吉林省肾综合征出血热发病预测研究

Prediction of hemorrhagic fever with renal syndrome in Jilin Province

目的通过对2007—2018吉林省肾综合征出血热病例进行流行特征描述,使用时间序列分析,预测吉林省未来的肾综合征出血热发病趋势。方法通过建立ARIMA模型,对2007—2018年间吉林省肾综合征出血热发病数进行拟合,使用2019—2021年发病数与预测结果之间进行拟合优度检验来验证拟合效果。结果2007—2018年间,吉林省共上报肾综合征出血热8844例。总体来说吉林省肾综合征出血热发病率呈现缓慢下降趋势,时间序列分析结果为:ARIMA(1,0,0)(0,1,2)12较好地拟合和预测吉林省肾综合征出血热发病数(R2=0.66,P<0.05)。2019年至2021年预测肾综合征出血热发病数分别为510人、449人和513人。将2019—2021年HFRS实际病例数并与ARIMA模型预测数据进行比较,计算相关系数R2=0.85,P<0.05。结论ARIMA模型能较好地拟合和预测吉林省HFRS发病数。

Objective The epidemiological characteristics of the cases of renal syndrome hemorrhagic fever(HFRS)in Jilin Province from 2007 to 2018 were characterized and time series analysis was used to predict the future trend of HFRS incidence in Jilin Province.Methods The number of incidences of HFRS in Jilin Province from 2007 to 2018 was fitted by building an ARIMA model,and the goodness-of-fit test between the number of incidences and the predicted results for 2019 to 2021 was used to verify the fit.Results From 2007 to 2018,a total of 8844 cases of HFRS were reported in Jilin Province.Overall,the incidence of renal syndrome haemorrhagic fever in Jilin Province presented a slow decreasing trend,and the results of the time series analysis were:ARIMA(1,0,0)(0,1,2)12better fitted and predicted the number of renal syndrome haemorrhagic fever cases in Jilin Province(R2=0.66,P<0.05).The predicted number of HFRS cases from 2019 to 2021 were 510,449 and 513,respectively.The incidence data of HFRS from 2019 to 2021 was compared with the predicted data of the model,and the correlation coefficient was calculated asR2=0.85,P<0.05.Conclusion The ARIMA model was able to better fit and predict the number of HFRS in cidences in Jilin Province.