老年吞咽障碍影响因素探究及其预测研究  

Influencing factors for dysphagia in the elderly and establishment of a predictive model

作　　者：彭朋 陈忻睿 周艺林 田小芹 唐于钦 邓丹[1] Peng Peng;Chen Xinrui;Zhou Yilin;Tian Xiaoqin;Tang Yuqin;Deng Dan(Department of Health Statistics,College of Public Health,Chongqing Medical University;Chongqing Jiangbei District Center for Disease Control and Prevention/Chongqing Jiangbei District Public Health Inspection Office)

机构地区：[1]重庆医科大学公共卫生学院卫生统计学教研室,重庆400016 [2]重庆市江北区疾病预防控制中心/重庆市江北区卫生监督所,重庆400025

出　　处：《重庆医科大学学报》2025年第4期501-510,共10页Journal of Chongqing Medical University

基　　金：重庆市自然科学基金面上资助项目(编号:CSTB2022NSCQ-MSX0804);重庆英才计划“包干制”资助项目(编号:cstc2022ycjh-bgzxm0015);重庆市教育委员会人文社会科学重点研究资助项目(编号:21SKGH027)。

摘　　要：目的:探究老年人吞咽障碍影响因素,构建吞咽障碍预测模型,为临床实践提供理论依据。方法:本研究采用病例对照研究设计,选取重庆医科大学附属第一医院老年病科2016年3月至2023年6月就诊的吞咽障碍患者为病例组,同期同科室就诊的非吞咽障碍病人为对照组,利用相关性分析、套索回归(least absolute shrinkage and selection operator,LASSO)及多因素logistic回归探究吞咽障碍影响因素,十折交叉验证训练极度梯度提升算法(extreme gradient boosting,XGBoost)模型用于预测吞咽障碍并运用沙普利加性解释(SHapley additive exPlanations,SHAP)方法对模型进行可视化处理。结果:病例组1 009例,对照组2 125例。相关性及LASSO回归共筛选12个因素进行多因素logistic回归,多因素结果显示患有肌肉减少症、年龄增加、照料者为子女或看护、老年人身体衰弱、口腔健康状况差、自我生活能力差、抑郁、认知障碍是吞咽障碍的危险因素(OR>1,P<0.05),女性、参加社区活动为吞咽障碍的保护因素(OR<1,P<0.05)。XGBoost模型预测效能较好,模型精确率为0.795、查准率为0.711、灵敏度为0.613、特异度为0.881,F1值为0.661,受试者工作特征(receiver operating characteristic,ROC)曲线下面积AUC=0.855。SHAP图显示,前5位重要特征分别为照料者、口腔评分、衰弱情况、日常生活能力及认知功能。结论:老年人吞咽障碍影响因素众多,临床上应更加关注口腔健康差、身体衰弱、日常生活需要依赖他人及认知功能障碍的老年人,XGBoost模型在老年人吞咽障碍预测上有较好的表现,可为临床实践提供参考。Objective:To investigate the influencing factors for dysphagia in the elderly,to construct a predictive model for dysphagia,and to provide a theoretical basis for clinical practice.Methods:In this case-control study,the patients with dysphagia who attended Department of Geriatrics in the first affiliated hospital of Chongqing Medical University from March 2016 to June 2023 were enrolled as case group,and the patients without dysphagia who attended the same department during the same period of time were enrolled as control group.The correlation analysis,least absolute shrinkage and selection operator(LASSO) regression,and multivariate logistic regression analysis were used to investigate the influencing factors for dysphagia;the 10-fold cross-validation Extreme Gradient Boosting(XGBoost) model was used to predict dysphagia,and the SHapley additive exPlanations(SHAP) method was used for model visualiza- tion.Results:There were 1009 cases in the case group and 2125 cases in the control group.The correlation analysis and LASSO regression analysis identified 12 factors for the multivariate logistic regression analysis,and the results showed that sarcopenia,increasing age,children or caretakers as caregivers,frail health,poor oral health,poor self-care ability,depression,and cognitive impairment were risk factors for dysphagia(odds ratio[OR]>1,P<0.05),and female sex and participation in community activities were protective factors against dysphagia(OR<1,P<0.05).The XGBoost model had a good predictive efficacy,with an accuracy rate of 0.795,a precision rate of 0.711,a sensitivity of 0.613,a specificity of 0.881,an F1 value of 0.661,and an area under the ROC curve of 0.855.The SHAP plot showed that the top five important characteristics were caregiver,oral score,frail health condition,activities of daily living,and cognitive function.Conclusion:There are various influencing factors for dysphagia in the elderly,and the elderly patients with poor oral health,frailty,dependence on others for daily life,and cognitive impairm

关 键 词：吞咽障碍 影响因素 预测模型 XGBoost模型 

分 类 号：R473.78[医药卫生—护理学]