Healthcare providers and payers are increasingly turning to clinical natural language processing (NLP) to enhance their compliance with the Healthcare Effectiveness Data and Information Set (HEDIS). According to Tim O’Connell, Co-Founder and CEO of emtelligent, the critical insights necessary for compliance are often hidden within vast amounts of unstructured clinical data rather than in easily accessible fields. O’Connell, who is also a practicing radiologist, emphasizes that leveraging NLP can significantly improve data extraction and reporting, ultimately impacting financial performance and accreditation.
Bridging the Unstructured Data Gap
Over 80% of healthcare data remains unstructured, encompassing physicians’ notes, discharge summaries, and radiology images. This leaves only 20% of health data in structured formats, which makes it difficult to extract vital information for compliance and reporting. O’Connell cautions that relying solely on structured data is inadequate, akin to reading every fifth page of a book, where crucial context and details are inevitably lost. He notes that many health plans have suffered from lower HEDIS scores and reduced Star Ratings due to their inability to effectively utilize the wealth of information contained within unstructured data.
Medical NLP emerges as a transformative solution to this challenge. It enables organizations to sift through thousands of pages of clinical notes efficiently, reducing the painstaking manual effort traditionally required. O’Connell points out that this technology is not merely theoretical; a recent peer-reviewed study showed that 81% of organizations employing clinical NLP achieved over 90% accuracy in information extraction and classification. This high level of precision is a result of modern NLP’s capability to understand language nuances, chronology, and relationships, thus facilitating comprehensive quality reporting.
Strategic Implementation of NLP for Compliance
Having an effective NLP engine is just one aspect of achieving HEDIS compliance. Organizations that excel in this area approach their data strategy holistically. This includes developing robust infrastructure that integrates both structured and unstructured data while ensuring transparency from the outset. O’Connell advocates for a framework where technology not only supports healthcare professionals but also minimizes manual workloads and enhances patient outcomes.
The successful implementation of NLP can lead to tangible improvements, including better HEDIS scores and stronger Star Ratings. These advancements not only help maintain accreditation but also bolster financial performance, relieving organizations from constant operational strain. Investing in auditable and explainable data pipelines prepares healthcare entities for current requirements as well as future changes, addressing the complex landscape of clinical data management.
O’Connell asserts that unlocking the full potential of unstructured clinical data is both a technical and strategic challenge. Organizations that prioritize the extraction and analysis of this data, employing NLP to uncover hidden insights, will be better positioned to excel in compliance, audits, and overall care quality.
About Dr. Tim O’Connell
Dr. Tim O’Connell is the founder and CEO of emtelligent, a clinical-grade artificial intelligence solution. In addition to his leadership role, he serves as the vice-chair of clinical informatics at the University of British Columbia. His dual expertise in radiology and health technology uniquely positions him to address the pressing challenges faced by healthcare organizations in managing clinical data effectively.
