应用ARIMA乘积季节模型对耐碳青霉烯类肺炎克雷伯菌流行趋势预测效果研究

doi:10.3969/j.issn.1006-5253.2023.03.003

摘要
图/表
参考文献
相关文章 (2)

全文: PDF (1803 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS) 背景资料

文章导读

摘要目的研究自回归求和移动平均（ARIMA）乘积季节模型在耐碳青霉烯类肺炎克雷伯菌（CRKP）流行趋势预测中的应用价值，为掌握医院CRKP流行趋势及制定防控对策提供参考。方法基于2016—2021年嘉兴市某三级甲等综合性医院CRKP月检出株数，应用SPSS 26.0软件构建ARIMA乘积季节模型，以2022年实际CRKP月检出株数作为评估模型的样本与预测值进行比较，评价模型的预测效果。结果 2016—2021年该院CRKP检出总数呈下降趋势，发病集中在每年的7—9月，9月为发病高峰，具有周期性和季节性。拟合的ARIMA（0,1,2）（0,1,0）12模型贝叶斯信息准则（BIC）为0.43，平均绝对百分误差（MAPE）为16.54，模型残差序列的Ljung-Box检验差异无统计学意义（Q=11.06，P=0.81），残差序列为白噪声，说明该模型拟合良好，确定为最优模型。2022年1—12月CRKP月检出株数用以验证模型的预测效果，结果显示实际发病趋势与预测曲线图较为吻合，实际值均位于预测值的95%置信区间内，预测误差-33.33%～25.00%，平均相对误差15.56%，说明模型的预测效果较好。结论 ARIMA乘积季节模型能较好地拟合该院CRKP检出株数的时间变化，可用于CRKP流行趋势的短期预测和动态分析，为医院CRKP感染的早期预警和防控提供理论依据。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	张绮萍
	季聪华
	陆锦琪
	王霄腾

关键词 ：自回归求和移动平均模型, 乘积季节模型, 耐碳青霉烯类肺炎克雷伯菌, 预测

Abstract：Objective To explore the application value of autoregressive integrated moving average (ARIMA) multiplicative seasonal model in predicting the epidemic trend of carbapenem resistant Klebsiella pneumoniae (CRKP), and to provide a theoretical basis for grasping the epidemic trend and formulating prevention and control measures of CRKP infection in hospitals.Methods Based on the monthly number of CRKP detected in a tertiary hospital in Jiaxing City from 2016 to 2021, the ARIMA multiplicative seasonal model was constructed with SPSS 26.0 software. The actual number of CRKP plants detected in 2022 was selected as the validation data set and compared with the predicted value, so as to evaluate the prediction efficiency of the model.Results The total number of CRKP detected in hospital showed a downward trend year by year from 2016 to 2021. The incidence was concentrated from July to September, and the peak was in September, which was seasonal and periodic. The Bayesian information criterion (BIC) of the fitted ARIMA (0, 1, 2) (0, 1, 0)12 was 0.43, and the mean absolute percent error (MAPE) was 16.54. The difference of Ljung-box test of model residual sequence was not statistically significant (Q=11.06, P=0.81). The residual sequence was white noise, indicating that the model fitted well and was determined as the optimal model. The number of CRKP strains detected from January to December 2022 was used to verify the prediction effect of the model. The results showed that the actual incidence trend was in good agreement with the prediction curve, and the actual values were all within the 95% confidence interval of the predicted values, the prediction error was -33.33%25.00%, and the average relative error was 15.56%, indicating that the prediction effect of the model was good.Conclusion ARIMA multiplicative seasonal model can fit the time change of the number of CRKP detected in the hospital well, so it can be used for short-term prediction and dynamic analysis of CRKP epidemic trend, and provide basis for the early warning and prevention of CRKP infection in hospitals.

Key words： autoregressive integrated moving average model multiplicative seasonal model Carbapenem resistant Klebsiella pneumoniae prediction

收稿日期: 2023-02-28

PACS:

基金资助:浙江省医药卫生科技计划项目（2021KY1107）；

嘉兴市科技计划项目（2021AD30163）；

嘉兴市第一医院院级课题（2021-YA-076）

通讯作者: 王霄腾，Email:rainbowlove44@126.com

引用本文:

张绮萍，季聪华，陆锦琪，王霄腾. 应用ARIMA乘积季节模型对耐碳青霉烯类肺炎克雷伯菌流行趋势预测效果研究[J]. 中国医院统计, 2023, 30(3): 173-178.
Zhang Qiping， Ji Conghua，Lu Jinqi， Wang Xiaoteng. Application of ARIMA multiplicative seasonal model in predicting the epidemic trend of carbapenem resistant Klebsiella pneumoniae

. journal1, 2023, 30(3): 173-178.

