Rear top view of young African man in shirt using computer while working in the office

" data-large-file="https://i0.wp.com/keebler.health/wp-content/uploads/2025/10/iStock-1194668223.jpg?fit=1024%2C683&ssl=1" class="aligncenter wp-image-1533 size-large" src="https://i0.wp.com/keebler.health/wp-content/uploads/2025/10/iStock-1194668223.jpg?resize=1024%2C683&ssl=1" alt="" width="1024" height="683">

 

Even as value-based care models mature, many organizations still struggle to consistently capture the complete clinical picture of each patient. The Medicare Payment Advisory Commission (MedPAC) estimates that the extra coding intensity in the Medicare Advantage (MA) program—versus traditional fee-for-service (FFS) Medicare—results in a net 10% higher risk score for MA beneficiaries after the statutory adjustment.

Artificial intelligence (AI) is transforming this landscape—helping teams see what traditional workflows often miss. This article explores five of the most persistent challenges in HCC medical coding, and how AI-driven systems are redefining how organizations approach risk adjustment.

 

1. Missing or Incomplete Documentation

 

Documentation is one of the biggest barriers to accurate coding. Relevant diagnoses are often buried in unstructured clinical notes, outside records, or scanned documents—making it easy for conditions to go uncaptured.

AI can bridge these gaps by reading and interpreting both structured and unstructured data. Instead of relying solely on coders to locate relevant information, AI surfaces potential risk conditions with contextual evidence from across the record. The result is more accurate, comprehensive CMS-HCC coding and a truer reflection of patient complexity.

 

2. The Burden of Manual Chart Review

 

Even the most skilled coders face an uphill climb when reviewing charts manually. Each patient record can span hundreds of pages, with relevant evidence dispersed throughout. This process is slow, labor-intensive, and prone to oversight—especially as coding volumes rise.

AI-assisted chart review changes the equation:

• It automatically identifies potential HCC codes and supporting evidence.
• Coders can validate suggestions instead of searching line by line.
• Review time is dramatically reduced without sacrificing accuracy.

By automating retrieval, AI frees teams to focus on validation and quality assurance, improving both efficiency and coder satisfaction.

 

3. Audit and Compliance Risk

 

MedPAC projects that coding intensity contributes approximately $40 billion of the total $84 billion in estimated over-payments to MA plans. With CMS intensifying its focus on documentation integrity, compliance has never been more critical. Unsupported codes can trigger audit findings, payment recoupments, and long-term reputational harm.

AI can strengthen audit readiness by:

• Linking every suggested code to its clinical evidence source.
  Tracking coding decisions with full transparency.
• Maintaining a digital audit trail aligned with CMS-HCC guidelines.

When organizations can clearly trace each diagnosis to its origin, they gain both defensibility and confidence in their data.

 

4. Fragmented Data Across Systems

 

HCC coding accuracy depends on having a complete view of each patient’s medical history. Yet many organizations are still working with fragmented data—split across multiple EHRs, specialist networks, and claims systems.

AI helps unify these data streams into a longitudinal, patient-centered record. By connecting structured and unstructured inputs through a FHIR-first framework, AI enables a Whole Patient Story—a single, integrated view of all relevant encounters, diagnoses, and evidence. That continuity improves not only coding accuracy but also clinical understanding.

 

5. Keeping Pace with CMS Model Changes

 

Each update to the CMS-HCC model—most recently Version 28 (v28)—reshapes how risk scores are calculated and which conditions are prioritized. Manually tracking these changes and applying them across systems is a major operational challenge.

AI systems can dynamically adapt to regulatory and model updates. As new versions are released, algorithms automatically adjust code groupings and weightings, ensuring consistency across patient populations. This capability allows organizations to stay compliant, maintain revenue integrity, and respond quickly to evolving risk adjustment rules.

Redefining Risk Adjustment Through Intelligence

The future of HCC coding lies in intelligence, not automation alone. AI’s greatest value is in amplifying human expertise—helping coders, clinicians, and compliance teams work from the same, complete dataset.

Organizations adopting AI-driven risk adjustment gain:

• Greater coding accuracy and documentation completeness.
• Faster turnaround with reduced manual review.
• Stronger audit readiness and transparency.
• Future-proofing for CMS-HCC V28 and beyond.

As the healthcare system continues its shift toward value-based care, one fact remains clear: accurate coding begins with understanding the patient as a whole. AI makes that possible—at scale, in real time, and with confidence.

 

Close every gap. Capture every code.

 

Keebler’s AI-driven platform captures every record—structured, unstructured, and everything in between—to deliver accurate, compliant insights across your population. Contact us to start coding with confidence today.

Get started with a Risk Free Trial today!