2010-2015年耐甲氧西林金黄色葡萄球菌医院流行趋势时间序列分析  

Epidemiological analysis on methicillin-resistant Staphylococcus aureus incidence from 2010 to 2015 by time series model

作　　者：单欢[1] 金凯玲[1] 叶金明[1] 张佩维[1] 林凯[1] 陈伟国[1] 储文杰[1] 肖震[2] SHAN Huan JIN Kai-Ling YE Jin-Ming ZHANG Pei-Wei LIN Kai CHEN Wei-Guo CHU Wen-Jie XIAO Zhen(Zhejiang Hospital, Hangzhou, Zhejiang 310013, China)

机构地区：[1]浙江医院医院感染管理科,浙江杭州310013 [2]浙江医院医学检验科,浙江杭州310013

出　　处：《中华医院感染学杂志》2016年第23期5308-5311,共4页Chinese Journal of Nosocomiology

基　　金：浙江医院医药卫生科学研究基金项目(2015YJ008)

摘　　要：目的探讨应用时间序列求和自回归滑动平均模型(ARIMA)进行耐甲氧西林金黄色葡萄球菌(MRSA)流行趋势预测的可行性,为降低MRSA定植或感染提供理论依据。方法使用2010-2014年浙江医院MRSA检出率拟合ARIMA模型,以2015年1-12月MRSA实际检出率作为预测模型的考核样本,验证模型的预测效果。结果 MRSA检出率ARIMA模型为Xt=0.3807Xt-1+Xt-12-0.3807Xt-13-0.02725;模型预测的平均相对误差为20.19%,预测的动态趋势与实际值基本吻合。结论 ARIMA模型对MRSA检出率拟合较为满意,预测效果良好,可为临床早期采取防控措施提供依据。OBJECTIVE To discuss the feasibility of using autoregressive integrated moving average models（ARIMA）to predict the incidence of methicillin-resistant Staphylococcus aureus（MRSA）and to provide theoretical basis for reducing the incidence of MRSA carriers or patients,so as to provide theoretical foundation for decreasing the colonization and infection of MRSA.METHODS ARIMA model was established by the monthly incidence rates of MRSA carriers or patients in a hospital of Zhejiang province from 2010 to 2014.The actual monthly incidence rates of MRSA carriers or patients from Jan.to Dec.2015 was used as inspection sample to predict the model,and the forecast result was also assessed.RESULTS The MRSA detection rate for ARIMA model was Xt=0.3807Xt-1＋Xt-12-0.3807Xt-13-0.02725.The average of the relative error between actual and predicted values was20.19%,but the dynamic trend of model prediction and the actual value was basically same.CONCLUSION The ARIMA model satisfies with the detection rate of MRSA with a good prediction effect,which could provide a reference to the prevention and control of MRSA carriers or patients.

关 键 词：耐甲氧西林金黄色葡萄球菌 求和自回归滑动平均模型 预测 

分 类 号：R378.11[医药卫生—病原生物学]