Comparison of six natural language processing approaches to assessing firearm access in Veterans Health Administration electronic health records
Journal of the American Medical Informatics Association · 01.01.2025 · 32 (1) · 113-118
DOI: 10.1093/jamia/ocae169 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift Journal of the American Medical Informatics Association, online erschienen am 12.11.2024. Verfasst von Joshua Trujeque und 14 weiteren Autor:innen. Beteiligte Einrichtung: Division of Pulmonary, Allergy, Critical Care and Sleep Medicine, Department of Medicine, University of Minnesota Medical School, Minneapolis, MN 554…. Hauptthemen laut MEDLINE: Natural Language Processing, Firearms, Electronic Health Records, United States Department of Veterans Affairs. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NIA NIH HHS, NHLBI NIH HHS, U.S. Department of Veterans Affairs Office of Suicide Prevention.
Abstract (Originalsprache)
Objective
Access to firearms is associated with increased suicide risk. Our aim was to develop a natural language processing approach to characterizing firearm access in clinical records. Materials and Methods We used clinical notes from 36 685 Veterans Health Administration (VHA) patients between April 10, 2023 and April 10, 2024. We expanded preexisting firearm term sets using subject matter experts and generated 250-character snippets around each firearm term appearing in notes. Annotators labeled 3000 snippets into three classes. Using these annotated snippets, we compared four nonneural machine learning models (random forest, bagging, gradient boosting, logistic regression with ridge penalization) and two versions of Bidirectional Encoder Representations from Transformers, or BERT (specifically, BioBERT and Bio-ClinicalBERT) for classifying firearm access as “definite access”, “definitely no access”, or “other”.
Results
Firearm terms were identified in 36 685 patient records (41.3%), 33.7% of snippets were categorized as definite access, 9.0% as definitely no access, and 57.2% as “other”. Among models classifying firearm access, five of six had acceptable performance, with BioBERT and Bio-ClinicalBERT performing best, with F1s of 0.876 (95% confidence interval, 0.874-0.879) and 0.896 (95% confidence interval, 0.894-0.899), respectively. Discussion and Conclusion Firearm-related terminology is common in the clinical records of VHA patients. The ability to use text to identify and characterize patients’ firearm access could enhance suicide prevention efforts, and five of our six models could be used to identify patients for clinical interventions.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Division of Pulmonary, Allergy, Critical Care and Sleep Medicine, Department of Medicine, University of Minnesota Medical School, Minneapolis, MN 55455, United States.
- Center for Care Delivery and Outcomes Research (CCDOR), Veterans Affairs (VA) Minneapolis Healthcare System, Minneapolis, MN 55417, United States.
- Institute for Health Informatics, University of Minnesota, Minneapolis, MN 55455, United States.
- School of Public Health, University of Minnesota, Minneapolis, MN 55455, United States.
- Medical School, University of Minnesota, Minneapolis, MN 55455, United States.
- Department of Surgery, University of Minnesota, Minneapolis, MN 55455, United States.
ORCID-Profile: Joshua Trujeque, Nicholas E Ingraham, Isai Ortiz, Rui Zhang
Förderung
- NIA NIH HHS
- NHLBI NIH HHS
- U.S. Department of Veterans Affairs Office of Suicide Prevention
- HSRD VA
- NIDDK NIH HHS
- VA
- AHRQ HHS
- University of Minnesota Medical School and Clinical and Translational Science Institute
Zitieren
APA
Trujeque, J., Dudley, R. A., Mesfin, N., Ingraham, N. E., Ortiz, I., Bangerter, A., Chakraborty, A., Schutte, D., Yeung, J., Liu, Y., Woodward-Abel, A., Bromley, E., Zhang, R., Brenner, L. A., & Simonetti, J. A. (2025). Comparison of six natural language processing approaches to assessing firearm access in Veterans Health Administration electronic health records. Journal of the American Medical Informatics Association, 32(1), 113-118. https://doi.org/10.1093/jamia/ocae169
Vancouver
Trujeque J, Dudley RA, Mesfin N, Ingraham NE, Ortiz I, Bangerter A, et al. Comparison of six natural language processing approaches to assessing firearm access in Veterans Health Administration electronic health records. Journal of the American Medical Informatics Association. 2025;32(1):113-118. doi:10.1093/jamia/ocae169
BibTeX
@article{Trujeque2025,
author = {Trujeque, Joshua and Dudley, R Adams and Mesfin, Nathan and Ingraham, Nicholas E and Ortiz, Isai and Bangerter, Ann and Chakraborty, Anjan and Schutte, Dalton and Yeung, Jeremy and Liu, Ying and Woodward-Abel, Alicia and Bromley, Emma and Zhang, Rui and Brenner, Lisa A and Simonetti, Joseph A},
title = {Comparison of six natural language processing approaches to assessing firearm access in Veterans Health Administration electronic health records},
journal = {Journal of the American Medical Informatics Association},
year = {2025},
volume = {32},
number = {1},
pages = {113--118},
doi = {10.1093/jamia/ocae169}
}
EUVIMED