老年慢性心力衰竭住院患者中重度心力衰竭发生风险预测模型的构建

Construction of a risk prediction model for moderate to severe heart failure in elderly inpatients with chronic heart failure

中文关键词: 电子病历  慢性心力衰竭  中重度  风险预测模型  护理

英文关键词: electronic medical record  chronic heart failure  moderate to severe  risk prediction model  nursing

中文摘要:

      目的 基于电子病历的方式,构建慢性心力衰竭(CHF)患者中重度心力衰竭发生风险预测模型,并验证该模型的预测效果.方法 2019 年1 月—2020 年12 月,选择上海市某三级甲等医院心血管临床医学中心诊断为CHF的299 例患者为研究对象,采用Logistic回归建立CHF患者中重度心力衰竭发生风险预测模型,采用Hosmer-Lemeshow和受试者操作特征曲线分别检验模型的拟合优度及预测效果,并纳入 100 例患者对模型进行验证.结果 心超(运动幅度降低)(OR =5.109)、双下肢水肿(OR =3.947)、心房颤动(OR =2.772)、血肌酐升高(OR =1.015)是CHF患者发生中重度心力衰竭的危险因素;血清白蛋白升高(OR =0.939)是保护因素;Hosmer-Lemeshow检验P =0.127,受试者操作特征曲线下面积为 0.858,约登指数为 0.528,最佳临界值为 0.805,灵敏度为 0.731,特异度为 0.797,实际应用的正确率为 77.00%.结论 本课题建立的CHF患者中重度心力衰竭发生风险的预测模型效果良好,有利于以后早期识别中重度心力衰竭发生风险电子系统报警提示程序的更新提供依据.

英文摘要:

      Objective To construct a risk prediction model for moderate to severe heart failure in patients with chronic heart failure(CHF)based on electronic medical records,and verify the predictive effect of the model.Methods A total of 299 patients diagnosed with CHF in cardiovascular clinical medical center of a grade Ⅲ-A hospital in Shanghai from January 2019 to December 2020 were selected as the research subjects.Logistic regression was used to establish the risk prediction mod-el for moderate to severe heart failure in CHF patients.Hosmer-Lemeshow and receiver operating characteristic curve(ROC)were used to test the goodness of fit and prediction effect of the model,respectively.A total of 100 patients were enrolled to verify the model.Results Cardiac ultrasound(reduced motion amplitude)(OR = 5.109),edema of both lower limbs(OR =3.947),atrial fibrillation(OR =2.772),and elevated serum creatinine(OR =1.015)were risk factorsfor moderate to severe heart failure in patients with CHF,while elevated serum albumin(OR =0.939)was a protective factor.Hosmer-Leme-show test showed that P =0.127,the area under the ROC curve was0.858,the Youden index was0.528,the optimal critical value was 0.805,the sensitivity was 0.731,the specificity was 0.797,and the accuracy of practical application was 77.00%.Conclusion The predictive model established in this study for the risk of moderate to severe heart failure in CHF patients ex-hibits good predictive performance,which is beneficial for the early identification of the risk of moderate to severe heart failure in the future,and provides a basis for updating the electronic system alarm program.