chronic Kidney Disease Stage 3a — ICD-10-CM code N18.31 maps to HCC 329 (Chronic Kidney Disease, Moderate Stage, V28). Coefficient: approximately 0.154.
  • Atrial Fibrillation — ICD-10-CM code I48.0 maps to HCC 238 (Cardiac Arrhythmias, V28). Coefficient: approximately 0.302.
  • For how ICD-10-CM codes map to HCC categories, see ICD-10 to HCC mapping explained.

    Step 3 — Confirm no hierarchy removes any HCC. Diabetes (HCC 37) and Chronic Kidney Disease (HCC 329) belong to separate disease hierarchies in V28 and are both counted. Atrial Fibrillation (HCC 238) is also a separate disease group. All three HCCs are retained.

    Step 4 — Check for interaction terms. HCC 37 (Diabetes with Chronic Complications) combined with HCC 329 (Chronic Kidney Disease, Moderate Stage) triggers a diabetes plus chronic kidney disease interaction term in V28. Coefficient for this interaction: approximately 0.152. This term is added on top of the individual HCC coefficients.

    Step 5 — Sum all components.

    Demographic baseline: 0.381

    HCC 37 (Diabetes with Chronic Complications): 0.302

    HCC 329 (Chronic Kidney Disease, Moderate Stage): 0.154

    HCC 238 (Atrial Fibrillation): 0.302

    Diabetes + Chronic Kidney Disease interaction: 0.152

    Approximate V28 RAF score: 1.291

    The actual CMS calculation applies the published coefficient tables and the exact blend — this example is illustrative. In a blended 2026 model, the final score combines the V24 result at 33% weight and the V28 result at 67% weight.

    Using a RAF Score Calculator for Gap Analysis

    A RAF score calculator is not just for end-of-year reconciliation. Coders and coding managers use it throughout the year for both prospective and retrospective purposes.

    Prospective gap analysis — Before a chart review, run a calculator with only the conditions already documented for a member. Then add a suspected condition that has supporting clinical language but has not yet been coded. The difference in RAF score tells you the potential documentation opportunity if the provider confirms and documents that condition. This is especially useful when preparing for Annual Wellness Visit encounters or Hierarchical Condition Category reviews.

    Retrospective verification — After coding a batch of charts, run each member's coded HCCs through the calculator and compare the output to the submitted RAF. A significant discrepancy signals either a mapping error, a missing interaction term, or a hierarchy application mistake.

    Population-level gap analysis — Coding managers can compare the average expected RAF for a panel (based on clinical complexity visible in the chart) to the average submitted RAF. A large gap across multiple members suggests a systemic coding pattern that may need training or audit attention.

    A calculator is only as good as the HCC list going in. If the wrong ICD-10-CM code is assigned, or if the code does not map to an HCC at all, the calculator will undercount the RAF. Accurate ICD-10 to HCC mapping upstream is what makes the downstream RAF calculation defensible.

    RAF Score Calculator in HCC Buddy

    HCC Buddy's built-in risk adjustment factor calculator is designed to fit into the coding workflow rather than sit alongside it.

    When you look up a code in the HCC Buddy encoder, the tool immediately shows you the V24 HCC assignment, the V28 HCC assignment, and the RAF coefficient for each model year. As you add codes to a patient session, the calculator accumulates them into a running RAF estimate — applying hierarchies and flagging interaction terms automatically.

    The demographic inputs are entered once per patient session. The calculator then displays the demographic baseline separately from the diagnosis contribution, so you can see at a glance how much of the RAF score is driven by chronic conditions versus age and enrollment factors.

    Because HCC Buddy runs as a Chrome extension, the calculator is available in the same browser window as the chart review. There is no tab switching, no separate spreadsheet, and no need to re-enter codes that are already being looked up in the encoder.

    Common Mistakes When Calculating RAF Scores

    Even experienced coders make these errors when working without a structured tool:

  • Failing to apply hierarchies before summing. Counting HCC 37 (Diabetes with Chronic Complications) and HCC 38 (Diabetes without Complication) for the same patient inflates the RAF. Only the highest-severity HCC within a hierarchy group is counted.
  • Using V24 coefficients only in 2026. The 2026 payment year uses a 33/67 blend. A calculation based entirely on V24 or entirely on V28 will not match the CMS-calculated RAF.
  • Including demographic factors that do not apply. Coding institutional status for a community-dwelling patient, or applying the originally disabled flag when the patient first enrolled at age 65 through age, adds unearned demographic weight.
  • Counting interaction terms twice. If a calculator automatically adds interaction terms, manually adding them again in a separate step doubles the contribution of those pairs.
  • Missing interaction terms entirely. Running only individual HCC coefficients without checking for valid interaction pairs understates the RAF, sometimes by 0.10 to 0.20 points.
  • HCC Buddy includes a built-in risk adjustment factor calculator that runs alongside the encoder — so every code you look up can feed directly into your RAF estimate. Try it free or see a full feature comparison at hccbuddy.com/crc.

    Daniel Plasencia

    Daniel Plasencia

    Founder & Developer

    Daniel Plasencia — Risk adjustment coding professional and software engineer who built the tool he wished existed, at a price coders can actually afford.

    Get HCC Coding Tips in Your Inbox

    Join our newsletter for coding tips, guideline updates, and tool announcements.

    Related Articles