Increase Confidence in Your Retrospective Risk Adjustment Program Performance
Apixio’s AI-powered risk adjustment solutions for retrospective coding and auditing help teams scale their chart review processes to capture a complete, defensible picture of patient risk.
![](https://www.apixio.com/wp-content/uploads/2023/11/d-Risk-Adjustment-Sub-1-@2x.png)
![](https://www.apixio.com/wp-content/uploads/2023/11/m-Risk-Adjustment-Sub-1-@2x.png)
Trustworthy Intelligence
Unmatched Efficiency
Scalable Technology
Flexible Services
The Impact
Reduction
80% reduction in coding and audit workload
Increase
23% accuracy increase with AI-assisted chart reviews
Accuracy
96%+ coding accuracy with up to 3 levels of QA
Features
AI-powered net-new HCC code identification
1-click evidence review and QA workflows
3+ levels of QA reviews
Targeted review workflows for potentially unsupported codes
Clear audit trail of chart reviews and results
Reviewer progress, productivity, and accuracy tracking
AI-as-a-Service for Retrospective Risk Adjustment
![](https://www.apixio.com/wp-content/uploads/2023/10/Testimonial-Image-Businesswoman-at-Office-Using-Desktop.jpg)
Complete, Accurate Coding and Auditing with Less Manual Effort
Regional Health Plan
Retrospective Risk Adjustment
Retrospective risk adjustment enables health plans and risk-bearing providers to retrospectively analyze previous claims. This process helps detect unreported or inaccurately submitted HCC codes that are substantiated by medical records.
Retrospective HCC coding is the process of uncovering any HCC codes missed during the submission based on evidence in the documentation and codes not substantiated in the documentation.
Artificial intelligence (AI) enables coders to analyze structured and unstructured data and quickly identify evidence in clinical documentation of missing codes previously unknown. In addition, AI can also identify codes without clinical evidence in the documentation that needs to be audited. To enhance this process, AI utilizes Natural Language Processing (NLP) and Machine Learning (ML). These techniques aid in grasping the clinical context and refining models for improved accuracy.
The V28 model is a revised 2024 CMS-HCC risk adjustment model with several changes from the V24 model, driven by the need to reflect better specificity and clinical relevance of diagnoses under the ICD-10 system. There was an expansion in the HCCs from 86 in V24 to 115 in V28, but 2,236 ICD-10 codes were no longer mapped to HCCs.
Apixio provides initial and follow-up coding and auditing services for retrospective risk adjustment, utilizing our AI-powered platform to ensure fast, accurate, and reliable results.