An Extraction Tool for Venous Thromboembolism Symptom Identification in Primary Care Notes to Facilitate Electronic Clinical Quality Measure Reporting: Algorithm Development and Validation (Preprint)
Authors:
- John Novoa-LaurentievJNJohn Novoa-Laurentiev
- Mica BowenMBMica Bowen
- Avery PullmanAPAvery Pullman
- Wenyu SongWSWenyu Song
- Ania SyrowatkaASAnia Syrowatka
- Jin ChenJCJin Chen
- Michael SainlaireMSMichael Sainlaire
- Frank ChangFCFrank Chang
- Krissy GrayKGKrissy Gray
- Purushottam PantaPPPurushottam Panta
- Luwei LiuLLLuwei Liu
- Khalid NawabKNKhalid Nawab
Preprint
- 28 June 2024
- preprint
- Published by JMIR Publications Inc. in JMIR Preprints
Abstract
BACKGROUND Diagnosis of venous thromboembolism (VTE) is often delayed and facilitating earlier diagnosis may improve associated morbidity and mortality. Clinical notes contain information not found elsewhere in the medical record that could facilitate timely VTE diagnosis and accurate quality measurement. However, extracting relevant information from unstructured clinical notes is complex. Today there are relatively few electronic clinical quality measures (eCQMs) in our national payment program and none that use NLP for data extraction. NLP holds great promise for making quality measurement more accurate and more efficient. OBJECTIVE We developed a rule-based NLP tool, VTExt, that extracts VTE symptoms from clinical note text, for use within an eCQM to quantify the rate of delayed diagnosis of VTE in primary care settings. METHODS We iteratively developed VTExt on an internal dataset using a rule-based approach to extract VTE symptoms from primary care clinical note text. The VTE symptoms lexicon was derived and modified with physician guidance and externally validated using data from two external healthcare organizations. RESULTS VTExt achieved near-perfect performance in extracting VTE symptoms from primary care notes sampled from records of patients diagnosed with and without VTE. In external validation VTExt achieved promising performance in two additional geographically distant organizations using different electronic health record systems. When compared against a deep learning-based model, VTExt exhibited similar or improved performance across all metrics. CONCLUSIONS This study demonstrates a data-driven NLP-based approach to clinical note information extraction that can be generalized to different electronic health record (EHR) systems across different institutions. VTExt is the first NLP application to be used in a nationally endorsed eCQM.Keywords
This publication has 58 references indexed in Scilit:
- Association of Hospital Public Quality Reporting With Electronic Health Record Medication Safety PerformanceJAMA Network Open, 2021
- Using Electronic Health Records for Surgical Quality Improvement in the Era of Big DataJAMA Surgery, 2015
- Venous Thromboembolism2012
- THE HITECH ACT—An OverviewAMA Journal of Ethics, 2011
- Risk Factors for Venous Thromboembolism in Hospitalized Patients With Acute Medical IllnessArchives of internal medicine (1960), 2004
- Epidemiology of venous thromboembolism in a Chinese populationBritish Journal of Surgery, 2004
- Relationship Between Deep Venous Thrombosis and the Postthrombotic SyndromeArchives of internal medicine (1960), 2004
- Trends in the Incidence of Deep Vein Thrombosis and Pulmonary EmbolismArchives of internal medicine (1960), 1998
- Deep Vein Thrombosis and Pulmonary Embolism in the General PopulationArchives of internal medicine (1960), 1997
- A Population-Based Perspective of the Hospital Incidence and Case-Fatality Rates of Deep Vein Thrombosis and Pulmonary EmbolismArchives of internal medicine (1960), 1991