时间标签：数据日期 2026-10-31 · 生成时间 2026-11-01 05:30（Asia/Shanghai）｜ 检索窗口 PubMed edat 2026/10/28–2026/10/31 ｜ 情报发生日期 2026-10-31
研究问题
E129 Predicting Future Long-Term Care Dependency with Explainable Artificial Intelligence Using Administrative Healthcare Data MICHIYO KAWAI1,2 , MAMORU HIROE2 , HISATAKA ANEZAKI2 , and YOICHI KUREBAYASHI2, * 1 Kobe University Graduate School of Science, Technology and Innovation, Kobe, Japan; 2 Department of Artificial Intelligence and Digital Health Sciences, Kobe University Graduate School of Medicine, Kobe, Japan *Corresponding author Received July 9, 2026/Accepted August 17, 2026 Keywords: Explainable artificial intelligence, Long-term care dependency, Administrative healthcare data, Machine learning, Risk prediction Preventing future long-term care dependency is a major public health challenge in aging societies.；INTRODUCTION With the rapid progression of population aging worldwide, the growing number of older adults requiring long- term care has become a major societal challenge (
方法
The model demonstrated good discriminative performance in the randomly split internal test cohort (AUC:0.870, sensitivity: 0.790, specificity: 0.800).；E130 MATERIALS AND METHODS Study population Among approximately 400,000 residents aged 65 years or older who were eligible for Kobe City’s Community- based Integrated Care System (12), a total of 284,660 individuals were included in the analysis after excluding those who had already been certified as requiring Long-Term Care Level 2 or higher during the baseline period (April 2016 to December 2017), as well as those who died or moved out of the city during the study period (April 2016 to December 2019).；Specifically, the 284,660 eligible individuals were randomly divided into a training cohort (n = 213,495) and a test cohort (n = 71,165).
核心结论
Conversely, the model identified factors associated with reduced predicted risk, including hypercholesterolemia, allergic rhinitis, and chronic periodontitis.；The model identified both high-risk individuals and disease-specific factors associated with predicted long-term care risk by quantifying feature importance and impact.；Unlike conventional black-box models, the proposed model enables transparent interpretation by providing not only predicted risk values but also the factors associated with the predicted risk. ‖ 结论：EXPLAINABLE AI FOR LONG-TERM CARE PREDICTION E135 In conclusion, the explainable AI model using the B3 algorithm accurately predicted future long-term care dependency while simultaneously identifying important factors associated with predicted long-term care risk.
局限
Therefore, these associations should be interpreted cautiously, and further studies are required to clarify the mechanisms underlying these negative feature impacts.；The present study has several limitations.；First, because the study population was limited to beneficiaries of the National Health Insurance and the Late-Stage Elderly Medical Care System in Kobe City, caution is required when generalizing the findings.
对本项目的价值
★☆☆☆☆ 方法学进展决定我们能否在有限样本下检出真实效应，属基础设施级收益。
全文来源
出版商 OA PDF（正文 31,988 字符，已过 DOI 逐字 / 标题指纹归属校验）
原文题录
- 英文标题：Predicting Future Long-Term Care Dependency with Explainable Artificial Intelligence Using Administrative Healthcare Data.
- 期刊与标识：Kobe J Med Sci · PMID 42950853 · DOI 10.24546/0100505679
A 级证据分级：A 级（同行评议人体研究）
本文属人体研究证据，但需注意：观察性或小样本干预的结论不等于已证实临床获益；效应量与长期安全性仍需更大规模、更长随访的试验验证。
关于本文的文献来源：本站依据 IMA 知识库「五刊监控·深度文献卡片」撰写，依据已获取的期刊全文（方法、样本量与结论已回到原文核对）；未能获取全文的同批条目另行标注「摘要级」，全程未做任何数值推算补全。