【摘要】 目的 基于不同算法构建合并多重慢性病的冠心病患者衰弱综合征风险预测模型，并进行验证。方法 选取2022年1月至2024年5月于郑州大学第一附属医院就诊的多重慢性病的冠心病患者430例，根据是否发生衰弱综合征分为衰弱组、无衰弱组，收集两组临床资料、实验室指标，通过单因素、多因素logistic回归分析确定进入各项模型的预测因子，并基于诺莫图、随机森林构建合并多重慢性病的冠心病患者衰弱综合征的预测模型，采用受试者工作特征(ROC)曲线验证其预测效能。结果 合并多重慢性病的冠心病患者衰弱综合征发生率为44.42%(191/430);年龄≥60岁、高纽约心脏病协会(NYHA)分级、营养不良、合并抑郁、有近1 a跌倒史、高成纤维细胞生长因子23(FGF23)水平、高全身炎症反应指数(SIRI)是合并多重慢性病的冠心病患者发生衰弱综合征的危险因素；随机森林模型的曲线下面积(AUC)为0.929(95%CI:0.901～0.951),大于诺莫图0.882(95%CI:0.848～0.911)。结论 相较诺莫图，基于随机森林模型建立合并多重慢性病的冠心病患者衰弱综合征的风险预测模型具有更好的预测效能，有利于衰弱综合征发生风险的评估，临床可据此制定针对性护理措施，降低衰弱综合征发生风险。更多还原

【Abstract】 Objective To construct and validate a risk prediction model for frailty syndrome in coronary heart disease patients with multiple chronic diseases based on different algorithms. Methods A total of 430 coronary heart disease patients with multiple chronic diseases who were treated at the First Affiliated Hospital of Zhengzhou University from January 2022 to May 2024 were selected. They were divided into the frail group and the non-frail group based on whether they had experienced the frail syndrome. Clinical data and laboratory indicators were collected from both groups. The predictive factors for entering each model were determined through univariate and multivariate logistic regression analysis. Based on nomogram and random forests, a predictive model for the frail syndrome in coronary heart disease patients with multiple chronic diseases was constructed, and its predictive performance was validated using the receiver operating characteristic(ROC) curve. Results The incidence of frailty syndrome in coronary heart disease patients with multiple chronic diseases was 44.42%(191/430). Patients with coronary heart disease who were aged ≥60 years, had a high New York Heart Association(NYHA) classification, were malnourished, had depression, had a history of falls in the past year, have high levels of fibroblast growth factor 23(FGF23), and had a high systemic inflammatory response index(SIRI) were at risk for developing frailty syndrome in patients with multiple chronic diseases. The area under the curve(AUC) of the random forest model was 0.929(95% CI: 0.901-0.951), which was greater than the nomogram’s 0.882(95% CI: 0.848-0.911). Conclusion Compared with nomogram, the risk prediction model of frailty syndrome in coronary heart disease patients with multiple chronic diseases based on random forest model has better predictive performance, which is beneficial for the assessment of the risk of frailty syndrome. Clinical nurses can develop targeted nursing measures based on this model to reduce the risk of frailty syndrome.更多还原