The equine influenza epidemic in Australia: Spatial and temporal descriptive analyses of a large propagating epidemic

Australia experienced a large outbreak of equine influenza in August 2007. Nearly 10 000 premises were infected during the epidemic. We used spatial and temporal analytical techniques to describe the epidemic, to quantify important descriptors of the epidemic, and to generate hypotheses about how the epidemic progressed and which control tools assisted in eradication. Spatio-temporal epidemic curves revealed three phases in the epidemic: dispersal, local spread and disease fade out. Spatial dispersal of infection rapidly declined immediately after national movement restrictions were introduced. The epidemic peak had passed before emergency vaccination could have induced substantial immunity in the equine population. Thirty seven clusters of epidemiologically linked premises were delineated using an interpolated surface of date of onset of clinical signs, geographic data and location of infected premises. These clusters were analysed individually to parameterise key epidemic measures: cumulative incidence, incidence rate, effective reproduction rate, nearest neighbour spread distances, epidemic length and the number of infected premises. These measures were summarised by landscape (rural versus peri-urban location) and standardised by disease management zone. Compared to rural areas, peri-urban areas appeared to have a higher density of equine premises (4.66 premises km−2 versus 0.80 premises km−2), longer epidemics (95 days versus 87 days), more infected premises (393 versus 339) and a shorter spread distance (1.27 km versus 2.38 km). However, effective reproduction rates (2.04 versus 1.99), cumulative incidence (27.4 versus 26.9) and incidence rates (1.36 versus 1.54) were similar. The relative impact of vaccination and national movement restrictions in controlling this epidemic needs further investigation.