湖南省流行性感冒与气象因素关系及预测  

Association of meteorological factors with influenza incidence in Hunan province: a time series model based analysis and prediction

作　　者：查文婷[1] 李渭通 何嘉慧 何嘉琪 吕媛[1] 刘颖 邹享玉[1] 易尚辉[1] ZHA Wen-ting;LI Wei-tong;HE Jia-hui(Hunan Province Key Laboratory for Molecular Epidemiology,School of Medicine,Hunan Normal University,Changsha,Hunan Province 410000,China)

机构地区：[1]湖南师范大学医学院"分子流行病学"湖南省重点实验室,湖南长沙410000

出　　处：《中国公共卫生》2021年第3期537-541,共5页Chinese Journal of Public Health

基　　金：湖南省教育厅项目(湘教通[2017]452号)。

摘　　要：目的分析湖南省气象因素与流行性感冒(流感)发病的关系,建立时间序列模型,并对预测效果进行评价,为今后制定更科学有效的流感防控策略提供参考依据。方法利用SPSS 20.0软件建立数据库,采用Spearman秩相关对气象因素和流感发病数进行相关性分析,以2008—2016年湖南省流感发病数为基础构建时间序列模型,对2017年1—12月流感发病数进行预测并评价预测效果。结果 (1)2008—2017年全年均有流感病例报告,高发季节为冬春季;(2)流感发病数与月平均气温(r=–0.195,P <0.05)、平均日照时数(r=–0.483,P <0.05)、平均日降水量(r=–0.116,P <0.05)呈负相关,与月平均气压(r=0.195,P <0.05)、平均相对湿度(r=0.260,P <0.05)、平均最大风速(r=0.279,P <0.05)呈正相关;(3)ARIMA(1,0,0)(0,1,1)_(12)模型和引入气象因素后的ARIMAX模型对2017年1—6月流感发病数预测的平均相对误差分别为43.90%和31.48%,对2017年7—12月流感发病数预测的平均相对误差为70.65%和67.17%。结论引入气象参数后的ARIMAX模型较ARIMA模型预测效果更好,但两个模型对2017年7—12月预测效果均不佳,今后将纳入更多的因素完善ARIMAX模型。Objective To explore the association of meteorological factors with influenza incidence in Hunan province with time series model analysis and to evaluate predictive effectiveness of the established model for developing influenza prevention and control strategies. Methods We retrieved the data on reported influenza incidents and daily meteorological parameters in Hunan province and divided the data into three sets: a dataset from January 2008 through December 2016 for establishing a time series model and two datasets of first and second half year of 2017 for assessing effectiveness of the constructed model in influenza incidence prediction. SPSS 20.0 software was used to process the data collected and Spearman correlation was adopted to analyze the correlation between meteorological factors and influenza incidence.Results(1) Influenza cases were reported throughout the period from 2008 to 2017 and more cases were reported during winter and spring seasons.(2) The monthly number of reported influenza cases was correlated negatively with monthly average temperature(r =–0.195, P < 0.05), sunshine hours(r =–0.483, P < 0.05), and amount of precipitation(r =–0.116,P < 0.05) but positively with monthly average air pressure(r = 0.195, P < 0.05), relative humidity(r = 0.260, P < 0.05), and maximum wind speed(r = 0.279, P < 0.05).(3) Aafter introducing meteorological factors, the average relative errors of autoregressive integrated moving average(ARIMA)(1, 0, 0)(0, 1, 1) _(12) model and ARIMAX model were 43.90% and31.48% for the prediction of influenza incidence from January to June 2017 and 70.65% and 67.17% from July to December2017. Conclusion The predictive effectiveness of ARIMAX model is better than that of ARIMA model after the introduction of meteorological parameters, but the two established models are not good when used to predict the influenza incidence for the period of July–December 2017, suggesting that more factors should be included for improving the ARIMAX model.

关 键 词：流行性感冒 气象因素 时间序列模型 预测 

分 类 号：R183.3[医药卫生—流行病学] R511.7[医药卫生—公共卫生与预防医学]