As value-based care arrangements proliferate throughout the healthcare industry, accurate risk adjustment has become more crucial for payers to ensure they are properly compensated for assuming greater financial risk for patients.
In many value-based contracting agreements, and notably Medicare Advantage, health plans receive set monthly payment amounts to manage the care of their members. Those monthly payment amounts are determined by risk scores that are associated specific clinical diagnoses, which are represented in claims data as ICD10-CM codes, bundled into various diagnostic groups called Hierarchical Condition Categories (HCCs).
When payers don’t capture the full spectrum of a patient’s diagnoses, they may be at risk for cost overruns associated with treating those unidentified conditions. As a result, it is critical for payers to gain 360-degree visibility into patient electronic health records (EHRs) to identify all diagnoses that may point to underlying conditions.
The challenge is that much of this key data is buried as unstructured text in patient records, making the information difficult to extract for risk adjustment. Indeed, it’s estimated that about 80% of medical data remains unstructured and untapped after creation.
Traditionally, to comb through patient records and obtain key risk-adjustment data, payers have relied on expensive and time-consuming chart reviews, in which medical workers manually read through pages upon pages of patient records to spot diagnoses. In other cases, payers outsource these reviews to third-party companies.
Today, more payers are looking to automate these manual processes using natural language processing (NLP) to capture key diagnostic data and risk-adjustment information. NLP is an artificial-intelligence-based technology that extracts and synthesizes the high-value information hidden in unstructured text. NLP-based text mining solutions can analyze unstructured text to identify the key facts, interpret the meaning, and extract and present facts in a structured form for payers to review, analyze, and summarize.
Why natural language processing is the right choice for risk adjustment
Our Natural Language Processing (NLP) Risk Adjustment solution enables payers and providers to connect technology with subject matter expertise in an easy-to-use workflow. This more accessible approach ensures a higher level of confidence in an organization’s risk adjustment submissions.
- Improves coding accuracy with > 90% precision and recall
- Significantly reduces medical chart review time
- Identifies and improves risk adjustable comorbid diagnoses with automated and semi-automated disease coding
- Provides a comprehensive audit trail of accepted ICD10-CM codes to support claims submitted to CMS.
In addition, the NLP Risk Adjustment solution is scalable, with the ability to process millions of records an hour. Dashboard tools provide timely, reliable intelligence that enriches coding teams’ analysis and tracks progress and productivity, while advanced AI-powered chart extraction drives accurate and actionable codes from unstructured data to reduce the risk of errors. Medical chart review time is further reduced with the solution’s intuitive one-click acceptance of suggested HCC or ICD10-CM codes.
Accurate risk adjustment is essential for payers in value-based care and Medicare Advantage plans. With our proven, flexible, and customizable NLP tools, payers can identify care gaps, improve documentation accuracy, and perform prospective reviews to lower costs and generate better patient outcomes.