基于SARIMA模型的小儿神经内科医院感染发病率预测  

Incidence prediction of healthcare-associated infection in pediatric neurology patients based on SARIMA model

作　　者：程颖[1] 肖书娜[1] 许慧[1] 黄娟[1] 冷福建[2] 吴艳[2] CHENG Ying;XIAO Shuna;XU Hui;HUANG Juan;LENG Fujian;WU Yan(Dept.of Pediatrics,Hubei Provincial Maternal and Child Heath Hospital,Wuhan 430070,Hubei,China;Dept.of Healthcare-Associated Infection Management,Hubei Provincial Maternal and Child Heath Hospital,Wuhan 430070,Hubei,China)

机构地区：[1]湖北省妇幼保健院儿科,湖北武汉430070 [2]湖北省妇幼保健院院感办,湖北武汉430070

出　　处：《武汉大学学报（医学版）》2021年第2期282-286,共5页Medical Journal of Wuhan University

基　　金：湖北省自然科学基金资助项目(编号:2018CFB735)。

摘　　要：目的:选择季节性差分自回归滑动平均(SARIMA)模型预测小儿神经内科病房院内感染发病率。方法:统计湖北省某三级甲等儿童专科医院小儿神经内科院内感染发病率数据,运用时间序列中的SARIMA(p,d,q)×(P,D,Q)s方法构建模型,对模型中的参数进行估计、检验,确定最佳模型,并采用该模型对小儿神经内科病房院内感染发病率进行预测及效果评价。结果:通过分析比较,确立最佳预测模型SARIMA(0,1,1)×(1,1,0)_(12)。以2019年1—6月科室院内感染发病率作为模型预测检验样本。实际情况与预测值的动态趋势基本一致。结论:SARIMA模型能较准确预测某院小儿神经内科院内感染发病率变化趋势,SARIMA模型的建立有助于医院感染管理部门及儿科有针对性地制定院感防控措施,进行早期干预,降低院内感染发生率。Objective: To explore the use of Seasonal Autoregressive Integrated Moving Average(SARIMA) method to establish a predictive model of pediatric neurology ward infection incidence, and provide theoretical basis for hospital infection management departments to formulate prevention and control strategies.Methods: The data of incidence of healthcare-associated infection(HAI) in pediatric neurology ward of a top first-class children′s hospital in Hubei Province were collected. The prediction model was established with SARIMA method, parameter estimation and model diagnosis were performed for the established model, and the optimal prediction model was selected. At the same time,the model was used to predict the infection incidence in pediatric neurology ward, and the prediction effect was evaluated.Results: Through comparison, the optimal prediction model was SARIMA(0,1,1)×(1,1,0)_(12). Data of January-May in 2019 was as validation sample for model prediction, the results showed that the dynamic trend of the predicted value was basically consistent with the actual condition, and the actual values were within the 95% confidence interval of the predicted value.Conclusion: The model can better predict the trend of the incidence of nosocomial infection in department of pediatric neurology, and it is helpful for the establishment and implementation of prevention and control measures related to nosocomial infection management.

关 键 词：医院感染 SARIMA模型 小儿神经内科 预测 

分 类 号：R174.6[医药卫生—妇幼卫生保健]