A Spatial Point Pattern Analysis of the 2003 SARS Epidemic in Beijing

Abstract

Beijing was the most prevalent city of SARS in China in 2003. The study on the spatial distribution and clustering characteristics is helpful to deeply understand the epidemic of SARS in Beijing. In this paper, the home addresses of SARS patients accquired by investigation were considered as the spatial location, deriving 2321 cases of the spatial distribution and incidence rate of infected patients. Kernel estimation method is used to obtain the density distribution of SARS patients. The results indicate that the distribution density of infected people is gradually attenuated from the center of the city to the suburbs. Ripley'K function is also used to explore the spatial clustering characteristics of SARS infection. In addition, the influence of gender, contact history and SARS Beijing Xiaotangshan Hospital towards the spatial clustering of patients are analyzed and thus shows that the spatial clustering of patients is the strongest at 11km distance. Gender and history of exposure to SARS infection in the spatial clustering are of a small impact, while SARS Beijing Xiaotangshan Hospital on SARS infection in the spatial clustering are of a strong impact. The clustering characteristics are significantly weaker after the establishment of the hospital that shows the importance of the establishment of the hospital on prevent and control of SARS epidemic in Beijing.