A decision-support system for Salmonella in broiler-chicken flocks

Abstract We built a decision-support system to assess the risk of contamination of chicken-broiler flocks by Salmonella at the end of the rearing period. This system was developed from the survey data from 85 chicken-broiler flocks located in western France. First, we estimated the probability of contamination of the house by Salmonella before placement of day-old chicks via a cleansing inspection using a visual-inspection grid, a decontamination evaluation using count-plates, and risk factors for Salmonella persistence in the barn after cleansing and disinfection. Second, we estimated (using a logistic model) the probability of prevalent contamination of the flock by Salmonella at the end of the rearing period. Validation was carried out on 60 flocks selected from seven production companies in western France. The risk estimated by the model was compared to the Salmonella status of the flock (gold standard) assessed by samples taken from the environment of the broilers and analysed with classical bacteriological methods. The sensitivity was 97.8% and the specificity 64.3%.