Verifiable summarization of electronic health records using large language models to support chart review
Journal of the American Medical Informatics Association · 13.11.2026
DOI: 10.1093/jamia/ocag181 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the American Medical Informatics Association, online erschienen am 13.11.2026. Verfasst von Ritchie Verma und 13 weiteren Autor:innen. Beteiligte Einrichtung: Massachusetts General Hospital, Harvard University, Boston, MA 02114, United States. Die Arbeit nennt 48 Literaturangaben. Gefördert von: Microsoft Research Accelerate Foundation Models Academic Research Program, NIH HHS, National Institutes of Health.
Abstract (Originalsprache)
Objectives
Information overload in electronic health records hampers clinicians’ ability to efficiently extract and synthesize critical information from a patient’s longitudinal health record, leading to increased cognitive burden and delays in care. This study explores the potential of large language models (LLMs) to address this challenge by generating problem-based admission summaries for patients admitted with heart failure, a leading cause of hospitalization worldwide. Materials and Methods We developed an extract-then-abstract approach guided by disease-specific “summary bundles” to generate summaries of longitudinal clinical notes that prioritize clinically relevant information. Through a mixed-methods evaluation using real-world clinical notes, we compared physicians’ ability to answer patient-specific clinical questions with the LLM-generated summaries vs standard chart review.
Results
While summary access did not significantly reduce overall questionnaire completion time, frequent summary use significantly contributed to faster questionnaire completion (P = .002). Individual physicians varied in how effectively they leveraged the summaries. Importantly, summary use maintained accuracy in answering clinical questions (88.0% with summaries vs 86.4% without). All physicians indicated they were “likely” or “very likely” to use the summaries in clinical practice, and 87.5% reported that the summaries would save them time. Preferences for summary format varied, highlighting the need for customizable summaries aligned with individual clinician workflows.
Discussion
This exploratory, mixed-methods pilot study provides one of the first extrinsic evaluations of LLMs for longitudinal summarization, demonstrating their potential to enhance clinician efficiency, alleviate workload, and support informed decision-making in time-sensitive care environments.
Conclusion
Our findings suggest that problem-based longitudinal summarization could improve chart review efficiency and standardize the presentation of clinical information, Variability in summary preferences and use underscores the need for user-centered evaluations to guide the development of LLM-generated summaries that align with clinician workflows.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Massachusetts General Hospital, Harvard University, Boston, MA 02114, United States.
- Brigham and Women's Hospital, Harvard University, Boston, MA 02115, United States.
ORCID-Profile: Ritchie Verma, Emily Alsentzer, Ozan Unlu, Jacqueline You, David Bates
Förderung
- Microsoft Research Accelerate Foundation Models Academic Research Program
- NIH HHS
- National Institutes of Health
Zitieren
APA
Verma, R., Alsentzer, E., Strasser, Z., Chang, L., Roman, K., Gershanik, E., Hernandez, C., Linares, M., Rodriguez, J., Thakral, D., Unlu, O., You, J., Zhou, L., & Bates, D. (2026). Verifiable summarization of electronic health records using large language models to support chart review. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocag181
Vancouver
Verma R, Alsentzer E, Strasser Z, Chang L, Roman K, Gershanik E, et al. Verifiable summarization of electronic health records using large language models to support chart review. Journal of the American Medical Informatics Association. 2026. doi:10.1093/jamia/ocag181
BibTeX
@article{Verma2026,
author = {Verma, Ritchie and Alsentzer, Emily and Strasser, Zachary and Chang, Leslie and Roman, Kirollos and Gershanik, Esteban and Hernandez, Camellia and Linares, Miguel and Rodriguez, Jorge and Thakral, Durga and Unlu, Ozan and You, Jacqueline and Zhou, Li and Bates, David},
title = {Verifiable summarization of electronic health records using large language models to support chart review},
journal = {Journal of the American Medical Informatics Association},
year = {2026},
doi = {10.1093/jamia/ocag181}
}
EUVIMED