论文部分内容阅读
目的应用空间自相关分析和空间扫描统计分析甘肃省2013年细菌性痢疾发病的空间分布特征,探讨空间自相关性和聚集范围。方法收集“中国疾病控制信息系统”中2013年甘肃省87个县(区)细菌性痢疾报告病例资料,采用Geoda 1.60软件进行空间全局和局部自相关分析,Sa TScan 9.1.1.0软件进行空间扫描,分析结果使用Arc GIS 10.2软件进行可视化地图展示。结果 2013年甘肃省细菌性痢疾报告发病数8191例,报告发病率为31.81/10万。总体层面上具有空间自相关性(Moran’s I=0.4555,Z=6.51,P=0.001);局部空间自相关分析,甘肃省东部的庆阳市和南部甘南州的9个县(区),呈高值聚集状态,为细菌性痢疾发病的“热点”区域,中西部的金昌市、张掖市和武威市的11个县(区),呈低值聚集状态,是细菌性痢疾发病“冷点”区域。空间扫描探测到的主要聚集区为兰州市及周边共5个县(区)(LLR=137.10,RR=2.38);庆阳市的9个县(区)(LLR=428.60,RR=2.40)。结论 2013年甘肃省细菌性痢疾空间分布呈非随机分布,具有空间自相关性,存在明显聚集性。
Objective To analyze the spatial distribution characteristics of bacterial dysentery in Gansu Province in 2013 by means of spatial autocorrelation analysis and spatial scanning statistics, and explore the spatial autocorrelation and aggregation range. Methods The data of bacterial dysentery cases in 87 counties (districts) of Gansu Province in 2013 were collected from China Disease Control Information System. The spatial and local autocorrelation analysis were performed using Geoda 1.60 software. Sa TScan 9.1.1.0 software was used for space Scan, analyze the results Use Arc GIS 10.2 software for visual map display. Results The number of reported cases of bacterial dysentery in Gansu Province in 2013 was 8,191, with a reported incidence rate of 31.81 / 100,000. Spatial autocorrelation (Moran’s I = 0.4555, Z = 6.51, P = 0.001); local spatial autocorrelation analysis showed that in Qingyang City in eastern Gansu Province and 9 counties in Southern Gannan Prefecture, Value aggregation state, the incidence of bacterial dysentery “hot spot ” area, Jinchang City in central and western China, Zhangye City and Wuwei City, 11 counties (districts), was a low value aggregation state, is the incidence of bacterial dysentery Point "area. The main agglomeration areas detected by space scanning are 5 counties (districts) in Lanzhou City and its surroundings (LLR = 137.10, RR = 2.38) and 9 counties (districts) in Qingyang City (LLR = 428.60, RR = 2.40). Conclusion The spatial distribution of bacterial dysentery in Gansu Province in 2013 showed a non-random distribution with spatial autocorrelation and obvious aggregation.