Risk adjustment matters because patients aren’t all equally healthy, and ignoring that reality distorts nearly everything in healthcare: how providers get paid, how hospitals are ranked, and whether insurance markets stay stable. Without it, a doctor’s office treating mostly elderly patients with multiple chronic conditions would receive the same payment as one treating young, healthy adults, creating a financial incentive to avoid sick patients altogether. Risk adjustment corrects for these differences by estimating how much it will cost to care for a given patient based on their specific health needs, then tying payments, quality scores, and resource planning to that estimate.
How Risk Adjustment Works
The most widely used system in U.S. healthcare is the Hierarchical Condition Categories (HCC) model, maintained by the Centers for Medicare & Medicaid Services. Here’s the basic logic: every diagnosis a patient has is coded using the ICD-10 system, and those codes are mapped to broader disease categories. Each category carries a numerical score reflecting how much future cost that condition typically drives. A patient’s scores are combined with demographic factors like age to produce a single number called a Risk Adjustment Factor (RAF) score.
That RAF score determines how much a health plan or provider is paid per patient in capitated or value-based payment arrangements. A higher score means more money follows the patient. Importantly, not every diagnosis triggers additional payment. Only about 10% of all ICD-10 codes map to “payment HCCs” that actually increase reimbursement. The rest are tracked but don’t change the dollar amount.
Fair Payment for Providers
The most straightforward reason risk adjustment exists is financial fairness. Providers who treat sicker, more complex patients cost more to operate. They need more staff time, more follow-up visits, more coordination with specialists. If payment doesn’t reflect that complexity, those providers face a choice: absorb losses or stop accepting high-need patients. Neither outcome is good for the healthcare system.
Risk adjustment removes this perverse incentive. A primary care practice managing patients with diabetes, heart failure, and chronic kidney disease receives higher per-patient payments than a practice whose panel is mostly healthy 30-year-olds. This doesn’t mean providers are rewarded for having sick patients. It means they aren’t penalized for it.
Making Hospital Quality Scores Meaningful
When you see a hospital ranked for readmission rates or mortality, those numbers have been risk-adjusted. Without that step, the rankings would be meaningless. A hospital in an affluent suburb would almost always outperform a safety-net hospital in a low-income urban area, not because the care is better, but because the patients walk in healthier.
Research published in JAMA Internal Medicine illustrates just how much patient characteristics move the needle. Among the factors that increase readmission risk, poverty stands out: hospitals in areas with higher poverty rates among residents 65 and older showed significantly elevated readmission numbers even before care quality entered the equation. The number of chronic conditions a patient carries is another powerful driver, as is disability status and whether the patient is also enrolled in Medicaid, a marker of low income. Without adjusting for these variables, pay-for-performance programs would systematically punish hospitals that serve disadvantaged communities.
Risk adjustment doesn’t eliminate all unfairness in quality measurement, but it narrows the gap enough that comparisons between hospitals become at least roughly useful for patients making choices and policymakers distributing penalties or bonuses.
Stabilizing Insurance Markets
In the individual and small-group insurance markets, risk adjustment plays a different but equally critical role. Health plans that happen to enroll sicker members face higher costs. Without a correction mechanism, those plans would need to raise premiums dramatically, pushing healthier members toward cheaper competitors. This triggers a cycle called adverse selection: the sick plan gets sicker, premiums spiral, and eventually the plan collapses.
Risk adjustment transfers funds from plans with lower-risk enrollees to plans with higher-risk enrollees, smoothing out the financial impact of uneven enrollment. This lets insurers compete on the quality and efficiency of their care networks rather than on how successfully they attract healthy people. It’s one of the structural pillars that keeps the Affordable Care Act marketplaces functioning.
Identifying Patients Who Need More Help
Beyond payment, risk scores serve as a practical tool for population health management. When a health system can see which patients carry the highest RAF scores, it can proactively direct resources toward them: care coordinators, medication management programs, home health visits, or chronic disease education. This is especially valuable in value-based care models where providers share financial risk. Catching a high-risk patient’s worsening condition early is better medicine and cheaper than an emergency admission later.
Risk scores also help health systems plan staffing and budgets. A clinic whose patient panel trends toward higher acuity can anticipate the volume of specialist referrals, lab work, and follow-up appointments it will need to handle in a given year.
The Role of Social Risk Factors
One of the most active debates in risk adjustment is whether social factors like income, housing stability, and education level should be folded into the models. Current HCC-based systems rely heavily on clinical diagnoses and demographics like age and sex. But social circumstances shape health outcomes in ways that clinical data alone can’t capture.
A federal analysis found that dual enrollment in Medicare and Medicaid (a proxy for low income) remains a powerful predictor of poor outcomes on quality and resource use measures, even after accounting for clinical risk. Functional status, meaning a person’s ability to perform daily tasks independently, is similarly predictive but isn’t consistently included in risk adjustment formulas. The Office of the Assistant Secretary for Planning and Evaluation has recommended that quality programs report outcomes separately for dually enrolled patients and include health equity measures, but has stopped short of recommending that quality scores be directly adjusted for social risk in public reporting. The concern is that doing so could mask genuine quality problems by giving low-performing providers a built-in excuse.
Accuracy and Oversight
Because risk adjustment directly determines payment, it creates an incentive to code aggressively. If adding a diagnosis to a patient’s record increases the RAF score, there is financial pressure to document every possible condition, and sometimes to document conditions that aren’t fully supported by the medical record. This is where oversight comes in.
CMS conducts Risk Adjustment Data Validation (RADV) audits of Medicare Advantage plans. During an audit, CMS pulls medical records and checks whether the diagnoses submitted for risk adjustment are actually supported by clinical documentation. If they aren’t, the agency can collect overpayments. These audits happen after the final risk adjustment data submission deadline for the contract year, giving plans time to review and correct their data first.
The system is also evolving. CMS began phasing in a new version of the HCC model (V28) in 2024, updating the diagnosis-to-category mapping and removing certain HCCs that were considered outdated or prone to overuse, including categories for protein-calorie malnutrition and certain forms of artery disease. These changes reflect an ongoing effort to keep the model clinically accurate and resistant to coding inflation.
What Happens Without Risk Adjustment
The simplest way to understand why risk adjustment matters is to imagine healthcare without it. Providers would be paid the same amount regardless of how sick their patients are, creating a race to attract the healthiest panels. Hospitals serving the poorest and sickest communities would be labeled as low quality based on raw outcome data, losing funding precisely when they need it most. Insurance markets would reward plans that cherry-pick healthy enrollees and punish those that don’t. Population health efforts would lack the data needed to target resources where they’d do the most good.
Risk adjustment is, at its core, an acknowledgment that health isn’t randomly distributed. People’s medical needs vary enormously based on their conditions, age, and circumstances. Any system that ignores those differences will produce unfair payments, misleading quality scores, and unstable markets. Risk adjustment doesn’t solve all of those problems perfectly, but it’s the mechanism that makes the rest of the system possible.