EUVIMEDEUVIMED entdecken
BeobachtungsstudieOpen Access · CC BY-NC-ND 4.0

Enhancing Adverse Event Reporting With Clinical Language Models: Inpatient Falls

Insook Cho iD, Hyunchul Park, Byeong Sun Park, Dong-Geon Lee iD

Journal of Advanced Nursing · 2025

DOI: 10.1111/jan.16812 ↗

Kurz gefasst

Beobachtungsstudie in der Zeitschrift Journal of Advanced Nursing, 2025. Verfasst von Insook Cho und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Nursing, Inha University, Incheon, Republic of Korea. Hauptthemen laut MEDLINE: Accidental Falls, Inpatients, Patient Safety. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Research Foundation of Korea, Inha University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Aims

To develop a method for computationally detecting fall events using clinical language models to complement existing self‐reporting mechanisms.

Design

Retrospective observational study.

Methods

Text data were collected from the unstructured nursing notes of three hospitals' electronic health records and the Korean national patient safety reports, totalling 34,480 records covering the period from January 2015 to December 2019. Note‐level labelling was conducted by two researchers with 95% agreement. Preprocessing data anonymisation and English translation were followed by semantic validation. Five language models based on pretrained Bidirectional Encoder Representations from Transformers (BERT) and Generative Pretrained Transformer (GPT)‐4 with prompt programming were explored. Model performance was assessed using F measurements. Error analysis was conducted for the GPT‐4 results.

Results

Fine‐tuned BERT models with the English data set outperformed GPT‐4, with Bio+Clinical BERT achieving the highest F1 score of 0.98. Fine‐tuned Korean BERT with the Korean data set also reached an F1 score of 0.98, while GPT‐4 achieved a competitive F1 score of 0.94. GPT‐4 with prompt programming showed much higher F1 scores than GPT‐4 with a standardised prompt for the English data set (0.85 vs. 0.39) and the Korean data set (0.94 vs. 0.03). The error analysis identified that the common misclassification patterns included fall history and homonyms, causing false positives and implicit expressions and missing contextual information, causing false negatives.

Conclusion

The clinical language model approach, if used alongside the existing self‐reporting, promises to increase the chance of identifying the majority of factual falls without the need for additional chart reviews. Impact Inpatient falls are often underreported, with up to 91% of incidents missed in self‐reports. Using language models, we identified a significant portion of these unreported falls, improving the accuracy of adverse event tracking while reducing the self‐reporting burden on nurses. Patient or Public Contribution Not applicable.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Accidental FallsInpatientsPatient SafetyHumansRetrospective StudiesRepublic of KoreaFemaleMaleElectronic Health RecordsAgedMiddle AgedAdultLanguage

Stichwörter der Autor:innen

information technologyinpatient fallsnursing notesoutcomes measurementquality improvement

Autor:innen und Institutionen

  • College of Nursing, Inha University, Incheon, Republic of Korea.
  • The Center for Patient Safety Research and Practice, Division of General Internal Medicine, Brigham and Women's Hospital, Boston, Massachusetts, USA.
  • Seoul School of Integrated Science & Technologies, aSSIST University, Seoul, Republic of Korea.
  • Graduate School of Artificial Intelligence, Pohang University of Science and Technology, Gyeongbuk, Republic of Korea.
  • Graduate School of Information & Telecommunications, Konkuk University, Seoul, Republic of Korea.
  • Department of Electrical and Computer Engineering, Inha University, Incheon, Republic of Korea.

ORCID-Profile: Insook Cho, Dong-Geon Lee

Förderung

  • National Research Foundation of Korea
  • Inha University

Zitieren

APA

Cho, I., Park, H., Park, B. S., & Lee, D. G. (2025). Enhancing Adverse Event Reporting With Clinical Language Models: Inpatient Falls. Journal of Advanced Nursing. https://doi.org/10.1111/jan.16812

Vancouver

Cho I, Park H, Park BS, Lee DG. Enhancing Adverse Event Reporting With Clinical Language Models: Inpatient Falls. Journal of Advanced Nursing. 2025. doi:10.1111/jan.16812

BibTeX
@article{Cho2025,
  author = {Cho, Insook and Park, Hyunchul and Park, Byeong Sun and Lee, Dong-Geon},
  title = {Enhancing Adverse Event Reporting With Clinical Language Models: Inpatient Falls},
  journal = {Journal of Advanced Nursing},
  year = {2025},
  doi = {10.1111/jan.16812}
}
Kostenlos starten →