Landscape, demographic and climatic associations with human West Nile virus occurrence regionally in 2012 in the United States of America
Landscape, demographic and climatic associations with human West Nile virus occurrence regionally in 2012 in the United States of America
复制标题
DOI：
10.4081/gh.2014.13
复制
发表时间：
2014-11-01
影响因子：
1.7
通讯作者：
Ecker, Mark
中科院分区：
文献类型：
--
作者：
DeGroote, John P.;Sugumaran, Ramanathan;Ecker, Mark
在美利坚合众国（美国）西尼罗河病毒（西尼罗河病毒）的低发病率数年之后，2012年在该国若干地区发生了大规模疫情。为了了解疫情动态、空间聚类和景观、人口和气候与美国西尼罗河病毒发生的关系，在区域一级进行了调查。以前的研究已经表明，在美国有少数几种主要的西尼罗河病毒蚊子媒介，它们具有不同的生态要求，负责西尼罗河病毒的传播。这些重要媒介的已发表的范围图被用于地理参考，并用于确定美国邻近地区的八个功能生态区域。为了计算2012年各县的西尼罗河病毒发病率，统计了各县的西尼罗河病毒病例数和人口数量。此外，根据每个县的西尼罗河病毒发病率高于或低于其所在地区的发病率，计算出每个县的二进制值（高/低）。利用全球Moran’s I和Anselin本地Moran’s I空间关联统计数据对各区域的WNV率和高/低等级进行聚类分析和可视化。利用Spearman rho相关系数和泊松回归模型，分析了不同来源的景观、人口和气候变量的空间数据与人类西尼罗河病毒的关系。研究结果表明，西尼罗河病毒具有显著的空间聚类性，并且西尼罗河病毒的发生与景观、人口和气候相关变量之间的关系存在显著的区域差异。这些区域间的关联与优势病媒生物的生态是一致的。在东南部地区发生大规模疫情之前，冬季和春季降水高于正常水平，夏季干燥炎热。
After several years of low West Nile virus (WNV) occurrence in the United States of America (USA), 2012 witnessed large outbreaks in several parts of the country. In order to understand the outbreak dynamics, spatial clustering and landscape, demographic and climatic associations with WNV occurrence were investigated at a regional level in the USA. Previous research has demonstrated that there are a handful of prominent WNV mosquito vectors with varying ecological requirements responsible for WNV transmission in the USA. Published range maps of these important vectors were georeferenced and used to define eight functional ecological regions in the coterminous USA. The number of human WNV cases and human populations by county were attained in order to calculate a WNV rate for each county in 2012. Additionally, a binary value (high/low) was calculated for each county based on whether the county WNV rate was above or below the rate for the region it fell in. Global Moran's I and Anselin Local Moran's I statistics of spatial association were used per region to examine and visualize clustering of the WNV rate and the high/low rating. Spatial data on landscape, demographic and climatic variables were compiled and derived from a variety of sources and then investigated in relation to human WNV using both Spearman rho correlation coefficients and Poisson regression models. Findings demonstrated significant spatial clustering of WNV and substantial inter-regional differences in relationships between WNV occurrence and landscape, demographic and climatically related variables. The regional associations were consistent with the ecologies of the dominant vectors for those regions. The large outbreak in the Southeast region was preceded by higher than normal winter and spring precipitation followed by dry and hot conditions in the summer.