摘要
Abstract
Objective To explore the relationship between hepatitis E and meteorological factors.Methods The monthly reported cases of hepatitis E and meteorological data in Changsha from 2004 to 2015 were collected.Through the comparison of discrete test and model fitting effect indicators,advantages and disadvantages of Poisson regression model and negative binomial regression model were evaluated.The optimal model was selected to analyze the meteorological factors of the incidence of hepatitis E.Results From 2004 to 2015,there were 1 355 cases reported with hepatitis E in Changsha.The annual incidence was (0.21-2.77) per 100 000 people.There were sporadic cases in each month,but most occurred between February and May.According to the results of correlation analysis of meteorological data,monthly total precipitation,monthly average air pressure,monthly average wind speed,monthly average temperature,monthly average relative humidity,monthly days of raining and monthly sunshine hours were enrolled into the model.The indicators of discrete test and model fitting effect showed that the effect of negative binomial regression model was better than that of Poisson regression model.Negative binomial regression model analysis showed that monthly average temperature 〔b=-0.024,95%CI (-0.036,-0.013)〕,monthly days of raining 〔b=0.047,95%CI(0.032,0.062)〕,monthly sunshine hours 〔b =0.003,95%CI (0.001,0.005)〕 were influence factors of hepatitis E in Changsha (P<0.05).Conclusion Lowered average temperature,more rainy days and longer sunshine hours are the meteorological factors of increased number of cases with hepatitis E.关键词
戊型肝炎/天气和发病/二项分布/影响因素分析Key words
Hepatitis E/Weather and invasion/Binomial distribution/Root cause analysis分类
医药卫生
