1998年的春天，台灣爆發全球最嚴重的第71型腸病毒疫情，且疫情迅速擴散，奪走多條幼兒的寶貴生命，造成全國民眾的恐慌，更耗費無數的社會資源與成本；然而如何有效且確切地掌握病例的空間分布情形已成為現今公部門在面臨疫情蔓延時的重要工作。在過去20年間，地理資訊系統(Geographic Information System, GIS)迅速發展，已擴大應用於醫學地理學及公共衛生、流行病學領域等，其分析功能，從視覺分析，進而探索(exploratory)和模式(model)分析。
本研究為了了解腸病毒是否具有顯著空間聚集的現象，因此採用空間自相關中的全域空間自相關(global spatial autocorrelation)先進行腸病毒一般病例和重症病例空間現象的整體分布狀況之分析，並判斷此現象是否有空間聚集特性存在，再將有空間聚集特性存在的年份利用區域空間自相關(local spatial autocorrelation)找出空間現象的聚集位置，發現歷年熱點區有往北移動的現象，且重症病例的熱點區擴散較一般病例更北移，其北移位置最北已來到彰化、台中地區。
In the spring of 1998, the most serious epidemic of Enterovirus 71 (EV71) broke out in Taiwan and spread rapidly throughout the whole of the island. It claimed a number of precious lives of young babies, and caused a nation-wide panic, at a huge expense of social costs and resources. Therefore, how to effectively and precisely understand the spatial distribution of cases has become a significant work facing the related government sections. Due to the rapid improvement of GIS (Geographic Information System technique) and its applications in medical geography, public health, and epidemiology, the analyzing function has been expanded from simply visual analysis, to the analyses of exploration and modeling.
In order to investigate if there was a significant difference in spatial clusters of Enterovirus risk, the technique of "spatial autocorrelation" was employed in this paper to test and analyze the data pertaining to the epidemics of enterovirus in Taiwan from 1999 to 2005. The results showed that the hot spots of the cases of Enterovirus generally shifted northward from south Taiwan, especially Tainan area, to central Taiwan, namely, Taichung and Changhua areas. More interesting thing is that the hot spots of the cases of Enterovirus 71 had moved to further north than that of other Enterovirus.
