河南省某三甲医院多药耐药菌感染率时间序列分析  

Time series of incidence of multidrug-resistant organism infection in a tertiary hospital in Henan province

作　　者：张佑健 李鹏[1] 刘云红[1] 李轶[2] 张付华[1] 王婷婷[1] 王菲菲 李妍[1] 孙明洁[1] ZHANG You-jian;LI Peng;LIU Yun-hong;LI Yi;ZHANG Fu-hua;WANG Ting-ting;WANG Fei-fei;LI Yan;SUN Ming-jie(Henan Provincial People’s Hospital,Zhengzhou,Henan 450003,China)

机构地区：[1]河南省人民医院医院感染管理科,河南郑州450003 [2]河南省人民医院检验科微生物室,河南郑州450003

出　　处：《中华医院感染学杂志》2020年第1期44-48,共5页Chinese Journal of Nosocomiology

基　　金：河南省科技攻关计划基金资助项目(162102310017)。

摘　　要：目的研究2013-2017年河南省人民医院多药耐药菌(Multi-drug resistant organism,MDRO)感染发生率的变化趋势,并预测2018年医院MDRO感染发生率。方法采用时间序列分析方法,对河南省人民医院MDRO感染发生率进行建模,选择最优模型,预测2018年MDRO感染发生率,并与实际监测结果相比较,验证其预测效果。结果采用专家建模器自动筛选为"简单季节性"模型,2018年MDRO感染发生率预测值与实际值趋势基本一致,预测值与实际值平均相对误差为14.16%。结论 MDRO感染率具有一定的季节性和周期性,其变化规律可以使用时间序列分析"简单季节性"模型预测,为早期采取感染防控措施提供科学依据。OBJECTIVE To investigate the incidence of multidrug-resistant organism(MDRO) infection in Henan Provincial People′s Hospital from 2013 to 2017, observe the changing trends and predict the incidence of MDRO infection in 2018. METHODS By means of time-series analysis, the models were established for the incidence of MDRO infection in Henan Provincial People′s Hospital, the optimal model was chosen to predict the incidence of MDRO infection in 2018, the prediction result was compared with the result of actual monitoring so as to verify the predictive effect. RESULTS The simple seasonal model was automatically screened by statistical package modeler, the predicted value of incidence of MDRO infection of 2018 is basically consistent with the trend of actual value, and the mean deviation between the predicted value and the actual value was 14.16%. CONCLUSION The incidence of MDRO infection shows certain seasonal and periodic characteristics, the change rules of which can be predicted by the simple seasonal model for time-series analysis so as to provide scientific basis for taking prevention measures in early stage.

关 键 词：时间序列 多药耐药菌 感染 模型 

分 类 号：R378[医药卫生—病原生物学]