Spatial Analysis Of Epidemiological Data In Henan Province On The Case Of Bacillary Dysentery

Spatial analysis is a process of extracting spatial information or deriving new data from existing data, and processing spatial data into useful information. In the field of epidemiology research, spatial analysis is applied to explore the spatiotemporal distribution of cases, identify high-risk areas and related risk factors, predict and control disease based on the time, space and environment factors information of cases. Currently, related studies on the application of spatial analysis in epidemic field is less in China, but increasing rapidly, and is becoming a new technical means of epidemiology and public-health research.Densely populated in Henan province, especially in the rural areas with relatively backward economic development and public-health conditions, so it is of great significance to detect time-space clusters with high incidence, quantitative analysis of the risk factors for controlling bacterial dysentery outbreak, and cut losses of public security and economic. It is an important innovation of this study to explore application of spatial analysis in epidemiology and public-health field from the perspective of time and space. The research methods in this study can provide references for related research, and the results of the study has important reference meanings for understanding the space-time distribution law and risk factors of bacillary dysentery, formulating relevant policies and measures of prevention and control, and allocating medical resources reasonably for public health departments in Henan province.This study conducted spatial analysis on bacterial dysentery cases data, by firstly discussing the main research contents basic theory and method the GIS spatial analysis method in the field of epidemiology, performing descriptive analysis on the basis of data statistics of bacillary dysentery, analyzing the incidence of spatial correlation and heterogeneity, conducting scan statistics analysis to detect time and space high-risk gather zones of bacterial dysentery, and exploring the relationship between incidence and socio-economic factors, such as population density, per capita GDP, and meteorological factors such as monthly average temperature, monthly rainfall, wind speed respectively by building a geographically weighted regression model and spatial panel data model.The main conclusions are as follows:(1) This paper summarized the main research contents which were applied by the current spatial analysis method in the field of public health, expounded related basic theory of spatial analysis and methods applied in the study of epidemic data, and introduced the main spatial analysis softwares or software packages used in studies on epidemiological.(2) Descriptive statistics analysis found that male proportion in the number of illness and incidence were higher than women’s on the distribution of gender, the men’s incidence were 1.54 times that of women’s; On age distribution, the incidence of 0 to 14 and more than 60 years old age group were much higher than that of 15 to 60 years old age group; On professional distribution, farmers and scattered children occupied larger proportion in total cases; On time distribution, summer and fall were peak time of bacterial dysentery with higher incidence.(3)Moran’s I indicated that the spatial distribution of bacillary dysentery incidence in Henan province existed positively spatial autocorrelation, and Moran’s I based on the common boundary or common point(Queen) relationship was the biggest; Moran’s I calculated based on distance was the largest when the distance is the default minimum, 55.77 km..(4) The spatial distribution of the local index spatial autocorrelation(LISA) showed that there were positive correlation in most of the counties(city), while some positive area surrounded by negative zone, indicating that bacillary dysentery incidence at the county level existed spatial heterogeneity.(5) The results of the space scan statistics and space-time scan statistics indicated that epidemic outbreak of bacillary dysentery in different levels appeared within the scope of the province, the most likely spatial cluster was located in most of the counties in Shangqiu and Zhoukou in the east, secondary cluster covered some counties in Luoyang, Sanmenxia, Pingdingshan in the west. Space-time scan statistics results showed the duration of epidemic outbreak of bacillary dysentery were mainly from May to October, and also showed that there were epidemic outbreak with smaller scale in Zhumadian, Anyang, Puyang, et al.(6) The analysis of geographically weighted regression model found that there were positive correlation between the incidence of bacillary dysentery and population density, and negative correlation between the incidence of bacillary dysentery and per capita GDP.(7) The statistical results of the spatial panel data model showed that the monthly incidence of bacillary dysentery was positively correlated with monthly average temperature and rainfall(P < 0.05), was positively correlated with wind speed(p<0.05), and the goodness of fit test of the spatial error model was better than panel data model and spatial lag model.