What Is Risk Adjustment?

Risk adjustment is one of those terms that sounds technical but touches nearly every corner of modern healthcare financing. Whether you’re running a Medicare Advantage plan, managing an ACO, or leading a health system’s value-based care strategy, understanding how risk adjustment works is essential for sustainable operations and fair compensation.

At its core, risk adjustment is a statistical and financial process that adjusts payments to health plans or providers based on the expected cost of caring for their enrolled population. Rather than paying the same amount for every member regardless of health status, risk adjustment ensures that organizations caring for sicker, more complex patients receive appropriate resources to deliver quality care.

This guide breaks down what risk adjustment is, how it functions across major U.S. programs, why it matters for payers and providers alike, and what organizations can do to improve their risk adjustment processes.

Definition of Risk Adjustment

Risk adjustment in healthcare and health insurance is a methodology that predicts how much it will cost to care for a specific patient or population over a defined period. It translates an individual’s health status, demographics, and clinical conditions into a numerical value—the risk score—that directly influences payment.

The risk adjustment process uses concrete inputs to calculate these scores. Age, sex, and diagnosis codes documented in the medical record serve as the foundation. In some programs, pharmacy data, lab results, and even social determinants of health factor into the calculation. The goal is to create a number representing the relative expected cost of caring for that person compared to an average beneficiary.

In U.S. programs like Medicare Advantage and the ACA marketplaces, risk scores are typically normalized around 1.0. A score of 1.0 represents the average medicare patient—someone with typical expected healthcare costs for their program. Scores above 1.0 indicate higher expected costs, while scores below 1.0 suggest lower expected costs. This normalization creates a level playing field for comparing patients and populations.

Different programs use different risk adjustment models calibrated to their specific populations. CMS-HCC is the primary model for the medicare advantage program, while the HHS HCC model governs ACA individual and small-group markets. Medicaid programs, commercial payers, and various value-based arrangements often adapt these models or develop their own based on their member demographics.

Consider a practical example: a 75-year-old with congestive heart failure and diabetes with chronic complications will receive a substantially higher risk score than a healthy 30-year-old. This makes intuitive sense—the older patient with multiple chronic conditions will require more care, more medications, and more healthcare services. Risk adjustment ensures the plan or provider caring for that complex patient receives higher payments to cover those expected costs.

How Risk Adjustment Works in Major U.S. Programs

Risk adjustment underpins payment and evaluation across Medicare Advantage, ACA Marketplaces, and increasingly, value-based care arrangements throughout the healthcare system. While the specific models differ, all share a common purpose: ensuring that payments reflect the actual health risk of enrolled populations rather than rewarding organizations for avoiding sicker patients.

Central to how risk adjustment works is the concept of hierarchical condition categories. HCCs group clinically related ICD-10-CM diagnosis codes into categories with different weights based on their associated cost and severity. When a provider documents a diagnosis and that diagnosis appears on a claim, the corresponding ICD-10 code is mapped to an HCC if one exists. Each HCC carries a coefficient that adds to the patient’s overall risk score.

For Medicare Advantage, CMS uses a prospective model that relies on prior-year diagnoses to predict costs for the coming year. The 2024 CMS-HCC version 28 model includes 115 HCCs and is being phased in through 2026. This model determines capitated payments to MA plans based on the cumulative risk adjustment factor of their enrolled members. Diagnoses must be documented and submitted to the centers for medicare and medicaid services through encounter data to count toward payment.

The ACA individual and small-group markets operate differently. The HHS-HCC model is concurrent, meaning it uses diagnoses from the same calendar year to predict that year’s plan liability. Since 2014, this model has helped stabilize premiums by transferring funds from plans with healthier individuals to plans with higher-risk enrollment. The department of health and human services oversees this process through External Data Gathering Environment (EDGE) servers.

