What Are Hierarchical Condition Categories
The hierarchical condition category system is the engine that drives Medicare Advantage risk adjustment payments. Hierarchical Condition Categories are a clinical classification system developed by CMS to group related medical conditions into categories that predict healthcare expenditure. Each HCC represents a clinically meaningful set of diagnoses that share similar cost patterns and treatment complexity. The HCC system is the foundation of the CMS-HCC risk adjustment model used to calculate RAF scores for Medicare Advantage payment.
The system works by mapping thousands of ICD-10-CM diagnosis codes to a smaller number of condition categories, then organizing those categories into hierarchies based on clinical severity. This mapping and hierarchy process converts the complexity of individual patient diagnoses into a structured risk profile that CMS uses to adjust payments to Medicare Advantage plans.
V28 Structure
CMS-HCC V28 contains 115 HCCs organized into 26 disease families, mapped from approximately 7,770 ICD-10-CM codes. Each HCC carries a CMS-published risk coefficient that reflects its predicted annual cost impact on healthcare utilization.
Financial Impact
HCCs are the link between clinical documentation and plan revenue. Each captured HCC adds its coefficient to the member's RAF score, which directly determines CMS capitation payments. A single missed HCC can represent $1,000-$4,000+ in annual revenue per member.
How the Hierarchy Works
The "hierarchical" in HCC refers to the severity ranking within each disease family. When a patient has multiple related conditions in the same hierarchy, the model applies only the most severe — preventing double-counting while ensuring the highest-impact condition drives the risk score.
- Severity Ranking: Within each disease family, HCCs are arranged from most severe (highest cost prediction) to least severe. A patient with both moderate and severe heart failure will have only the severe heart failure HCC counted in their RAF score — the moderate HCC is "trumped" by the higher-ranking condition in the hierarchy.
- Cross-Family Independence: Hierarchies only apply within disease families, not across them. A patient with severe diabetes AND severe heart failure has both HCCs counted because they belong to different disease families. The hierarchy prevents double-counting within diabetes, and separately within heart failure, but does not limit the accumulation of conditions across different systems.
- Practical Example: Consider a patient with Type 2 diabetes with diabetic nephropathy (HCC 37), Type 2 diabetes with peripheral circulatory complications (HCC 38), and congestive heart failure (HCC 85). Within the diabetes hierarchy, only the highest-ranking diabetes HCC counts. The heart failure HCC counts independently because it belongs to a separate disease family.
- V28 Constraining Modification: Under V28, constraining assigns identical coefficients to related HCCs within certain disease families regardless of severity. All diabetes HCCs (except pancreas transplant) carry the same weight of approximately 0.166. The hierarchy still determines which HCC is counted, but the financial impact is the same regardless of which level within the constrained family is documented.
- Documentation Implications: Even with constraining, documenting the most severe condition within a hierarchy remains important for clinical accuracy, audit defensibility, and future model changes. CMS may revise constraining in subsequent model updates, and organizations that maintained accurate severity documentation will be positioned to capture differential coefficients if constraining is modified.
The 26 Disease Families in V28
CMS-HCC V28 organizes its 115 HCCs into 26 disease families, each representing a clinically related group of conditions. Understanding these families is essential for RAF score calculation and risk adjustment strategy.
- High-Prevalence Families: Diabetes (HCCs 35-38), Cardiovascular (including heart failure HCCs 85-88, coronary artery disease, and vascular disease), and Chronic Kidney Disease (HCCs 326-329) represent the disease families encountered most frequently in Medicare Advantage populations. These families account for the majority of RAF value in a typical plan.
- High-Coefficient Families: Some disease families carry higher individual HCC coefficients despite lower prevalence — including transplant status, severe hematological conditions, and certain neoplasms. While affecting fewer members, each captured HCC in these families generates substantial per-member RAF value.
- Mental Health and Substance Use: V28 expanded the representation of behavioral health conditions compared to V24, with dedicated disease families for substance use disorders and psychiatric conditions. These HCCs are frequently undercaptured because behavioral health documentation often occurs in specialty settings with incomplete claims capture.
- Musculoskeletal and Pain: Conditions like rheumatoid arthritis, inflammatory connective tissue disease, and severe musculoskeletal conditions have dedicated HCC families in V28. These conditions often affect coding complexity because of overlap between pain management documentation and the underlying condition documentation.
- New V28 Families: V28 introduced several disease families that did not exist in V24, including expanded infectious disease categories and reorganized autoimmune condition classifications. Organizations transitioning from V24 assumptions must map their populations against the new family structure to identify capture opportunities that did not exist under the prior model.
ICD-10 to HCC Mapping Process
The journey from clinical documentation to RAF score begins with ICD-10-to-HCC mapping — the process that converts individual diagnosis codes into the condition categories that drive risk scoring.
- Code-to-HCC Crosswalk: CMS publishes annual crosswalk files that define which ICD-10-CM codes map to which HCCs. Under V28, approximately 7,770 of the 73,000+ ICD-10-CM codes map to HCCs. The remaining codes — while clinically valid — do not affect risk adjustment. This means fewer than 11% of all available diagnosis codes are HCC-relevant.
- Many-to-One Mapping: Multiple ICD-10 codes can map to the same HCC. For example, dozens of specific diabetes complication codes may all map to the same diabetes HCC. This many-to-one relationship means that coding specificity determines which HCC is triggered, but multiple codes can reach the same destination.
