From Manual Burden to Intelligent Automation - How HealthEdge is Partnering with Health Plans to Modernize Retrospective Risk Adjustment

Modernizing Retrospective Risk Adjustment Coding with AI

Health plans process millions of medical charts annually for retrospective risk adjustment coding, a resource-intensive task designed to ensure accurate reimbursement for complex patient populations. This process, while critical, is often repetitive and manual, leading to high administrative costs and operational inefficiencies. HealthEdge® is changing this with an innovative, AI-powered solution that automates chart classification, reduces unnecessary reviews, and allows certified coders to focus on high-value work.

The Challenge of Traditional Retrospective Risk Adjustment Coding

Accurately capturing Hierarchical Condition Categories (HCCs) is a financial imperative for health plans. HCCs are part of a risk adjustment model from the Centers for Medicare & Medicaid Services (CMS) that helps predict patient costs and determine reimbursement based on chronic conditions. Missing or improperly documenting an HCC can lead to significant reimbursement losses.

The traditional approach to retrospective risk adjustment coding has several drawbacks:

• High Volume and Redundancy: Health plans often review millions of charts, sometimes sending the same chart to multiple vendors to ensure accuracy.

• Manual Burden: The process is heavily manual, requiring thousands of coders and significant operational oversight. This introduces human variability and fatigue.

• Inefficient Resource Allocation: A 2025 KFF analysis found that a vast majority of reviewed charts required little to no additional coding. This means valuable coder time is spent on low-intervention charts, increasing costs and delaying action on more complex cases.

The Solution: AI-Powered Chart Classification

HealthEdge developed its HCC Identifier solution to address the inefficiencies of manual review. By combining proprietary technology with advanced AI, the solution delivers content-driven chart classification with greater than 99% accuracy.

How Technology-First Chart Classification Works

The HCC Identifier solution rapidly processes both structured and unstructured medical record data to classify charts and route them to the most efficient workflow.

1. Seamless Integration: Health plans submit chart files through their existing workflows, requiring no process overhauls.

2. Automated Processing: The system uses multiple automated processing and validation layers to classify charts based on their content.

3. Intelligent Routing: Charts that don't require intervention are automatically flagged to eliminate unnecessary manual review, while more complex cases are routed to certified coders for expert analysis.
4. Human Oversight: Certified coders validate technology results, and ongoing quality assurance reviews ensure the model maintains high performance and accuracy over time.

This technology-first approach puts human expertise where it matters most. AI handles straightforward classifications, freeing up coders to focus on complex work that requires clinical expertise, such as untangling overlapping conditions or validating ambiguous documentation.

Proven Results and Financial Impact

To prove the model, HealthEdge partnered with a large national health plan for a pilot program that quickly scaled to millions of charts.

Partnering with a Health Plan

The initial pilot involved 10,000 Affordable Care Act (ACA) charts. The goals were to eliminate unnecessary human review for approximately 50% of charts and achieve greater than 99% technology accuracy.

• The Outcome: Following calibration and training, the HCC Identifier achieved 99.4% accuracy, validated by certified human coders.

• Scaling Success: The program scaled to more than two million charts across multiple health plan partners while maintaining performance outcomes exceeding 99% accuracy.

Lower Costs, Faster Decisions

By automating a significant portion of the review process, the solution provides clear financial benefits.

The Journey Toward Automated Coding at Scale

The economic and strategic impact of automating retrospective risk adjustment coding grows with chart volume. Assuming 50% of charts are removed from unnecessary review, the benefits become increasingly significant.

Looking Ahead: The Future of Risk Adjustment

The long-term vision is to move this intelligence closer to the point of care, empowering providers to identify and document conditions correctly the first time. This proactive approach would reduce the need for retrospective reviews and lower downstream administrative burdens for the entire industry. By combining AI-powered automation with human expertise, HealthEdge helps health plans transform retrospective risk adjustment coding from a cost center into a strategic opportunity to reduce waste, improve efficiency, and redirect resources toward better member care.