AI-Driven Risk Adjustment for Value-Based CareAI-Driven Risk Adjustment for Value-Based Care

The client is a US-based healthcare technology organization within a venture-backed provider growth platform that has invested over $6 billion in healthcare assets. The platform serves nearly 100,000 healthcare professionals through strategic partnerships and technology-enabled solutions. They focus on AI-powered coding, audit, training, and risk adjustment solutions for Medicare Advantage and commercial payor models. Their platform supports revenue cycle operations while helping healthcare teams improve compliance, documentation accuracy, and financial outcomes.

Accurate HCC coding plays a major role in reimbursement, care planning, population risk visibility, and value-based care performance. The client’s existing platform supported CPT coding based on medical decision-making and time spent. Stronger capabilities were needed for longitudinal diagnosis recapture, RAF scoring, and risk-adjusted documentation.

Key Challenges

The HCC risk adjustment process was manual, inconsistent, and difficult to scale.

1. Only 40% year-over-year capture of previously coded HCC conditions led to patient complexity, reduced reimbursement visibility, and challenges in population risk assessment.
2. Providers and coders relied on retrospective audits and correction cycles.
3. CMS V28 and hybrid scoring rules added complexity to coding decisions.
4. Missed diagnoses reduced RAF accuracy and reimbursement potential.

The client sought a partner with knowledge of healthcare processes, AI-native engineering capabilities, experience in secure data architecture, and strong EHR integration expertise.

Coditas brought a controlled risk adjustment approach using longitudinal diagnosis mapping, real-time RAF calculation, HCC scoring logic, FHIR and HL7 ingestion, HIPAA-compliant Azure infrastructure, role-based access control, and audit-ready interaction logging.

Coditas built a secure AI-powered HCC diagnosis assistant to support documentation, RAF scoring, and diagnosis recapture inside the EHR experience. The system processed three years of historical EHR data, including diagnoses, claims, encounters, and patient metadata.

The assistant identified chronic conditions requiring current-year recapture, applied hybrid CMS scoring logic using 67% V24 and 33% V28, and suggested more specific ICD-10 codes for provider review. Providers could accept, dismiss, or replace diagnoses while viewing RAF score changes in real time. The assistant also added accepted codes to the Assessment and Plan section with compliant documentation prompts.

React embedded UI components, Python orchestration service, custom HCC scoring API, Azure-hosted Diagnosis Data Warehouse, FHIR and HL7 ingestion pipelines, JSON-based HCC logic compatibility, HIPAA-aligned Azure infrastructure, role-based access control, scheduled ETL jobs, and DDI-powered V24 and V28 RAF scoring logic

The engagement improved HCC recapture, RAF precision, coding efficiency, audit readiness, and value-based care readiness. Key outcomes included

• 4X improvement in accurate HCC recapture
• 42% increase in RAF score precision
• 50% reduction in post-encounter coding edits
• 60% reduction in provider-coder revision cycles
• Real-time RAF visibility at the point of care
• Audit-ready logging for diagnosis suggestions, provider actions, RAF deltas, and final coding outcomes

Risk adjustment is now central to the success of value-based care. Incomplete diagnosis recapture affects reimbursement, patient risk visibility, and payer confidence.

Coditas tackled the problem with a more accurate, compliant, and scalable risk adjustment model through AI native engineering, interoperability, secure cloud architecture, and provider-centered design.