链接本文:

http://manu58.magtech.com.cn/Jwk_zgyytj/CN/10.3969/j.issn.1006-5253.2023.03.003 或 http://manu58.magtech.com.cn/Jwk_zgyytj/CN/Y2023/V30/I3/173

［1］REYES J, AGUILAR A C, CAICEDO A. Carbapenem-resistant Klebsiella pneumoniae: Microbiology key points for clinical practice［J］. Int J Gen Med, 2019, 12:437-446. DOI: 10.2147/ijgm.s214305.

［2］权敏，吕晓菊．碳青霉烯耐药菌感染的诊治进展［J］．中华传染病杂志，2022，40(4)：193-197．DOI: 10.3760/cma.j.cn311365-20210715-00253.

［3］DURANTE-MANGONI E, ANDINI R, ZAMPINO R. Management of carbapenem-resistant Enterobacteriaceae infections［J］. Clin Microbiol Infect, 2019, 25(8):943-950.DOI: 10.1016/j.cmi.2019.04.013.

［4］黄丹艳，史秋橙，蓝鹏，等．浙江地区碳青霉烯类耐药肺炎克雷伯菌分布、流行及耐药基因分析［J］．中华传染病杂志，2018，36(1)：7-11．DOI: 10.3760/cma.j.issn.1000-6680.2018.01.004.

［5］党耀国,王正新,钱吴永,等.灰色预测技术方法［M］.北京：科技出版社，2016:15-19.

［6］CDC.Facility guidance for control of carbapenem-resistant Enterobacteriaceae(CRE)［EB/OL］.［2023-01-22］.https://www.cdc.gov/hai/pdfs/cre/cre-guidance-508.pdf.

［7］ZENG Q L, LI D D, HUANG G, et al. Time series analysis of temporal trends in the pertussis incidence in Mainland China from 2005 to 2016［J］. Sci Rep, 2016, 6:32367. DOI: 10.1038/srep32367.

［8］刘峰，朱妮，邱琳，等．ARIMA乘积季节模型在陕西省手足口病预测中的应用［J］．中华流行病学杂志，2016，37(8)：1117-1120．DOI: 10.3760/cma.j.issn.0254-6450.2016.08.013.

［9］王燕．应用时间序列分析［M］．4版．北京：中国人民大学出版社，2015．

［10］陈亚男，刘菁，李爱民，等．基于多学科诊疗模式降低重症监护病房碳青霉烯类耐药肠杆菌科细菌感染的效果评价［J］．中华临床感染病杂志，2020，13(3)：182-188．DOI: 10.3760/cma.j.issn.1674-2397.2020.03.005.

［11］中华预防医学会医院感染控制分会，中华医学会感染病学分会，中国医院协会医院感染管理专业委员会，等．中国碳青霉烯耐药革兰阴性杆菌(CRO)感染预防与控制技术指引［J］．中华医院感染学杂志，2019，29(13)：2075-2080．DOI: 10.11816/cn.ni.2019-191088.

［12］MIT- B-Y-K S,瘙塁 EYDA E. Improving forecasting accuracy of time series data using a new ARIMA-ANN hybrid method and empirical mode decomposition［J］. Neurocomputing, 2019, 361:151-163. DOI: 10.1016/j.neucom.2019.05.099.

［13］蔡晶，黄淑琼，杨雯雯，等．2010—2018年湖北省猩红热流行病学特征及发病趋势预测［J］．中华疾病控制杂志，2020，24(2)：134-138，150．DOI: 10.16462/j.cnki.zhjbkz.2020.02.003.

［14］储文杰，金凯玲，林凯，等．基于ARIMA乘积季节模型预测产超广谱β-内酰胺酶大肠埃希菌流行趋势研究［J］．预防医学，2018，30(7)：680-684．DOI: 10.19485/j.cnki.issn2096-5087.2018.07.008.

［15］陈欣，吴晓敏，包名家，等．佳木斯市2004—2017年流行性腮腺炎流行特征分析及趋势预测［J］．中华疾病控制杂志，2019，23(2)：185-190．DOI: 10.16462/j.cnki.zhjbkz.2019.02.013.

［16］尹丹萍，周国清．耐碳青霉烯类肠杆菌科细菌临床分布特点及耐药性研究［J］．中华医院感染学杂志，2018，28(18)：2752-2755．

［17］DAVID S, REUTER S, HARRIS S R, et al. Epidemic of carbapenem-resistant Klebsiella pneumoniae in Europe is driven by nosocomial spread［J］. Nat Microbiol, 2019, 4(11):1919-1929. DOI: 10.1038/s41564-019-0492-8.

［18］李忠奇，陶必林，占梦瑶，等．时间序列模型应用于新型冠状病毒肺炎疫情预测效果比较研究［J］．中华流行病学杂志，2021，42(3)：421-426．DOI: 10.3760/cma.j.cn112338-20201116-01333.

［19］樊雯婧，陈健，楼冬洁，等．基于ARIMA模型对海南某医院的医院感染发病率的预测研究［J］．中国消毒学杂志，2020，37(9)：670-673．DOI: 10.11726/j.issn.1001-7658.2020.09.010.