摘要
目的研究老年住院患者抑郁状态的危险因素,并构建风险预测模型。方法选取2017年9月至2020年7月云南省第一人民医院老年医学科老年住院患者(≥60岁)525例为研究对象,采用云南省第一人民医院老年医学科自主研发的“老年综合评估系统软件”对患者进行老年综合评估,并利用其中的SDS量表对住院患者进行抑郁状态的评估。采用二元Logistic回归分析探讨老年抑郁的影响因素,建立风险预测模型,采用ROC曲线下面积及Hosmer-Lemeshow检验评估模型预测效果。结果在525例新入院患者中,非抑郁组418例(79.6%),抑郁组107例(20.4%);2组间的年龄、性别、慢病数量、MNA-SF、SAS、MMSE、BADL、ALS、Morse、RomaⅢ量表,差异均有统计学意义(P<0.05);且慢病数量(OR=1.473,95%CI:1.050~2.067,P=0.025)、失能(OR=1.450,95%CI:1.073~1.958,P=0.015)、焦虑(OR=2.442,95%CI:1.829~3.259,P<0.001)、认知功能障碍(OR=1.488,95%,CI:1.072~2.066,P=0.018)、失眠(OR=1.620,95%CI:1.147~2.288,P=0.006)、社会支持障碍(OR=2.544,95%,CI:1.165~5.555,P=0.019)是老年住院患者抑郁状态的独立危险因素;另外,通过将模型预测概率作为新变量绘制ROC曲线,结果显示AUC为0.859,临界值为0.595(CI:0.822~0.897,P<0.001),预测模型的灵敏度为0.818,特异度为0.782。结论老年抑郁状态与多种危险因素密切相关,基于该危险因素所建立的风险预测模型亦具有较高的准确性。
Objective To study the risk factors of depression in elderly inpatients and build a risk prediction model.Methods From September 2017 to July 2020,525 elderly patients(≥60 years old)in the Department of Geriatrics of the First People’s Hospital of Yunnan Province were selected as the research objects.The comprehensive geriatric assessment system software independently developed by the geriatric medical department of the First People’s Hospital of Yunnan province was used for the comprehensive geriatric assessment,and the SDS scale in it was used to evaluate the depressive state of hospitalized patients.Binary Logistic regression analysis was used to explore the influencing factors of depression in the elderly,and a risk prediction model was established.The area under the ROC curve and the Hosmer-Lemeshow test were used to evaluate the prediction effect of the model.Results Among 525 newly admitted patients,there were 418 cases(79.6%)in the non-depressed group and 107 cases(20.4%)in the depressed group.There were statistically significant differences in age,gender,number of chronic diseases,MNA-SF,SAS,MMSE,BADL,ALS,MORSE and RomaⅢscales between the two groups(P<0.05).The number of chronic diseases(OR=1.473,95%CI:1.050~2.067,P=0.025),daily dysfunction(OR=1.450,95%CI:1.073~1.958,P=0.015),anxiety(OR=2.442,95%CI:1.829~3.259,P<0.001),cognitive dysfunction(OR=1.488,95%CI:1.072~2.066,P=0.018),insomnia(OR=1.620,95%CI:1.147~2.288,P=0.006),social support disorder(OR=2.544,95%,CI:1.165~5.555,P=0.019)were independent risk factors for depression in elderly inpatients.In addition,the ROC curve was drawn by using the prediction probability of the model as a new variable,the results showed that the AUC was 0.859,with a critical value of 0.595(CI:0.822~0.897,P<0.001),and the sensitivity and specificity of the prediction model were 0.818 and 0.782.Conclusion The state of senile depression is closely related to a variety of risk factors,and the risk prediction model based on the risk factors also has high accuracy.
作者
戴靖榕
李婕
何旭
李杨
李燕
DAI Jing-rong;LI Jie;HE Xu;LI Yang;LI Yan(Medical School,Kunming University of Science and Technology,Kunming Yunan 650500;Dept.of Geriatrics,The First People’s Hospital of Yunnan Province,Kunming Yunan 650032,China)
出处
《昆明医科大学学报》
CAS
2021年第11期20-26,共7页
Journal of Kunming Medical University
基金
国家自然科学基金资助项目(81760109)
国家重点研发计划(2018YFC2002103)
云南省临床医学开发项目(2019LCZXKF-NM08)
云南省卫生科技计划项目(2017NS221,2017NS222)。
关键词
老年综合征
抑郁
预测模型
住院患者
Geriatric symptoms
Depression
Prediction model
Inpatient
