Beyond the Code: Rethinking Integrity in Risk Adjustment

Beyond the Code: Rethinking Integrity in Risk Adjustment

Article Reference Code: NAMAS.09.25.2026

Written By: Alexis Isom CPC, CPMA, CRC, CHONC, AAPC Approved Instructor

For many healthcare professionals, today’s conversations surrounding Medicare Advantage risk adjustment are increasingly shaped by a news cycle of RADV audits, Department of Justice (DOJ) settlements, Office of Inspector General (OIG) findings, and heightened Centers for Medicare & Medicaid Services (CMS) oversight. Read enough headlines, and it’s easy to conclude that risk adjustment has become little more than a hunt for overpayments.

I Would Argue That Perspective Misses the Larger Story

Risk adjustment was designed to account for differences in disease burden because patients with greater healthcare needs often require more intensive management, closer monitoring, and greater coordination of care.

Diagnoses submitted to CMS become part of the clinical picture used to describe a population’s chronic disease burden. Hierarchical Condition Categories (HCCs) are the mechanism through which qualifying diagnoses are incorporated into the risk-adjustment model, contributing to risk scores used in risk-adjusted payment. The code, then, is not the endpoint. It becomes data, and that data is a representation of the population being served.

Therefore, Payment Integrity Must Run in Both Directions

Much of today’s regulatory attention and compliance infrastructure surrounding risk adjustment is conditioned to detect unsupported diagnoses and disproportionately focuses on the risk of overpayment. An unsupported diagnosis may overstate disease burden, while a clinically supported condition that is missed, inadequately documented, or reported without sufficient specificity may understate it. In either direction, the data may no longer accurately represent the patient’s health status and expected healthcare needs that the model is intended to reflect.

If we’re serious about Medicare risk-adjustment program integrity, it has to mean more than finding money that shouldn’t have been paid.

For coding professionals, technical proficiency has always been the floor, but this broader perspective expands the lens from coding accuracy to documentation integrity.

Knowing the Code	AND… Understanding the “Why”
Applying ICD-10-CM guidelines to the documented diagnosis.	Recognizing when accurate coding may still leave chronic conditions underrepresented.
Applying the specificity supported by documentation.	Recognizing when clinical indicators or unclear severity may warrant a query for clarification.
Verifying that CMS requirements for risk-adjustment reporting are met.	Understanding that both over- and under-reporting can compromise model integrity.
Completing an accurate code assignment.	Understanding how coded data represents chronic illness and informs reimbursement and resource planning.

 

Technically Accurate Coding Can Coexist with Incomplete Documentation

A provider may document, “Patient in today for follow-up of chronic kidney disease.”

A coder focused solely on code assignment may report N18.9, chronic kidney disease, unspecified, as documented. On that statement alone, the assignment is accurate.

Now consider that the record also contains an eGFR consistent with advanced renal disease. Although coders cannot independently interpret laboratory values and assign a CKD stage, we can recognize the opportunity for greater specificity and query the provider for clarification when clinically appropriate.

Without that clarification, the severity reflected elsewhere in the chart may never be represented in the diagnosis reported to CMS. Under CMS-HCC V28, N18.9, chronic kidney disease, unspecified, does not map to an HCC. Greater diagnostic specificity, when clinically supported and documented by the provider, can materially change how the patient’s CKD is represented within the risk-adjustment model and ultimately affect risk-adjusted payment. Nothing was necessarily coded incorrectly, yet the data leaving the encounter may still describe a less complex patient than the record itself suggests.

Consequences That Have Teeth

In May 2026, the OIG reported that CMS potentially overpaid Medicare Advantage organizations $462 million based on certain unsupported acute stroke diagnosis codes reported during the 2020 service year. The finding illustrates the opposite failure: instead of clinically supported complexity being insufficiently represented, diagnoses were reported

without corresponding evidence in the medical record. The direction of the error is different, but the integrity standard remains the same: the data should not represent patients as sicker or healthier than the medical record supports.

For organizations developing coding teams in this environment, that standard should influence how we educate, audit, and measure performance:

• Teach beyond the guideline. Risk adjustment coding professionals should understand what happens after a diagnosis is coded and reported and be equipped to translate that knowledge into meaningful provider education. Complete and specific documentation does more than support code assignment; it contributes to how the patient is represented within the risk-adjustment model and, ultimately, informs risk-adjusted payment intended to account for differences in expected healthcare costs.
• Audit for what is missing, not only what is wrong. Compliance programs must identify unsupported diagnoses, but documentation integrity also requires attention to clinically supported complexity that the medical record fails to fully communicate.
• Measure more than productivity. Charts per hour and turnaround times matter, but so does a coder’s ability to recognize documentation gaps, query appropriately, and exercise sound judgment when the record warrants a closer look.

Together, these practices strengthen the risk-adjustment lifecycle from the clinical encounter to the data ultimately submitted to CMS. As regulatory scrutiny intensifies, the opportunity is not simply to identify errors after they occur, but to build documentation and coding processes that preserve integrity at every stage.

References/Works Cited:

1. Office of Inspector General, US Department of Health and Human Services. CMS Potentially Overpaid Medicare Advantage Organizations $462 Million Based on Certain Unsupported Acute Stroke Diagnosis Codes. A-02-23-01020. May 28, 2026. https://oig.hhs.gov/reports/all/2026/cms-potentially-overpaid-medicare-advantage-organizations-462-million-based-on-certain-unsupported-acute-stroke-diagnosis-codes/
2. Centers for Medicare & Medicaid Services. Risk adjustment. Published August 14, 2023. https://www.cms.gov/priorities/innovation/key-concepts/risk-adjustment
3. Centers for Medicare & Medicaid Services. 2026 model software/ICD-10 mappings. https://www.cms.gov/medicare/payment/medicare-advantage-rates-statistics/risk-adjustment/2026-model-software-icd-10-mappings

---

Connect with Alexis on LinkedIn by Clicking her Name Below:

Alexis Isom CPC, CPMA, CRC, CHONC, AAPC Approved Instructor

---

NAMAS BLOG Disclaimer:

The NAMAS Blog features content written by both NAMAS staff and guest contributors. Guest contributors may present opinions or perspectives that differ from those officially instructed or encouraged by NAMAS. We believe in providing space for a range of informed viewpoints to foster dialogue, reflection, and deeper understanding within the auditing and compliance community.

Some contributors may use artificial intelligence (AI) tools in the development of their content. The decision to incorporate AI is left to the discretion of the author and does not reflect an endorsement or directive from NAMAS.

If you have questions, comments, or concerns about a specific blog post, we encourage you to contact the individual author directly. Their name and contact information are provided at the end of each post.