Classification and regression tree model study on correlation between influenza epidemic and meteorological factors in different areas of Gansu, 2010 to 2016

ObjectiveTo explore the influence of meteorological factors on the influenza epidemic intensity in Gansu province and provide scientific evidence for the improvement of timeliness of influenza prevention and control.

MethodsThree surveillance areas were selected, i.e. Baiyin, Tianshui and Jiuquan. The influenza surveillance data and meteorological data of the three areas were collected, and the lag time of various meteorological factors on the incidence of influenza was analyzed by cross correlation analysis. Classification and regression tree (CART) was used model to select the meteorological factors which had the greatest impact on the epidemic intensity.

ResultsMeteorological factors correlation studies showed that temperature and humidity were negatively correlated with influenza epidemic intensity. The influenza epidemic intensity in Tianshui increased by 1.89 times when the 3 weeks moving average of maximum temperature was below 19.11 ℃; the influenza epidemic intensity in Jiuquan increased by 2.01 times when the 3 weeks moving average of average temperature was below 8.03 ℃; the influenza epidemic intensity of Baiyin increased by 2.24 times when the 3 weeks moving average of average temperature was below 8.68 ℃ and the 14 weeks moving average of relative humidity was below 61.04%.

ConclusionMeteorological factors could affect the incidence of influenza, cold and dry weather would be suitable for the spread of influenza in population.