Applying Bayesian network modelling to understand the links between on-farm biosecurity practice during the 2007 equine influenza outbreak and horse managers’ perceptions of a subsequent outbreak.

This research article discusses an approach to modelling the factors that influence how effective horse farm managers perceived biosecurity measures during an outbreak of equine influenza in Australia in 2007, in terms of their potential effectiveness during a future outbreak.

Research Background

• The genesis of this research was Australia’s first-ever Equine Influenza outbreak in August 2007 that saw horses on 9,359 premises get infected over five months.
• This disease was eradicated using a mix of horse movement controls, on-farm biosecurity and vaccination. A case-control study of this outbreak at the premises level highlighted the protective effect of several variables denoting on-farm biosecurity practices.
• Further, factors that influenced horse managers’ perceptions of the viability and effectiveness of these biosecurity measures were also identified separately.

Research Objective and Methods

• The focus of this analysis is to apply additive Bayesian network modelling to draw a comprehensive and nuanced picture of the associations between variables denoting on-farm human behaviour during the outbreak and the perceptions of horse managers of the effectiveness of such measures in the event of a subsequent outbreak.
• Using Heuristic structure discovery, a robust statistical model was created for 31 variables representing biosecurity practices and perceptions of the owners and managers of 148 premises.

Research Findings and Significance

• The resulting Bayesian graphical network model statistically describes the associations linking horse managers’ on-farm biosecurity practices during the outbreak risk period in 2007 and their perceptions on the effectiveness of such measures during a potential future outbreak.
• It was found that practicing barrier infection control measures were linked with a heightened perception of preparedness. Meanwhile, horse managers who considered their on-farm biosecurity measures during the outbreak to be more stringent than their normal practices had a increased perception of the efficacy of other measures like controlling access to the premises.
• The research offers insights into how past performance under crisis could be a reliable predictor of future perceptions and provides guidance on biosecurity practices from the perspective of horse owners and managers for better infection control.