Not every diagnosis code influences risk scores. Of the nearly 74,000 ICD-10-CM codes available, only about 7,700 map to payment HCCs in the CY2024 CMS-HCC model. The remaining codes, while clinically important for treatment and documentation, do not directly affect risk adjustment payments. Understanding which conditions are risk-adjustable—and ensuring they’re properly documented—is critical for accurate risk adjustment.

Demographic factors also contribute to the calculation. Age and sex cells establish a baseline risk score before any diagnoses are added. Additionally, interaction terms capture the compounding effect of certain condition combinations. For example, diabetes plus heart failure together may carry additional weight beyond their individual HCC values, reflecting the documented reality that patients with multiple chronic conditions typically require more intensive care management.

Key Risk Adjustment Models and Scores

This subsection highlights the most widely used risk adjustment models and how they produce the risk scores that drive payment.

The risk adjustment factor, sometimes called the plan liability risk score, is the numerical output of a risk adjustment model. It represents the relative expected cost of a beneficiary compared to the average. A RAF of 1.2 suggests expected costs 20% above average, while a 0.8 indicates costs 20% below. For health plans, these scores directly translate to revenue through capitated payments.

CMS-HCC for Medicare Advantage determines how much Medicare pays MA plans monthly for each enrollee. The model requires that diagnoses come from acceptable encounter types—specific claim forms, provider types, and service categories qualify while others do not. Plans must submit complete and accurate encounter data to ensure proper coding is reflected in their members’ risk scores. The model is recalibrated periodically; major updates occurred in 2014, 2019, and 2024 to reflect changing cost patterns and clinical practices.

The HHS-HCC model for ACA markets uses 127 of its 264 HCCs in the payment calculation. These conditions were selected based on criteria including clinical meaningfulness, reliability of coding, and resistance to discretionary interpretation. The model focuses on conditions where accurate documentation genuinely reflects patient’s health complexity rather than coding practices. Plans participating in ACA markets receive risk adjustment transfers based on how their members’ risk scores compare to state and market averages.

Commercial payers and Medicaid programs frequently adapt these models or develop proprietary versions. Some incorporate pharmacy-based risk adjustment using prescription fill data to assign risk scores or validate diagnosis-based scores. Others integrate lab values, health risk assessments, or social risk factors to refine predictions. Medicaid services often use models like CDPS that weight conditions differently for disabled adults, non-disabled adults, and children.

Why Risk Adjustment Matters

Risk adjustment is foundational to fair payment, sustainable insurance markets, and credible performance measurement across healthcare. Without it, the financial incentives in health insurance would systematically disadvantage organizations serving sicker populations—and ultimately harm the patients who need care most.

By adjusting payments based on expected cost, risk adjustment reduces the incentive for risk selection—the practice of attracting healthier individuals while avoiding those with expensive health conditions. Before ACA risk adjustment was implemented, insurers had strong financial reasons to design products and marketing that discouraged enrollment by people with pre-existing conditions. Today, plans compete on quality and value rather than their ability to cherry-pick healthy members.

Risk adjustment stabilizes premiums in ACA individual and small-group markets by balancing payments among plans. If one health plan attracts members with higher-than-average risk scores (perhaps due to a strong specialty network), it receives transfer payments from plans with lower-risk enrollment. This mechanism has operated since 2014 and supports health care coverage for approximately 25 million people in individual markets who might otherwise face prohibitive costs.

In Medicare Advantage, where over 54% of Medicare beneficiaries now receive their benefits through private plans, risk adjustment determines how federal dollars flow. The roughly 32-33 million MA enrollees in 2024 represent trillions in federal outlays distributed based on cumulative risk scores. Plans serving members with heart failure, COPD, diabetes, and other chronic conditions receive higher payments to support the additional resources those patients compared to healthier members require.

Risk adjustment also enables meaningful performance measurement in value-based care. When comparing quality metrics, total cost of care, or shared savings across organizations, the patient populations being served must be accounted for. A primary care practice managing patients with multiple chronic conditions shouldn’t be penalized relative to one serving younger, healthier patients. Risk-adjusted benchmarks create appropriate care expectations aligned with actual patient complexity.

