摘要
目的分析神经重症患者鼻饲后腹泻的影响因素,构建预测模型列线图并验证预测效能。方法采用回顾性研究方法,选取2020年5月—2023年11月天津市第一中心医院神经重症监护室收治的797例鼻饲患者为研究对象,将患者以7∶3的比例分为建模集(n=558)和验证集(n=239)。根据单因素分析筛选影响因素,通过多重共线性诊断检测变量间的相关性,再利用二项Logistic回归分析确定危险因素,并创建列线图模型。采用受试者工作特征曲线下面积(AUC)评估模型的预测能力;使用Hosmer-Lemeshow检验评估模型的拟合程度,并绘制校准曲线;通过临床决策曲线评价模型在临床中的实际应用价值。结果二项Logistic回归分析结果显示,格拉斯哥昏迷评分(OR=0.891)、急性生理与慢性健康状况评价系统Ⅱ评分(OR=1.063)、镇静镇痛剂(OR=0.326)、白蛋白水平(OR=0.856)、抗生素使用天数(OR=3.338)、抑酸剂(OR=3.260)是神经重症患者鼻饲后腹泻的影响因素(P<0.05)。在验证集中,神经重症患者鼻饲后腹泻风险列线图模型的AUC为0.82(95%CI:0.77~0.87),灵敏度为0.72,特异度为0.72。临床决策曲线显示,列线图模型具有较高的应用价值。结论本研究构建的列线图模型具有良好的预测效能,能够帮助神经重症专科医护人员早期识别发生腹泻的高风险患者,从而为临床制订个体化预防策略提供参考。
ObjectiveTo explore the factors influencing diarrhea after nasal feeding in patients with severe neurological illness,construct a nomogram model and verify its predictive performance.MethodsA retrospective study was conducted to select 797 nasal feeding patients admitted to the Neurological Intensive Care Unit of the Tianjin First Central Hospital from May 2020 to November 2023 as study subjects.Patients were divided into a modeling set(n=558)and a validation set(n=239)in a 7∶3 ratio.Influencing factors were screened based on univariate analysis,correlations between variables were detected by multicollinearity diagnostics,and then risk factors were identified using binomial Logistic regression,and a nomogram model was created.The predictive capacity of the model was assessed using the area under the receiver operating characteristic curve(AUC),the degree of fit of the model was evaluated using the Hosmer-Lemeshow test and calibration curves were plotted,and the practical application value of the model in clinical practice was appraised by means of clinical decision curve analysis.ResultsBinomial Logistic regression analysis showed that Glasgow Coma Scale score(OR=0.891),Acute Physiology and Chronic Health EvaluationⅡscore(OR=1.063),sedative-analgesic(OR=0.326),albumin level(OR=0.856),days of antibiotic use(OR=3.338),and acid suppressants(OR=3.260)were the factors influencing diarrhea after nasal feeding in patients with severe neurological illness(P<0.05).In the validation set,the AUC for the nomogram model of the risk of diarrhea after nasal feeding in patients with severe neurological illness was 0.82[95%CI(0.77,0.87)],with a sensitivity of 0.72 and a specificity of 0.72.Clinical decision curve showed nomogram model had high application value.ConclusionsThe constructed nomogram model has good predictive performance and can help healthcare professionals in neurocritical care specialties to identify patients at high risk of diarrhea at an early stage,thus informing the development of individualized prevention strategies in clinical practice.
作者
王皓月
刘彦慧
常连霞
杨艳
宿威
Wang Haoyue;Liu Yanhui;Chang Lianxia;Yang Yan;Su Wei(School of Nursing,Tianjin University of Traditional Chinese Medicine,Tianjin 301617,China;Neurological Intensive Care Unit,Tianjin First Central Hospital,Tianjin 300192,China;Nursing Department,Tianjin First Central Hospital,Tianjin 300192,China;Plastic and Burn Surgery,Tianjin First Central Hospital,Tianjin 300192,China)
出处
《中华现代护理杂志》
2025年第9期1215-1222,共8页
Chinese Journal of Modern Nursing
基金
国家自然科学基金面上项目(82072212)。
关键词
肠内营养
神经重症护理
腹泻
预测模型
Enteral nutrition
Neurocritical care
Diarrhea
Prediction model
