老年患者导尿管相关尿路感染风险预测评分模型构建与验证

Establishment and validation of risk prediction scoring model for catheterassociated urinary tract infection in elderly patients

目的构建老年患者导尿管相关尿路感染(CAUTI)风险预测评分模型,为筛选高危人群,有效预防与控制老年患者泌尿系统感染提供依据。方法采用回顾性研究方法,收集2015年1月1日—2017年11月30日住院期间曾留置导尿管的老年患者病历资料(2015—2016年数据作为建模组,2017年数据作为验证组),构建logistic回归模型进行危险因素分析,根据OR值赋予各危险因素相应的分值,建立感染风险评分模型,利用受试者工作特征(ROC)曲线评价模型的预测效果。依据建立的感染风险评分模型对验证组病例进行评分,利用ROC曲线评价模型的预测效果。结果 logistic回归分析结果表明,年龄(≥80岁)、肾功能障碍、联用抗菌药物、特殊类抗菌药物使用天数(≥15 d)、术后住院时间(≥15 d)、导尿管置管天数(≥7 d)是老年患者发生CAUTI的独立危险因素,风险评分模型中相应的分值分别为3、3、3、5、7、9分。评分>15分为高风险人群,评分模型在建模组数据中的曲线下面积(AUC)为0.91(95%CI:0.90～0.93,P<0.05),灵敏度84.8%,特异度85.3%。依据验证组患者风险得分情况绘制ROC曲线,AUC为0.92(95%CI:0.88～0.95,P<0.05),灵敏度84.7%,特异度86.6%。结论建立的风险评分模型在建模组和验证组均具有较好的判别效度,可用于识别老年患者CAUTI易感高危人群,以期早期预防与控制。

Objective To establish a risk prediction scoring model for catheter-associated urinary tract infection(CAUTI)in elderly patients,provide evidence for screening high-risk population,and effectively prevent and control urinary tract infection in elderly patients.Methods A retrospective study was conducted to collect medical data of elderly patients with indwelling urinary catheters during hospitalization between January 1,2015 and November 30,2017(data from 2015 to 2016 was as modeling group,in 2017 as validation group),logistic regression model was constructed to analyze risk factors,according to OR value,the corresponding point of each risk factor was scored,an infection risk scoring model was established,receiver operating characteristic(ROC)curve was used to evaluate prediction effect of model.According to the established infection risk scoring model,patients in validation group were scored,ROC curve was used to evaluate the predictive effect of the model.Results Logistic regression analysis showed that age(≥80 years),renal dysfunction,combined use of antimicrobial agents,days of special antimicrobial use(≥15 days),length of postoperative hospital stay(≥15 days),duration of indewlling catheter(≥7 days)were independent risk factors for CAUTI in elderly patients,the corresponding points in the risk scoring model were 3,3,3,5,7,and 9 respectively.Score>15 points was high-risk group,the area under the curve(AUC)of scoring model was 0.91(95%CI:0.90-0.93,P<0.05),sensitivity and specificity were 84.8%and 85.3%respectively.ROC curve was drawn according to the risk score of validation group,AUC was 0.92(95%CI:0.88-0.95,P<0.05),sensitivity and specificity were 84.7%and 86.6%respectively.Conclusion The established risk scoring model has good discriminatory validity for both modeling group and validation group,and can be used in the identification of susceptible high-risk population for CAUTI in elderly patients,so as to achieve early prevention and control.