Introduced in 2004, Hierarchical Condition Categories (HCCs) lie at the heart of CMS’s methodology for determining capitated payments for Medicare programs. The HCCs allow population expenditures to be risk-adjusted based on patient complexity. The methodology uses a 12-month diagnostic coding history to predict future financial utilization and risk. Each relevant code relates to an ICD diagnosis and has a weighted measure of the resources required to treat it. Therefore, the more complex and resource-intensive the group of beneficiaries are, the higher the predicted expenditures are for Medicare plans and Medicare Shared Savings Programs. In other words, the patient’s diagnoses codes create a risk adjustment factor (RAF) “score” for a patient that reflects his or her complexity.

It is a growing imperative for health systems to focus on accurate HCC coding. Accurate coding information helps create a more complete picture of the complexity of a patient population, improves the value of the problem list, and enables better management of a patient’s chronic diseases. Additionally, documentation that captures the full complexity of the patient often results in appropriately predicted expenditures (benchmark) for Medicare Shared Savings Programs.

When setting cost targets and savings rates, CMS factors in two things: historical payments it has made to a provider and the RAF score for the relevant Medicare population in the three years prior to the first performance year.

The HCC system may seem complex, but it really boils down to providers doing one thing: keeping problem lists up-to-date and comprehensive for every Medicare patient. For example, if a patient had a heart attack five years ago, it should be on the problem list of that patient’s medical chart. Because patients with undocumented HCCs must be re-evaluated and re-documented each year, the organization’s population risk score starts at zero in January.

On average, an organization leaves 30% of the reimbursement value of at-risk populations on the table due to inaccurate HCC capture and documentation.

Characteristics of CMS-HCC Model

  • ICD-10 diagnosis codes are sorted into HCCs to represent the most severe manifestation of the disease burden on the patient.
  • The HCC model contains 79 diagnostic categories, reflective largely of chronic conditions experienced in a Medicare population.
    • An HCC is made up of many ICD codes with each ICD code mapping directly to only one HCC. There are ICD codes that do not map to any existing HCC.
  • Risk scores for each diagnosed HCC are added together for each patient, as are additional factors for comorbidities (ex. CHF and COPD) to calculate a RAF score.
    • The model is also calibrated to account for disease severity, requiring coding specificity (ex. diabetes unspecified vs. diabetes with renal complications).
  • Financial benchmarks are set based on HCC and RAF logic.
  • Conditions must be documented and billed at least once per calendar year for inclusion in the risk model.