Predicting future falls in older people using natural language processing of general practitioners’ clinical notes

Author:

Dormosh Noman12,Schut Martijn C134,Heymans Martijn W56,Maarsingh Otto78,Bouman Jonathan9,van der Velde Nathalie1011,Abu-Hanna Ameen12

Affiliation:

1. Amsterdam UMC location University of Amsterdam Department of Medical Informatics, , Amsterdam, The Netherlands

2. Amsterdam Public Health, Aging and Later Life & Methodology Amsterdam , Amsterdam, The Netherlands

3. Amsterdam UMC location Vrije Universiteit Amsterdam Department of Clinical Chemistry, , Amsterdam, The Netherlands

4. Amsterdam Public Health, Methodology & Quality of Care , Amsterdam, The Netherlands

5. Amsterdam UMC location Vrije Universiteit Amsterdam Department of Epidemiology and Data Science, , Amsterdam, The Netherlands

6. Amsterdam Public Health, Methodology & Personalized Medicine , Amsterdam, The Netherlands

7. Amsterdam UMC location Vrije Universiteit Amsterdam Department of General practice, , Amsterdam, The Netherlands

8. Amsterdam Public Health, Aging and Later Life & Mental Health , Amsterdam, The Netherlands

9. Amsterdam UMC location University of Amsterdam Department of General Practice, , Amsterdam, The Netherlands

10. Amsterdam UMC location University of Amsterdam Department of Internal Medicine, Section of Geriatric Medicine, , Amsterdam, The Netherlands

11. Amsterdam Public Health, Aging and Later Life , Amsterdam, The Netherlands

Abstract

AbstractBackgroundFalls in older people are common and morbid. Prediction models can help identifying individuals at higher fall risk. Electronic health records (EHR) offer an opportunity to develop automated prediction tools that may help to identify fall-prone individuals and lower clinical workload. However, existing models primarily utilise structured EHR data and neglect information in unstructured data. Using machine learning and natural language processing (NLP), we aimed to examine the predictive performance provided by unstructured clinical notes, and their incremental performance over structured data to predict falls.MethodsWe used primary care EHR data of people aged 65 or over. We developed three logistic regression models using the least absolute shrinkage and selection operator: one using structured clinical variables (Baseline), one with topics extracted from unstructured clinical notes (Topic-based) and one by adding clinical variables to the extracted topics (Combi). Model performance was assessed in terms of discrimination using the area under the receiver operating characteristic curve (AUC), and calibration by calibration plots. We used 10-fold cross-validation to validate the approach.ResultsData of 35,357 individuals were analysed, of which 4,734 experienced falls. Our NLP topic modelling technique discovered 151 topics from the unstructured clinical notes. AUCs and 95% confidence intervals of the Baseline, Topic-based and Combi models were 0.709 (0.700–0.719), 0.685 (0.676–0.694) and 0.718 (0.708–0.727), respectively. All the models showed good calibration.ConclusionsUnstructured clinical notes are an additional viable data source to develop and improve prediction models for falls compared to traditional prediction models, but the clinical relevance remains limited.

Funder

Dutch Research Council

Publisher

Oxford University Press (OUP)

Subject

Geriatrics and Gerontology,Aging,General Medicine

Reference42 articles.

1. EuroSafe: injuries in the European Union, summary on injury statistics 2012–2014;EuroSafe;EuroSafe,2016

2. Trends in nonfatal falls and fall-related injuries among adults aged ≥65 years—United States, 2012–2018;Moreland;MMWR Morb Mortal Wkly Rep,2020

3. Consequences of falling in older men and women and risk factors for health service use and functional decline;Stel;Age Ageing,2004

4. Societal consequences of falls in the older population: injuries, healthcare costs, and long-term reduced quality of life;Hartholt;J Trauma Inj Infect Crit Care,2011

5. New horizons in falls prevention and management for older adults: a global initiative;Montero-Odasso;Age Ageing,2021

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3