The Effect Of Meteorological Factors On The Incidence Of Influenza Among Residents In Shenyang From 2010 To 2019 And The Analysis Of SARIMA Time Series Model

Objectives:To grasp the epidemic situation of influenza-like cases（ILI）in Shenyang,and to discuss the relationship between meteorological factors and influenza-like cases in Shenyang and its lag effect.The time series model is used to predict the incidence trend of influenza and provide some theoretical support for reducing the loss caused by influenza.Methods:Data of influenza-like cases in Shenyang from 2010 to 2019 were collected from 18 sentinel hospitals in Shenyang,and meteorological factors in the same period were collected from the National Meteorological Data Sharing Center（http://www.cma.gov.cn）.Descriptive epidemiological methods were used to describe and analyze the characteristics of disease data and meteorological data.The distributed lag nonlinear model（Distributed lag non-linear model,DLNM）was used to evaluate the influence of meteorological factors on influenza-like cases.The impact of extreme weather on HFMD was quantified by comparing the 90th or 10th percentile of meteorological variables to the median,and the seasonal differential autoregressive moving average model（SARIMA）was used to predict the percentage of influenza-like cases（ILI%）per month in Shenyang in 2020.Result:（1）There were 887346 influenza-like cases in Shenyang from 2010 to 2019,and the percentage of influenza-like cases was 3.21%.There are two high peaks of influenza throughout the year,winter and spring and summer,and the number of cases is significantly higher than other times.The patients are mainly old people and children,among which children younger than 4 years old have the highest incidence.（2）meteorological factors in high temperature,high pressure,high humidity,low humidity,sunshine for the risk factors for long,increase the risk of influenza-like illness,in selected lag period’s biggest RR value respectively（lag:7 RR=1.13,95%CI:1.05～1.20）,（lag:14 RR=1.31,95%CI:1.21～1.41）,（lag:14 RR=1.12,95%CI:1.05～1.21）,（lag:14 RR=1.13,95%CI:1.06～1.21）,（14 RR=1.30,95%CI:1.21～1.38）.Low temperature was the protective effect,and the minimum RR was（lag:5 RR=0.88,95%CI:0.83-0.95）.（4）The model SARIMA（1,1,1）（1,1,1）₁₂ is constructed on a weekly basis to predict the monthly ILI%in 2020.Conclusion:（1）Shenyang epidemiological characteristic is seasonal,with a double peak in winter,spring and summer.Children and the elderly are the most susceptible population.（2）There is a nonlinear lag relationship between meteorological factors and influenza-like cases,in which high temperature,high pressure,high humidity and low humidity increase the risk of influenza,and there is a cumulative effect.Low temperature has a protective effect on the onset of influenza-like illness.（3）SARIMA model fitting prediction effect is different from the true value.