There is increasing evidence that the European wild rabbit (Oryctolagus cuniculus) is a wildlife reservoir for paratuberculosis and infected populations may contribute to the persistence of infection in livestock. The aim of this study was to test the hypothesis that farms with difficulties controlling paratuberculosis in their cattle herds have a higher prevalence of Mycobacterium avium subsp. paratuberculosis (MAP) infection in their rabbit populations. A total of 281 rabbits from 13 beef farms in the East of Scotland were randomly sampled in early spring 2007. Participating farms were in paratuberculosis control programmes under the Premium Cattle Health Scheme (PCHS), and were classified as 'responder' (paratuberculosis under control) or 'low responder' (a persistent number of paratuberculosis-positive cattle despite control measures in place) farms. Of the rabbits sampled, 23.8% tested positive for MAP, with those on 'low responder' farms having a greater probability of being infected (0.4) relative to rabbits on 'responder' farms (0.1). The association suggests that MAP-infected rabbits may contribute to the persistence of paratuberculosis in domestic livestock and undermine control strategies that focus on livestock alone. This study provides the first evidence of an association between the persistence of paratuberculosis in livestock despite the implementation of disease control strategies, and MAP-infected sympatric wild rabbit populations.