论文部分内容阅读
目的构建ICU患者医院感染logistic回归预测模型,并对模型进行评价。方法以入住ICU>48 h的患者为研究对象,构建医院感染logistic回归模型,对模型进行拟和优度检验、ROC曲线下面积分析。结果入住ICU天数、气管插管、前列腺肥大、动静脉插管、基础疾病(肿瘤)等变量进入logistic回归方程,模型ROC曲线下面积为0.856。结论 logistic回归模型对ICU患者医院感染预测拟合度较好。
Objective To construct a logistic regression model of nosocomial infection in ICU patients and evaluate the model. Methods Patients admitted to ICU> 48 h were selected as the research object to establish a logistic regression model of nosocomial infection. The model was tested for the goodness of fit and the area under the ROC curve was analyzed. Results Into ICU days, tracheal intubation, prostatic hypertrophy, arteriovenous intubation, and other underlying diseases (tumors) entered the logistic regression equation. The area under the ROC curve was 0.856. Conclusion The logistic regression model is good for predicting nosocomial infection in ICU patients.