Inaccurate risk scores create problems in both directions. Understated scores lead to underpayment—organizations don’t receive adequate resources to care for their members, potentially affecting access and quality. Overstated scores represent overpayment and compliance risk, inviting regulatory scrutiny and potential enforcement action. Neither outcome supports the goal of delivering high-quality, sustainable healthcare.

Impact on Payers and Health Plans

For Medicare Advantage plans, ACA issuers, and Medicaid managed care organizations, risk adjustment directly determines revenue. Monthly capitated payments flow based on enrolled members’ cumulative RAF scores. A plan with average member scores of 1.1 receives roughly 10% more per member than one with average scores of 1.0. At scale, these differences represent hundreds of millions of dollars annually.

Accurate risk scores enable fair reimbursement for serving high-risk populations. Plans that develop strong clinical programs for members with heart failure, COPD, or diabetes with complications can be appropriately compensated for the greater medical services those members require. Without risk adjustment, such plans would face unsustainable losses that would force them to either exit markets or find ways to discourage high-need enrollment.

This financial impact shapes strategic decisions throughout plan operations. Network design, benefit structures, care management investments, and pricing strategies all account for expected member risk profiles. An MA plan serving dual eligibles (members eligible for both Medicare and Medicaid) will design very different operations than one serving primarily healthy retirees—and risk adjustment provides the financial basis for serving both populations.

Payers increasingly invest in analytics, documentation improvement programs, and provider education to ensure proper coding and complete risk capture. These investments aren’t about inflating scores inappropriately; they’re about ensuring that documented conditions are accurately reflected in submitted data so payments align with actual member needs.

Impact on Providers and Care Delivery

Risk adjustment influences provider payments in virtually every capitated or shared-savings arrangement. ACOs participating in MSSP or ACO REACH receive benchmarks adjusted for their attributed population’s risk. Primary care practices in CMS Primary Care First receive care management fees tied partly to panel-level complexity. Hospitals and health systems in global budget arrangements see their targets calibrated to patient risk.

For providers, accurate documentation and coding serve dual purposes. They support appropriate payment for the clinical work being performed, and they ensure that performance metrics fairly reflect patient complexity. A practice that manages many patients with congestive heart failure, diabetes with complications, and multiple chronic conditions should receive credit for that complexity in both payment and evaluation.

Consider a primary care practice participating in a value-based contract. If providers consistently document the full clinical picture—including all chronic conditions, their severity, and relevant complications—the practice’s attributed risk score will accurately reflect patient needs. This supports appropriate benchmark targets, fair quality comparisons, and adequate resources for care coordination and care management.

Poor documentation creates the opposite effect. A practice might appear inefficient or low-performing in cost metrics simply because its data doesn’t reflect the true patient’s risk of its panel. Meanwhile, the clinical team delivers appropriate care to complex patients but lacks the resources that accurate risk capture would provide.

Regulatory and Compliance Dimensions

U.S. regulators closely scrutinize risk adjustment submissions, particularly in Medicare Advantage. The Office of Inspector General, Department of Justice, and CMS itself have prioritized enforcement around suspected upcoding and unsupported diagnoses. False Claims Act cases have targeted plans alleged to have submitted diagnosis codes not adequately supported by the medical record.

This scrutiny reflects the substantial financial impact of risk adjustment. Every 0.01 increase in average RAF across 30+ million MA beneficiaries represents significant federal outlays. Regulators have a strong interest in ensuring that submitted data accurately reflects documented clinical reality rather than aggressive coding practices designed to maximize payment.

CMS is addressing some concerns through model changes. The 2024 CMS-HCC v28 model being phased in through 2026 tightens clinical logic for many conditions, reduces payment for some previously high-value codes, and aims to decrease susceptibility to discretionary interpretation. These changes place greater emphasis on specificity and clinical accuracy in documentation.

