Text-based fall prediction in hospital: Development and internal validation of a model to predict in-hospital falls in older patients using free text from daily nursing records
Marjolein Groeneveld , Wendy L.M. Leurs , Arthur R.A. Bouwman , Jimmy Schenk , Loes Lammers , Angelique Dierick , Erik Korsten , Carolien C.M.J. van der Linden
{"title":"Text-based fall prediction in hospital: Development and internal validation of a model to predict in-hospital falls in older patients using free text from daily nursing records","authors":"Marjolein Groeneveld , Wendy L.M. Leurs , Arthur R.A. Bouwman , Jimmy Schenk , Loes Lammers , Angelique Dierick , Erik Korsten , Carolien C.M.J. van der Linden","doi":"10.1016/j.apnr.2025.151923","DOIUrl":null,"url":null,"abstract":"<div><h3>Aim</h3><div>The aim of this study was to evelop a predictive model, estimating the probability of an in-hospital fall using previously identified associated words, and word combinations in daily nursing records. To assess the difference in discriminatory ability between the predictive model and currently used screening questions.</div></div><div><h3>Background</h3><div>Hospital falls are a persistent challenge. Identifying patients at high risk before fall incidents occur is essential to optimize preventive measures and reduce the burden on nursing staff.</div></div><div><h3>Method</h3><div>Words from daily nursing records were used as predictive variables to construct and validate the model. The DeLong's test was used to determine statistical differences between the developed model and the current screening questions.</div></div><div><h3>Results</h3><div>A total of 3255 consecutive admissions of patients aged 70 and over were included, of whom 110 experiences a fall. Upon internal validation, the predictive text model demonstrated moderate discriminatory ability (AUC-ROC 0. 737 (CI 95 % 0. 683–0.791)) and good calibration across a range of the risk groups. Compared to the screening questions (AUC-ROC 0.603 (CI 95 % 0.555–0.652)) the text model (AUC-ROC 0.734 (CI 95 % 0.679–0.788)) showed significantly better discriminatory ability (DeLong's − 3.93, <em>p</em> ≤0.001).</div></div><div><h3>Conclusion</h3><div>Daily nursing records can be used to estimate the probability of in-hospital falls. A text-based predictive model outperforms the currently employed screening questions and provides insights for the efficient use of fall prevention interventions. Further research should focus on improving the accuracy and external validation of the model and implementation strategies.</div></div>","PeriodicalId":50740,"journal":{"name":"Applied Nursing Research","volume":"82 ","pages":"Article 151923"},"PeriodicalIF":2.7000,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Nursing Research","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0897189725000254","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"NURSING","Score":null,"Total":0}
引用次数: 0
Abstract
Aim
The aim of this study was to evelop a predictive model, estimating the probability of an in-hospital fall using previously identified associated words, and word combinations in daily nursing records. To assess the difference in discriminatory ability between the predictive model and currently used screening questions.
Background
Hospital falls are a persistent challenge. Identifying patients at high risk before fall incidents occur is essential to optimize preventive measures and reduce the burden on nursing staff.
Method
Words from daily nursing records were used as predictive variables to construct and validate the model. The DeLong's test was used to determine statistical differences between the developed model and the current screening questions.
Results
A total of 3255 consecutive admissions of patients aged 70 and over were included, of whom 110 experiences a fall. Upon internal validation, the predictive text model demonstrated moderate discriminatory ability (AUC-ROC 0. 737 (CI 95 % 0. 683–0.791)) and good calibration across a range of the risk groups. Compared to the screening questions (AUC-ROC 0.603 (CI 95 % 0.555–0.652)) the text model (AUC-ROC 0.734 (CI 95 % 0.679–0.788)) showed significantly better discriminatory ability (DeLong's − 3.93, p ≤0.001).
Conclusion
Daily nursing records can be used to estimate the probability of in-hospital falls. A text-based predictive model outperforms the currently employed screening questions and provides insights for the efficient use of fall prevention interventions. Further research should focus on improving the accuracy and external validation of the model and implementation strategies.
期刊介绍:
Applied Nursing Research presents original, peer-reviewed research findings clearly and directly for clinical applications in all nursing specialties. Regular features include "Ask the Experts," research briefs, clinical methods, book reviews, news and announcements, and an editorial section. Applied Nursing Research covers such areas as pain management, patient education, discharge planning, nursing diagnosis, job stress in nursing, nursing influence on length of hospital stay, and nurse/physician collaboration.