Porcine reproductive and respiratory syndrome (PRRS) has a worldwide distribution. This economically important endemic disease causes reproductive failure in breeding stock and respiratory tract illness in young pigs. In Ontario restricted fragment length polymorphism (RFLP) 1-18-4 has been determined as one of the most common virus genotypes. Individual-level models (ILMs) for infectious diseases, fitted in a Bayesian MCMC framework, have been used to describe both the spatial and temporal spread of diseases. They are an intuitive and flexible class of models that can take into account population heterogeneity via various individual-level covariates. The objective of this study is to identify relative importance of risk factors for the spread of the genotype 1-18-4 from monitoring data in southern Ontario using ILMs. Specifically, we explore networks through which resources are obtained or delivered, as well as the ownership structure of herds, and identify factors that may be contributing to high risk of infection. A population of 316 herds which experienced their PRRS outbreaks between September 2004 and August 2007 are included in the analyses, in which 194 (61%) are sow herds. During the study period, 45 herds (27 sow herds) experienced their first outbreak due to RFLP 1-18-4. Our results show that the three relatively most important factors for the spread of 1-18-4 genotype in Ontario swine herds were sharing the same herd ownership, gilt source and market trucks. All other networks had relatively smaller impact on spread of this PRRSV genotype. Spatial proximity could not be identified as important contributor to spread. Our findings also suggest that gilt acclimation should be practiced whenever possible and appropriate to reduce the risk for the herd and for others as it is already widely implemented and recommended in the North American swine industry.
Copyright © 2013 Elsevier B.V. All rights reserved.