Modelling has been used to extrapolate outbreak and experimental virus transmission data to predict vaccine-based control in the field. This predicts that if vaccination is optimised and clinical surveillance effectively removes herds with diseased animals, then the number of undisclosed infected herds and animals should be small with few carriers [43], [44] and [45]. Undetected infected

animals would be found mainly in non-vaccinated sheep VX-809 mouse herds and vaccinated cattle and sheep herds. However, after serosurveillance, carried out according to the EU Directive, vaccination and pre-emptive culling strategies yielded comparable low numbers of undetected infected BGB324 manufacturer animals [45]. Schley et al. emphasised that following effective vaccination, the quality of inspection is the principal factor influencing whether or not undisclosed carrier herds occur, supporting the importance of other control

measures [44]. Further studies are required to model virus persistence in vaccinated populations through transmission from acutely infected animals, rather than from carrier animals, as the former represent a more significant risk for new FMD outbreaks [12]. NSP serosurveillance of a large number of animals will give rise to many false positive test reactors, since the tests have imperfect specificity (Sp of 98–99.7% for cattle; [41]) and Se/Sp limitations cannot be overcome easily by using a combination of different NSP tests [46]. Furthermore, true positive test results cannot be distinguished readily from false positive ones [47], although a cluster analysis [48] and the use of likelihood ratios to weight the strength of seroconversion might improve the possible discrimination [49]. This makes classification of the infection status of large herds difficult. Arnold et al. concluded that in this situation, the best compromise between maximising the sensitivity for carrier detection, whilst minimising unnecessary culling,

will be met by adopting an individual-based testing regime in which all animals in all vaccinated herds are tested and positive animals rather than herds are culled unless [43]. The remaining risk with this approach is that any carriers that are missed will be free to move to unvaccinated herds on national territory once outbreak restrictions are lifted and those non-vaccinated animals may be traded. Requirements for recovering the FMD-free status where vaccination is not practised are laid out in the OIE Terrestrial Animal Health Code (Supplementary Table 1; [19]) and for EU Member States in the EU FMD Directive [9]. With stamping out (culling) of affected herds and suitable surveillance, the FMD-free status can be regained 3 months after the last case.