老年烧伤患者医院感染发生情况及相关影响因素Logistic回归模型对临床防治的指导意义  

Incidence and related factors of nosocomial infection in elderly burn patients guiding significance of logistic regression model for clinical prevention and treatment

作　　者：何瑛 蒋芩[1] 张晓 张学群 HE Ying;JIANG Qin;ZHANG Xiao;ZHANG Xuequn(Department of Burn and Plastic Surgery,Suzhou Municipal Hospital,Suzhou Jiangsu 215000,China)

机构地区：[1]苏州市立医院烧伤整形科,江苏苏州215000

出　　处：《中国急救复苏与灾害医学杂志》2022年第7期934-938,共5页China Journal of Emergency Resuscitation and Disaster Medicine

基　　金：江苏省卫健委卫生财务研究项目(编号:CW201856)。

摘　　要：目的 探究老年烧伤后医院感染发生情况及相关影响因素Logistic回归模型对临床防治的指导意义。方法回顾性选取苏州市市立医院2017年3月—2020年10月老年烧伤患者208例作为研究对象,统计感染发生情况,根据是否发生医院感染分为感染组与无感染组,收集两组临床资料,分析老年烧伤患者发生感染的影响因素,构建Logistic多因素回归模型,采用似然比卡方、Wald卡方、拟合优度检验评价Logistic回归模型,受试者工作特征(ROC)曲线评价模型的预测价值。结果 208例老年烧伤患者,医院感染发生率为17.79%(37/208),其中血液感染14例,占37.84%;皮肤软组织感染13例,占35.14%;下呼吸道感染10例,占27.03%。构建Logistic多因素回归模型显示,年龄>70岁、Ⅳ度烧伤、烧伤面积>50%、重度烧伤、侵入性操作、使用有创呼吸机、抗生素应用时间>48 h是老年烧伤患者医院感染发生的独立危险因素(P<0.05);似然比卡方检验,模型建立具有统计学意义,Wald卡方检验,模型构建有效,拟合优度检验,模型拟合效果较好;ROC曲线分析,Logistic多因素回归模型预测老年烧伤患者发生感染的AUC为0.964(95%CI为0.937~0.991),敏感度为89.19%,特异度为91.81%。结论 老年烧伤患者医院感染发生率较高,以血液感染、皮肤软组织感染、下呼吸道感染较为常见,构建Logistic多因素回归模型能有效预测感染发生风险,为临床制定防治措施提供可靠依据。Objective To explore the incidence of nosocomial infection after burn in the elderly and its related influencing factors,and the guiding significance of logistic regression model for clinical prevention and treatment.Methods 208 elderly burn patients in our hospital from March 2017 to October 2020 were retrospectively selected as the research object.The incidence of infection was counted.They were divided into infection group and non infection group according to whether nosocomial infection occurred.The clinical data of the two groups were collected,the influencing factors of infection in elderly burn patients were analyzed,and a logistic multivariate regression model was constructed.Likelihood ratio chi square,Wald chi square The goodness of fit test evaluates the logistic regression model,and the receiver operating characteristic(ROC) curve evaluates the predictive value of the model.Results The incidence of nosocomial infection in 208 elderly burn patients was 17.79%(37/208),including 14 cases of blood infection,accounting for 37.84%,13 cases of skin and soft tissue infection,accounting for 35.14%,and 10 cases of lower respiratory tract infection,accounting for 27.03%.Logistic regression model showed that age >70 years old,fourth degree burn,burn area >50%,severe burn,invasive operation,use of invasive ventilator and application time of antibiotics >48 hours were independent risk factors for nosocomial infection in elderly burn patients(P<0.05);Likelihood ratio chi square test,model establishment has statistical significance,Wald chi square test,model construction is effective,goodness of fit test,model fitting effect is good;ROC curve analysis and logistic multivariate regression model predicted that the AUC of infection in elderly burn patients was 0.964(95%CI 0.937 ~ 0.991),the sensitivity was 89.19%,and the specificity was 91.81%.Conclusion The incidence of nosocomial infection in elderly burn patients is high,and blood infection,skin and soft tissue infection and lower respiratory tract infection are comm

关 键 词：老年烧伤 感染 影响因素 LOGISTIC回归模型 临床防治 

分 类 号：R644[医药卫生—外科学]