For health plans and providers, maintaining strong compliance programs is essential. This includes clear policies on acceptable documentation and coding practices, robust audit trails, regular internal reviews, and mechanisms for identifying and correcting errors. Organizations that invest in compliance infrastructure position themselves to withstand regulatory scrutiny while still achieving accurate risk adjustment through legitimate means.

Common Challenges in Risk Adjustment

Although the concept is straightforward—adjust payments based on patient health status—risk adjustment is operationally complex. Organizations struggle with data quality, documentation gaps, model complexity, and the constant tension between capturing accurate risk and avoiding compliance pitfalls.

The primary challenge areas span the entire risk adjustment lifecycle. Incomplete or inaccurate documentation means conditions aren’t captured in claims data. Fragmented data sources create blind spots where relevant clinical information exists but isn’t accessible. Complex models with hundreds of HCCs, demographic factors, and interaction terms require sophisticated analytics to manage. Evolving regulations demand constant updates to processes, training, and technology.

These challenges manifest differently across programs. Medicare Advantage emphasizes prospective annual capture, creating pressure to document chronic conditions every calendar year. ACA concurrent models require accurate same-year submissions through EDGE servers. Medicaid programs vary by state with different models and requirements. Despite these differences, the root causes—data quality issues and process maturity gaps—remain consistent.

Data Quality and Capture Gaps

Many risk-relevant conditions simply never appear in claims or encounter data. A patient may have well-documented diabetes in their EHR problem list, but if the diagnosis isn’t captured on a qualifying claim, it won’t contribute to the risk score. This disconnect between clinical reality and administrative data creates systematic undercapture of patient complexity.

Common data source problems include encounters billed with non-risk-adjustable codes, diagnoses recorded in clinical notes but never coded on claims, and services that don’t meet program requirements. For example, CMS-HCC historically required audio-video telehealth for certain encounter types—audio-only visits didn’t qualify even if clinically appropriate.

Some conditions appear only in member surveys, health risk assessments, or provider notes without ever reaching claims. Analytics might identify these as “suspect” conditions—likely present based on medication use, lab values, or clinical history, but missing from risk-adjusted data. For instance, a member taking insulin without any documented diabetes diagnosis represents a clear capture gap.

Chronic conditions pose particular challenges. Unlike acute events that naturally appear on claims when treated, chronic diseases like diabetes, heart failure, and COPD must be actively redocumented each year to count toward risk scores. The annual reset on January 1 means last year’s documented conditions don’t automatically carry forward. Plans and providers must ensure chronic conditions are evaluated and documented in qualifying encounters each calendar year—a significant operational burden.

Complex Models and Evolving Rules

Modern risk adjustment models are remarkably complex. CMS-HCC v28 includes 115 HCCs, multiple age/sex cells, numerous interaction terms, hierarchies that override lower-severity conditions, and detailed rules about which claim types, provider types, and service categories qualify. Understanding how a specific patient’s diagnoses translate to a risk score requires sophisticated analytics.

Frequent model updates compound the complexity. The 2024 CMS-HCC changes restructured how diabetes, mental health conditions, and vascular diseases are categorized. Statistical models were recalibrated using newer data. Some conditions increased in value while others decreased. Each update requires organizations to recalibrate their analytics, revise coding policies, and retrain clinical and administrative staff.

Eligibility rules add another layer. Not every diagnosis on every claim counts toward risk adjustment. CMS specifies acceptable claim forms, required data elements, qualifying provider types, and facility bill type requirements. These rules differ between Medicare and other programs. Frontline staff—coders, providers, billers—may struggle to track which encounters will actually contribute to risk scores.

Emerging factors like social determinants of health present additional complexity. Clinical evidence increasingly supports the importance of SDoH in predicting costs and outcomes. Z codes can capture housing instability, food insecurity, and other social risk factors. However, these codes often don’t map to payment HCCs, creating a gap between clinically relevant documentation and risk adjustment impact.

