2.8. Screening of Exchanged Animals

Screening of exchanged animals is applied when the time between the departure of the selling herd and the arrival of an animal at the buyer’s herd exceeds six days, as well as in herds classified as “at risk”. Exchanges that exceeded six days were not taken into account in our model due to their rarity (personal communication from the national TB coordinator), the lack of data about them, and the very low probability that they could lead to the detection of infection in a herd. Reliable centralized data were not available on the proportion of animals sold from “at risk” herds that are actually screened. Therefore, we assumed that all animals sold from herds classified as “at risk” (with the exception of animals for fattening) were screened. For this herds the proportion of animals sold to another herd was estimated for each herd type of each area thanks to the BDNI data base using 2016 data. The probability of an animal being sold during the year was therefore modeled according to the area and type of herd, using an asymmetrical normal law designed to correspond as closely as possible to the distribution observed in the sample corresponding to the type of herd and the area concerned.

The probability for each tested animal to be found TB-infected was modeled the same way than for the compliant quick-path protocol of periodic screening (accounting for ICCT sensitivity considering veterinarians’ practices and laboratory analyses results of all animals with non-negative result with the ICCT (histology, PCR, and bacteriology)) [4,6].