Risk adjustment coding is the process of translating a patient’s documented medical conditions into specific diagnosis codes that determine how much a health plan gets paid. The core idea is straightforward: insurers covering sicker patients should receive more money than those covering healthier ones. Risk adjustment coding is the mechanism that makes this happen, linking clinical documentation to reimbursement through a standardized scoring system managed by the Centers for Medicare and Medicaid Services (CMS).
How Risk Scores Are Calculated
Every enrollee in a risk-adjusted plan receives a numeric risk score based on two categories of information: demographics (age, sex, eligibility status) and diagnosed health conditions. Each qualifying diagnosis and demographic factor carries a specific weight reflecting its estimated contribution to future healthcare costs. The final risk score is the sum of all those weighted factors.
A score of 1.0 represents the average expected cost for a beneficiary. A patient with a score of 1.4 is predicted to cost 40% more than average, so the plan receives a proportionally higher payment. A relatively healthy patient might score 0.5, generating half the average payment. This is what makes the coding so financially significant: every condition that’s properly documented, coded, and submitted directly influences how much money flows to the health plan.
Hierarchical Condition Categories (HCCs)
The coding system at the center of risk adjustment uses Hierarchical Condition Categories, or HCCs. These are groups of related ICD-10 diagnosis codes organized by disease type and severity. Not every diagnosis code maps to an HCC. Only conditions that are significant predictors of future healthcare spending qualify.
The “hierarchical” part means that when multiple related conditions exist within the same disease group, only the most severe one counts toward the risk score. If a patient has both mild and severe forms of diabetes documented, the model captures the severe form and drops the mild one, preventing double-counting within the same category. Unrelated conditions across different categories do stack, though. A patient with diabetes, heart failure, and chronic kidney disease would accumulate the HCC weight for each.
In 2024, CMS began transitioning Medicare Advantage payments to a new version of the model (V28), which significantly reduced the number of diagnosis codes that map to an HCC while increasing the number of HCC categories used to calculate payments. This restructuring changed which conditions drive reimbursement and made precise coding even more critical for plans and providers.
Where Risk Adjustment Applies
Risk adjustment coding operates in two major insurance markets. Medicare Advantage plans use the CMS-HCC model, which was specifically built to predict costs for the Medicare population. The Affordable Care Act (ACA) marketplace uses a separate model called the HHS-HCC model, designed for the commercially insured population. Medicaid managed care plans also use risk adjustment in many states, though the specific models vary.
The underlying logic is the same across programs: measure disease burden through diagnosis codes, calculate a risk score, and adjust payments accordingly. The practical differences lie in which conditions are weighted, how heavily, and what data submission systems are used.
The Role of Clinical Documentation
Risk adjustment coding is only as accurate as the medical record behind it. A diagnosis that isn’t documented in the clinical note can’t be coded, and a condition that isn’t coded doesn’t exist in the risk adjustment model. This creates a direct link between what a provider writes during a patient visit and the financial health of the insurance plan.
The industry standard for adequate documentation is known by the acronym MEAT. For a condition to be validly captured for risk adjustment, the medical record needs to show that the provider did at least one of the following during the encounter:
- Monitored signs, symptoms, disease progression, or regression
- Evaluated test results, medication effectiveness, or response to treatment
- Assessed or addressed the condition through ordered tests, counseling, or record review
- Treated the condition with medications, therapies, or other interventions
Simply listing a diagnosis in the problem list without any supporting clinical activity in the visit note is not sufficient. The documentation must show that the provider actively managed or acknowledged the condition during that specific encounter. This distinction matters because risk adjustment conditions must be recaptured every calendar year. A diagnosis of heart failure documented in 2023 but never addressed in a 2024 visit won’t count toward the 2024 risk score, even if the patient still has heart failure.
The Annual Data Cycle
Risk adjustment operates on a structured annual timeline with multiple submission windows. Health plans submit diagnosis data to CMS through designated systems, and CMS runs the data through the risk adjustment model at specific points during the year to calculate and reconcile payments.
For a given payment year, the cycle typically works like this: an initial risk score run uses claims data from a window that begins six months before the payment year starts. If a plan misses the initial submission deadline, there’s a mid-year run that captures later-submitted data. A final reconciliation run closes out the payment year, incorporating a full calendar year of claims. For example, the 2026 initial run uses service dates from July 2024 through June 2025, with data due by September 2025. The 2026 final run covers all of calendar year 2025, with a submission deadline in early February 2027.
This staggered structure means that plans have multiple opportunities to capture and submit diagnoses, but each deadline is firm. Data submitted after 8 PM Eastern on a deadline date won’t be included in that particular run.
Audits and Compliance
Because risk adjustment directly controls payment amounts, CMS audits the data for accuracy. The primary audit program for Medicare Advantage is called Risk Adjustment Data Validation, or RADV. In a RADV audit, CMS selects a sample of enrollees, pulls their medical records, and checks whether the submitted diagnosis codes are actually supported by the documentation.
When an audit finds unsupported diagnoses, the plan must repay the resulting overpayments. Starting with payment year 2018, CMS began extrapolating audit findings, meaning the error rate found in a sample can be applied across the plan’s entire membership to calculate a total recovery amount. For audits covering payment years 2011 through 2017, CMS only collects on the specific unsupported codes found in the sample without extrapolation. Recovery amounts are deducted from the plan’s monthly payments, spread across multiple months if the total exceeds a single month’s payment.
CMS also eliminated what was known as the Fee-For-Service Adjuster, a factor that had previously offset some audit findings by accounting for coding differences between Medicare Advantage and traditional Medicare. Removing this adjuster means plans now face the full financial impact of any documentation gaps found during audits.
Who Does Risk Adjustment Coding
Risk adjustment coding involves multiple roles across the healthcare system. Medical coders with specialized training in HCC coding review clinical documentation and assign the appropriate ICD-10 codes. Many health plans employ risk adjustment coders who conduct retrospective chart reviews, looking through completed medical records to identify conditions that were clinically documented but may not have been coded on the original claim.
Providers play an equally important role on the front end. Physicians and advanced practice providers generate the documentation that coders rely on. Many health plans run provider education programs focused on helping clinicians understand which conditions matter for risk adjustment and what level of documentation is needed to support them. Some organizations also use clinical documentation improvement specialists who work alongside providers in real time to close documentation gaps before the record is finalized.
The financial stakes are substantial. Undercoding means a plan receives less than it needs to cover the actual cost of caring for its members. Overcoding, whether intentional or through sloppy documentation, creates audit liability and potential fraud exposure. Accurate risk adjustment coding sits at the intersection of clinical care, compliance, and financial sustainability.