Risk analysis of classical swine fever introduction

Keywords:Classical swine fever; Cost-effectiveness; European Union; Pathway diagram; Prevention; Risk analysis; Scenario tree model; Sensitivity analysis; The Netherlands; Virus introductionThe research described in this thesis aimed to support decision-making on preventing classical swine fever virus (CSFV) introduction into disease-free regions of the European Union (EU). A risk analysis of CSFV introduction was performed. The main objective was to provide quantitative insight into the major risk factors contributing to the probability of CSFV introduction (P CSFV ). First, a pathway diagram was constructed presenting an overview of all factors that might possibly contribute to the P CSFV . Then, based on this pathway diagram, a scenario tree model was developed to calculate the annual P CSFV into member states of the EU. The main aim of this model was to analyse quantitatively those pathways that contribute most to the annual P CSFV, and the origin of these pathways. Pathways included in the model were import of pigs and pork products, returning livestock trucks, and contacts with wild boar. All (old) 15 EU member states were included as possible sources of CSFV introduction. Because the model contained many uncertain input parameters, an extensive sensitivity analysis was performed to indicate which of these parameters most influenced model results. The results indicated that only four out of 257 uncertain input parameters changed the ranking of risk factors, viz. the expected number of classical swine fever epidemics in Germany, Belgium, and the United Kingdom, and the probability that CSFV survives in an empty livestock truck travelling over a distance of 0-900 km. Model calculations based on the situation as at 2003 showed that returning livestock trucks contributed most to the annual P CSFV into the Netherlands with 50%. The most likely sources of CSFV introduction were Germany, Belgium and Spain. Finally, a cost-effectiveness analysis was performed for six measures aimed at preventing CSFV introduction into the Netherlands. Results showed that using separate livestock trucks for national and international transport was most cost-effective, especially when a worst-case scenario was assumed (i.e. for 0.95 percentile values of the annual P CSFV