Hierarchical Condition Categories (HCCs) which is CMS’ risk adjustment reimbursement methodology for Medicare Advantage (MA) (Medicare Part C) has undergone a significant overhaul for CY 2024 in the form of the release of a new model version.
CMS has explained the rationale for a new model as follows:
“In the MA payment system, all diagnosis codes are mapped to categories of diagnoses (called HCCs, Hierarchical Condition Categories) that are clinically related and have similar ability to predict Medicare costs. The existing HCCs (used in the current 2020 model) are built using ICD-9 codes. However, ever since the health system transitioned to ICD-10 in 2015, the codes coming in from plans to CMS for MA payment have been ICD-10 codes. Thus, what CMS had to do is map ICD-10 codes to ICD-9 condition categories (HCCs) while it waited for ICD-10 coding practices to stabilize.
For 2024, CMS undertook a CMS-HCC reclassification that involved newly building condition categories from the ground up, reviewing each diagnosis and determining the best grouping of diagnoses to be clinically sound and their ability to predict Medicare costs, with iterative input from our clinical expert panel over multiple years. The goal of updating the underlying data and the clinical reclassification was to improve predictive ability by better reflecting current disease patterns, treatment methods and costs, and diagnosis and coding practices.” https://www.cms.gov/files/document/2024-announcement-pdf.pdf
A new version of the CMS-HCC model, V28, was finalized in 2023 for implementation in CY 2024.
The V28 model includes these significant changes:
V24 | V28 |
HCC 19 Diabetes without Complication | HCC 38 Diabetes with Glycemic, Unspecified, or No Complications |
HCC 34 Chronic Pancreatitis | HCC 79 Chronic Pancreatitis |
No HCC | HCC 279 Severe Persistent Asthma |
Shown here is a summary of the quantity of HCCs, and the number of each HCC per disease group, for V28 vs. V24.
Disease Group | Quantity of HCCs | HCC Numbers | ||
V24 | V28 | V24 | V28 | |
Infections | 3 | 3 | 1-2, 6 | 1-2, 6 |
Neoplasm | 5 | 7 | 8-12 | 17-23 |
Diabetes | 3 | 4 | 17-19 | 37-38 |
Metabolic | 3 | 4 | 21-23 | 48-51 |
Liver | 3 | 5 | 27-29 | 62-65, 68 |
GI | 3 | 5 | 33-35 | 77-81 |
Musculoskeletal | 2 | 3 | 39-40 | 92-94 |
Blood | 3 | 7 | 46-48 | 107-115 |
Cognitive | 2 | 3 | 51-52 | 125-127 |
Substance Use Disorder | 3 | 5 | 54-56 | 135-139 |
Psychiatric | 4 | 5 | 57-60 | 151-155 |
Spinal | 3 | 3 | 70-71 | 180-182 |
Neurological | 8 | 12 | 72-80 | 191-193, 195-202 |
Arrest | 3 | 3 | 82-84 | 211-213 |
Heart | 3 | 10 | 86-88 | 221-229, 238 |
Cerebrovascular | 4 | 4 | 99-100, 103-104 | 248-249, 253-254 |
Vascular | 3 | 3 | 106-108 | 263-264, 267 |
Lung | 5 | 7 | 110-112, 114-115 | 276-280, 282-283 |
Eye | 2 | 2 | 122, 124 | 298, 300 |
Kidney | 5 | 4 | 135-138 | 326-329 |
Skin | 5 | 7 | 157-159, 161-162 | 379-383, 385, 387 |
Injury | 5 | 6 | 166-167, 169-170, 173 | 397-399, 401-402, 405 |
Complications | 1 | 0 | 176 | N/A |
Transplant | 1 | 1 | 186 | 454 |
Openings | 1 | 1 | 188 | 463 |
Amputation | 1 | 1 | 189 | 409 |
V28 is being phased in over a three-year period. During the transition period risk adjustment factors will be calculated based on a model that blends the value of the score under the V24 version and the value of the score under the V28 version as shown in the table below.
Calendar Year | Version 24 | Version 28 |
2024 | 67% | 33% |
2025 | 33% | 67% |
There are two major ways that the transition to V28 will impact the calculation of RAF scores; removal of codes that no longer map to an HCC and constraining of HCC coefficients.
With the removal of 2,294 codes that map to an HCC, some diagnoses that were formerly considered in calculation of a patient’s RAF score will no longer be counted.
A considerable change comes in the form of constraining the coefficient values for HCCs. Under constraining, related HCCs are given the same coefficients. For example, in V28 the coefficient is the same (0.166) for both uncomplicated diabetes and diabetes with an acute complication whose coefficient under V24 would have been significantly higher at 0.312.
These changes mean that a patient’s RAF score, and consequently the payment received for treating the patient, may decline even if the patient has not had a change in health status.
Although the coefficient for some HCCs will increase, which may increase the RAF score for some patients, depending on their clinical picture, the overall loss of value from constraining will not overcome those gains as Medicare predicts an overall decrease in payment of 3.12% with a projected net savings of 11 billion dollars.
These CMS links will take you to a complete listing of ICD-10-CM codes that map to an HCC 2024 Model Software/ICD-10 Mappings | CMS and their corresponding coefficients for 2024 (see Table VIII-1, page 183). Announcement of Calendar Year (CY) 2024 Medicare Advantage (MA) Capitation Rates and Part C and Part D Payment Policies (cms.gov)
The expansion of the number of HCCs is reflective of the greatest specificity available in ICD-10 vs. ICD-9. Coders will need to be vigilant in assigning codes to the highest degree of specificity supported by documentation in the health record. In turn, documentation from providers should be regularly monitored or queries posed for the specificity that allows the assignment of codes that most accurately and completely communicate the patient’s clinical picture. On the corporate level, organizations will need to determine what the predicted impact to RAF scores means for them based on their particular case mix.
References
Since 1992, HIA has been the leading provider of compliance audits, coding support services and clinical documentation audit services for hospitals, ambulatory surgery centers, physician groups and other healthcare entities. HIA offers PRN support as well as total outsource support.