Compliance, Audit Risk, and Public Scrutiny

Organizations face real tension between closing legitimate risk gaps and avoiding practices that cross ethical or legal lines. The difference between appropriate capture of existing conditions and prohibited upcoding can seem subtle, but the consequences are significant.

Inspector general toolkits and targeted audits focus on high-risk areas with historically elevated error rates. Certain HCCs—particularly those with vague diagnostic criteria or high payment value—receive extra scrutiny. Organizations must document not just that a diagnosis appears in the record, but that it reflects a genuine clinical assessment made by a qualified provider during a face-to-face encounter.

The distinction between “suspect” conditions and documented diagnoses matters enormously. Analytics might flag that a patient likely has a condition based on medications, labs, or clinical patterns. But plans cannot simply add that diagnosis to claims—it must be confirmed through clinical evaluation and documented appropriately. Ensure proper coding means coding what providers have actually assessed and documented, not what analytics suggest might be present.

Public scrutiny has intensified. Media coverage, Congressional hearings, and policy debates have highlighted concerns about Medicare Advantage spending and the role of risk adjustment in driving costs. This visibility creates additional pressure on plans to demonstrate that their risk adjustment practices reflect genuine clinical reality rather than aggressive revenue optimization.

Related Solutions and Best Practices

Organizations can address risk adjustment challenges through better data infrastructure, clinical workflows, analytics capabilities, and compliance practices. The most effective approaches simultaneously support accurate payment, quality measurement, and patient-centered care rather than treating risk adjustment as purely a revenue exercise.

Moving from reactive chart reviews to proactive, integrated workflows represents the most significant opportunity for improvement. Rather than chasing diagnoses after encounters occur, leading organizations embed risk-relevant documentation into routine clinical care where it supports both coding accuracy and care management.

Clinical Documentation and Coding Excellence

Provider education on risk-relevant documentation forms the foundation of accurate risk adjustment. Clinicians must understand not just which conditions matter, but how to document them with the specificity that ICD-10-CM and HCC logic require. Diabetes “with complications” versus “without complications” carries different risk weight; documentation must clearly establish the clinical reality.

Annual wellness visits and comprehensive assessments provide natural opportunities to evaluate and document chronic conditions. A thorough annual review of a patient’s active problems—confirming current status, assessing severity, and documenting any complications—ensures that persisting conditions are captured in that year’s claims. This practice supports both risk adjustment and good clinical care by ensuring the care team has current information about each chronic condition.

Multidisciplinary workflows improve capture rates while reducing provider burden. Physicians, nurse practitioners, coders, nurses, and care managers each bring different perspectives to ensuring complete documentation. Pre-visit chart reviews can identify conditions that may need reassessment. Post-visit quality checks can catch documentation gaps before claims are submitted.

Prospective support within the EHR—problem list reconciliation tools, documentation prompts, structured templates for common conditions—reduces reliance on retrospective chart review. When clinicians have the right information and prompts at the point of care, they can document completely during the encounter rather than requiring subsequent outreach and amendment.

Integrated Data, Analytics, and “Suspecting”

Aggregating data from multiple sources creates a more complete picture than any single data source can provide. Combining EHR data, claims history, lab results, pharmacy fills, and member-reported information builds a longitudinal patient record that reveals gaps and patterns invisible in isolated datasets.

“Suspecting” workflows use analytics to identify likely but undocumented conditions. The example of insulin use without a diabetes diagnosis is classic—pharmacy data clearly suggests diabetes, but the condition may be missing from claims. Other examples include abnormal lab values (elevated HbA1c, reduced kidney function) without corresponding diagnoses, or medication patterns that suggest conditions not appearing in recent encounters.

These analytics-driven insights must feed into clinical workflows rather than just coding queues. The goal isn’t simply to add codes to claims; it’s to prompt clinical evaluation that confirms (or rules out) the suspected condition. When a clinician reviews a patient flagged for possible undocumented heart failure, assesses the patient, and documents findings, the resulting code is clinically legitimate and audit-defensible.

