摘要
目的通过运用2010年-2016年的既往监测数据分析和比较指数平滑法与求和自回归移动平均(ARI-MA)乘积季节模型对产超广谱β-内酰胺酶(ESBLs)肺炎克雷伯菌检出人数的预测效能,分析与预测产ESBLs肺炎克雷伯菌的流行趋势。方法使用浙江某医院2010年1月-2015年12月的产ESBLs肺炎克雷伯菌的月度监测数据,分别建立ARIMA模型与指数平滑法,以平均绝对百分误差(MAPE)及贝叶斯信息准则(BIC)作为评价指标评价模型的可行性。以2016年1-12月产ESBLs肺炎克雷伯菌的检出人数作为预测模型的样本数据以验证模型的预测效果。结果 ARIMA乘积季节模型筛选出的最优模型为ARIMA(0,1,1)(2,1,0)12,模型的MAPE为44.92,BIC为3.63,平均预测误差为14.34%。指数平滑法所筛选出的最优模型为简单季节模型,模型的MAPE为37.03,BIC为3.00,平均预测误差为17.65%。结论同简单季节模型相比,ARIMA乘积季节模型对该院既往监测数据拟合效果更为理想,预测精度更高,可用于预测产ESBLs肺炎克雷伯菌感染情况的预测和预警。
OBJECTIVE To observe and compare the efficacy of exponential smoothing method and multiple seasonal autoregressive integrated moving average(ARIMA)model in prediction of the number of people who are tested with extended-spectrumβ-lactamases(ESBLs)-producing Klebsiella pneumoniae by means of the previous monitoring data in 2010-2016 and analyze the prevalence of the ESBLs-producing K.pneumoniae.METHODS The ARIMA model and exponential smoothing method were respectively established based on the monitoring data of ESBLs-producing K.pneumoniae strains that was collected from a hospital in Zhejiang from Jan 2010 to Dec 2015,the feasibility of the model was evaluated by using the Mean Absolute Percent Error(MAPE)and Bayesian Information Criterion(BIC),and the predictive effect of the model was verified by taking the number of the people who were tested with ESBLs-producing K.pneumoniae strains as the sample data of the prediction model.RESULTS The optimal model that was screened out by the multiple seasonal ARIMA model was ARIMA(0,1,1)(2,1,0)12,with the MAPE 44.92,BIC 3.63,the mean prediction error 14.34%.The optimal model that was screened out by the exponential smoothing method was the simple seasonal model,with the MAPE 37.03,BIC 3.00,the mean prediction error 17.65%.CONCLUSION The multiple seasonal ARIMA model has higher imitative effect of the previous monitoring data and higher prediction accuracy than the simple seasonal mode does,and it can be used for the prediction of prevalence of ESBLs-producing K.pneumoniae infection.
作者
储文杰
叶金明
张佩维
金凯玲
林凯
单欢
陈伟国
CHU Wen-jie;YE Jin-ming;ZHANG Pei-wei;JIN Kai-ling;LIN Kai;SHAN Huan;CHEN Wei-guo(Zhejiang Hospital,Hangzhou,Zhejiang310003,China)
出处
《中华医院感染学杂志》
CAS
CSCD
北大核心
2019年第5期788-792,共5页
Chinese Journal of Nosocomiology
基金
浙江医院医药卫生科学研究基金项目(2015YJ008)