- Codes That Lost Mapping: V28 removed approximately 2,294 ICD-10 codes from HCC mapping that were previously valid under V24. Organizations still coding based on V24 mapping assumptions are submitting codes that generate no HCC and no RAF value — creating invisible revenue leakage. Annual ICD-10 updates further modify the mapping landscape each October.
- Specificity Requirements: V28's reduced code set means that less specific codes are more likely to fall outside HCC mapping. Coding at the highest supported specificity level is no longer a best practice — it is a revenue requirement. Unspecified codes that mapped to HCCs under V24 may not map under V28.
- Mapping Validation: Every submitted diagnosis code should be validated against current-year mapping tables before CMS submission. Automated mapping validation catches codes that do not trigger HCCs, codes that map to unexpected HCCs, and codes that have been removed from the current-year crosswalk.
How HCCs Drive RAF Scores
HCCs are the clinical input that, combined with demographic factors, produce the RAF score used for CMS payment calculation. Understanding the connection between HCC capture and revenue is essential for every risk adjustment stakeholder.
- Coefficient Assignment: Each HCC carries a CMS-published risk coefficient reflecting its predicted cost impact. Coefficients range from approximately 0.06 for lower-impact conditions to over 1.0 for the most severe conditions. These coefficients are derived from historical claims data analysis and represent the incremental cost above the demographic baseline.
- Additive Model: RAF Score = Demographic Baseline + Sum of HCC Coefficients + Disease Interaction Factors. HCCs add incrementally to the score — a member with three captured HCCs has a higher score than one with two (assuming they are from different disease families). This additive nature means every additional HCC captured directly increases the RAF score and corresponding revenue.
- Disease Interactions: Certain HCC combinations trigger interaction factors that add additional value beyond the individual HCC coefficients. Diabetes combined with heart failure, for example, generates an interaction bonus because the combination predicts higher costs than either condition alone. Identifying and documenting interaction-eligible conditions represents high-value opportunity.
- Normalization: CMS applies a normalization factor to maintain budget neutrality. The raw RAF score is divided by this factor to produce the final payment score. Normalization means that as average RAF scores increase across the industry, the per-point value adjusts — though the relative importance of accurate capture remains constant.
- Revenue Translation: The final RAF score multiplied by the county base rate (approximately $10,400 in 2026) determines the annual CMS payment for that member. Each 0.1 increase in RAF score translates to roughly $1,040 in annual revenue per member. For population-level impact, multiply by total membership.
Annual Recapture Requirements
The CMS-HCC model resets every payment year, meaning no HCC carries forward automatically regardless of the clinical permanence of the underlying condition. This annual reset creates the recapture requirement — the operational necessity to re-document every chronic condition every year.
- Full Annual Reset: On January 1 of each payment year, every member's HCC profile resets to zero. Chronic conditions documented in the prior year must be re-documented through a qualified encounter in the current year to count toward the current RAF score. There are no exceptions for permanent conditions.
- Qualified Encounter Requirements: HCC-supporting documentation must come from a face-to-face encounter with a qualified provider. Telehealth visits, when conducted with appropriate audio and video components, qualify. Phone-only encounters, lab orders without a clinical visit, and prescription refills without an encounter do not satisfy the documentation requirement.
- MEAT Documentation Standards: Each recaptured condition must demonstrate that it was Monitored, Evaluated, Assessed, and Treated (MEAT) during the encounter. Simply listing a condition on the problem list or assessment without clinical context fails the MEAT standard and will not survive audit validation.
- Recapture Timing: While HCCs can be recaptured at any point during the payment year, earlier recapture is operationally preferable. Conditions recaptured in Q1-Q2 provide more time for supplemental data submission and mid-year payment adjustments. Q4 recapture efforts face tighter deadlines and reduced flexibility.
- Recapture Rate Benchmarks: High-performing organizations achieve 90%+ chronic HCC recapture rates through systematic risk stratification and targeted outreach. Plans without dedicated recapture programs typically achieve 70-75%, losing 25-30% of prior-year chronic HCC value annually.
Common HCC Misunderstandings
Several persistent misconceptions about the HCC system lead to operational errors, compliance risk, and missed revenue. Correcting these misunderstandings is a prerequisite for effective risk adjustment.
- "All diagnosis codes map to HCCs." Under V28, fewer than 11% of ICD-10-CM codes map to HCCs. The vast majority of codes, while clinically valid, have no risk adjustment impact. Organizations that do not distinguish between HCC-mapped and non-mapped codes waste resources documenting and coding conditions with no RAF value.
- "Higher severity always means higher payment." Under V28 constraining, related HCCs within constrained disease families carry identical coefficients. Coding a more severe condition within a constrained family does not generate more revenue than coding a less severe condition. This eliminates the financial incentive for severity upcoding while maintaining clinical accuracy requirements.
- "HCCs and ICD-10 codes are the same thing." ICD-10 codes are diagnosis classifications; HCCs are risk categories. Multiple ICD-10 codes map to each HCC, and the mapping changes annually. Confusing the two leads to coding strategies that optimize for ICD-10 volume rather than HCC accuracy.
- "Once captured, an HCC stays on the score." The annual reset means every HCC must be re-documented every year. Organizations that assume prior-year conditions carry forward experience predictable RAF score declines that are entirely preventable through systematic recapture programs.
- "The HCC system has not changed significantly." V28 is a fundamental redesign — new disease families, constrained coefficients, reduced mapping paths, and a rebuilt ICD-10-native architecture. Organizations applying V24 assumptions to V28 will systematically misestimate revenue and misallocate risk adjustment resources.