Cost-effectiveness of surveillance and biosecurity scenarios for preventing CSF in Switzerland

Classical Swine Fever (CSF) is a porcine viral disease that has severe consequences on animal health, welfare, and production. In Switzerland six broad levels of on-farm intervention measures are considered depending on the assessment of the epidemiological status of the focal farm, of the neighbouring farms and of the country; Basic biosecurity (I0), Basic biosecurity with reinforced surveillance (I0b), Surveillance zone (I1), Protection Zone (I2), Quarantine (I3) and Culling (I4). In 2014 an exclusion component was added to allow testing for CSF without implementing quarantine or movement restrictions. The objective of this study was to assess the cost-effectiveness of this current system. A stochastic farm-based model was developed which simulates the spread of CSF through Swiss pig farms in space and time, taking account of the implementation of intervention measures based on the system of surveillance. Two measures of effectiveness of the current system were assessed 1) the number of infected farms per outbreak and 2) the probability of implementing an intervention strategy (I0-I4) that is not strong enough given the true epidemiological status. Costs of the outbreak included the cost of implementing interventions and the costs of culling all animals determined to be infected over the course of the outbreak The median number of infected farms per outbreak was 80. Among all the simulations, 1.4% (14/1000 simulations) did not lead to an outbreak. The median probabilities of errors, over all simulations, for the different possible intervention measures to be implemented at the time when a farm becomes infectious were highest for basic biosecurity level with reinforced surveillance (I0b), then surveillance zone measures (I1), basic biosecurity level (I0), and finally protection zone measures (I2) with the lowest probability. The median costs endorsed at federal level were 3.4 million CHF per outbreak. The cost-effectiveness of the system was evaluated as medium. The exclusion component was evaluated as good, balancing its low price of implementation, the level of biosecurity implemented at farms and the risk of CSF incursion in Switzerland. We suggest that changes in the surveillance strategy that would increase the probability of reinforced surveillance (I0b, I1, I2) to be implemented at the time a farm becomes infectious would increase the cost-effectiveness of the system. Such changes could include increasing awareness among stakeholders or developing new actions after the first outbreak, such as more intense investigations among connected farms