Estimation of the individual slaughterhouse surveillance sensitivity for bovine tuberculosis in Catalonia (North-Eastern Spain)

- Estimation of bTB slaughterhouse surveillance sensitivity in Catalonia (Spain).
- Probability that bTB-infected animal presents compatible visible lesions (CVL).
- Probability that CVL were detected by the routine meat inspection.
- Probability that a veterinary officer suspects of bTB and takes sample.
- High heterogeneity of the bTB surveillance sensitivity between slaughterhouses.

The achievement of the Officially Tuberculosis Free (OTF) status in regions with low bovine Tuberculosis (bTB) herd prevalence, as is the case of North-Eastern Spain (Catalonia), might be a likely option in the medium term. In this context, risk-based approaches could be an alternative surveillance strategy to the costly current strategy. However, before any change in the system may be contemplated, a reliable estimate of the sensitivity of the different surveillance components is needed. In this study, we focused on the slaughterhouse component. The probability of detection of a bTB-infected cattle by the slaughterhouses in Catalonia was estimated as the product of three consecutive probabilities: (P1) the probability that a bTB-infected animal arrived at the slaughterhouse presenting Macroscopically Detectable Lesions (MDL); (P2) the probability that MDL were detected by the routine meat inspection process and (P3) the probability that the veterinary officer suspected bTB and sent the sample for laboratory confirmation.The first probability was obtained from data collected through the bTB eradication program carried out in Catalonia between 2005 and 2008, while the last two were obtained through the expert opinion of the veterinary officers working at the slaughterhouses who fulfilled a questionnaire administered during 2014.The bTB surveillance sensitivity of the different cattle slaughterhouses in Catalonia obtained in this study was 31.4% (CI 95%: 28.6-36.2), and there were important differences among them. The low bTB surveillance sensitivity was mainly related with the low probability that a bTB-infected animal arrived at the slaughterhouse presenting MDL (around 44.8%). The variability of the sensitivity among the different slaughterhouses could be explained by significant associations between some variables included in the survey and P2. For instance, factors like attendance to training courses, number of meat technicians and speed of the slaughter chain were significantly related with the probabilities that a MDL was detected by the meat inspection procedure carried out in the slaughterhouse. Technical and policy efforts should be focused on the improvement of these factors in order to maximize the slaughterhouse sensitivity.