Effective solutions present insights where clinicians can act on them—in pre-visit summaries, EHR alerts, or care management worklists. Batch reports to coding teams after the fact are less effective than real-time information that shapes the encounter itself.

Governance, Compliance, and Continuous Improvement

Formal governance for risk adjustment should involve compliance leadership, clinical leadership, analytics teams, and revenue cycle stakeholders. This cross-functional structure ensures that risk adjustment strategies align with both financial objectives and ethical obligations.

Routine internal audits detect patterns in both directions. Under-coding leaves legitimate complexity uncaptured, affecting payment and performance measurement. Over-coding creates compliance risk even when unintentional. Peer comparisons—benchmarking coding patterns against similar organizations—help identify outliers that warrant investigation.

Standardized policies establish guardrails for acceptable practices. These should address how chart reviews are conducted, what types of provider outreach are appropriate, how documentation must support diagnoses, and how suspected conditions should be handled. Policies aligned with CMS and OIG guidance provide defensible frameworks for staff to follow.

Feedback loops to clinicians close the improvement cycle. Dashboards showing documentation quality, risk capture rates, and audit findings help providers understand the impact of their documentation practices. When clinicians see how incomplete documentation affects their panel’s risk profile—and by extension, resources for their patients—they’re more likely to invest in thorough documentation.

How Innovaccer Supports Risk Adjustment

Innovaccer is a healthcare data and analytics company focused on unifying data and enabling value-based care across payers and providers. The platform addresses many of the data and workflow challenges that undermine accurate risk adjustment.

At its foundation, Innovaccer aggregates data from EHRs, claims, labs, pharmacy systems, and other sources into a unified longitudinal member record. This integrated data source eliminates the blind spots that occur when clinical and administrative data remain siloed. Organizations gain visibility into their members’ complete health picture rather than fragments scattered across systems.

The platform’s analytics capabilities identify suspected but undocumented conditions, chronic disease gaps, and missed HCC opportunities. By analyzing patterns across claims, medications, labs, and clinical data, Innovaccer surfaces the high-value insights that enable accurate risk capture. These might include members with medication patterns suggesting undocumented conditions, chronic diseases not reassessed in the current year, or documentation that lacks the specificity needed for accurate coding.

Critically, Innovaccer delivers these insights at the point of care. Rather than generating reports for coding teams to review after encounters, the platform provides information through provider worklists, in-workflow prompts, and care management tools. Clinicians have the information they need to evaluate and document conditions during the encounter—supporting both risk accuracy and clinical care.

The platform supports multiple programs with program-specific logic. Whether an organization participates in Medicare Advantage, MSSP or ACO REACH ACOs, Medicaid managed care, or commercial value-based contracts, Innovaccer’s configurations reflect the relevant models, rules, and requirements. This eliminates the need for organizations to maintain separate tools for different program requirements.

Innovaccer’s approach connects risk adjustment with broader value-based care capabilities. Care management, quality improvement, and health equity analytics work from the same unified data foundation. Organizations can improve financial performance through accurate risk capture while simultaneously advancing patient outcomes and addressing disparities—rather than treating these as separate, potentially conflicting objectives.

Call to Action

Effective risk adjustment is essential for fair payment, sustainable value-based care, and better outcomes for high-risk populations. Organizations that achieve accurate risk capture can invest appropriately in care for their most complex members while meeting compliance obligations and demonstrating genuine value.

Health plans, ACOs, health systems, and provider groups seeking to strengthen their risk adjustment strategy should explore how unified data platforms and advanced analytics can close gaps, improve documentation workflows, and support both financial and clinical success.

Ready to transform your risk adjustment approach? Schedule a demo with Innovaccer to see how integrated data and intelligent analytics can help your organization achieve accurate, compliant, and sustainable risk adjustment.