A voluntary, risk-based control program for paratuberculosis in dairy herds was started in 2006 in Denmark. The program does not include non-dairy herds, and the occurrence of Mycobacterium avium subsp. paratuberculosis (MAP) in these herds is unknown. The objective of this study was to establish the prevalence of MAP infected adult non-dairy cattle in Denmark. Serum samples were collected between October, 2008 and January, 2009 from every 6th animal over 24 months of age, sent to slaughter to Danish slaughterhouses from non-dairy herds. The final sample included 2345 cattle of 13 different breeds, with the largest breed group being crossbreds (of unknown breeds) (30%) and three dairy breeds (Danish Holstein, Danish Jersey and Danish Red Cattle) comprising 27% of the samples. The serum samples were tested using a MAP specific antibody ELISA (IDScreen) and positives were defined as the sample-to-positive ratio greater than 0.60. Estimation of the breed-specific apparent prevalences, true prevalences (TP), and true prevalences with a random effect of breed was done in a Bayesian analysis. Information about test sensitivity and specificity were based on literature data and expert information. Regardless of the method of analysis, the estimated prevalences showed similar differences between breeds. The dairy breeds Danish Jersey, Danish Holstein and Danish Red Cattle were ranked highest (i.e. with highest prevalence) (TP medians: 13, 10, and 6, respectively). Combined, the dairy breeds had a significantly higher prevalence than the other breeds, median TP (dairy)=15.7% vs. median TP (non-dairy)=0.8%. For the individual non-dairy breeds, the median estimates were generally higher, illustrating the problems of ranking groups based on relatively small sample sizes.
Copyright 2010 Elsevier B.V. All rights reserved.