Spatial Dynamics of the Severe Acute Respiratory Syndrome (SARS) Epidemic in Hong Kong in 2003

Abstract

The Severe Acute Respiratory Syndrome (SARS) epidemic in 2003 was the first infectious disease outbreak caused by a novel pathogen in the twenty-first century. The outbreak in Hong Kong was the second largest worldwide and was characterised by a large proportion of hospital infections and a super-spreading event caused by environmental factors in residential buildings. Hospitals treating SARS cases were at high risk for transmission. I found that hospital outbreaks triggered community transmission as well as the formation of spatial clusters of community cases. The size of the community outbreak in an area increased with the size of the outbreak in the nearest hospital treating SARS, and an area was more likely to have no community-infected cases if it was far from hospitals treating SARS, or had less hospital-infected cases within the area. To quantify the transmission between hospital and community, I developed a spatial epidemic-tree-reconstruction method that uses gravity models to spatially define the probability of contact between individuals in the community. From the reconstructed probabilistic infection tree, I estimated that 24% of community transmission was likely to be infected by cases infected in hospitals, with infected patients discharged during their incubation period and hospital visitors the most important drivers of transmission from healthcare settings to the community. Healthcare workers were key drivers of hospital transmission, with the hospital-to-hospital reproduction number, excluding a single hospital super-spreading event, estimated to be 0.8. A typical community-acquired case was estimated to generate 0.6 cases in the community and 0.2 cases in the hospital in which they were subsequently hospitalised. My findings suggest that hospital infection control could be improved. Restricted hospital visitor policies could have been imposed for longer time during the outbreak and quarantine could be considered for those who recently visited or have been discharged from hospitals treating SARS cases.