ABSTRACT
In this paper we analyze the impact of the sensitivity and specificity of a Mycobacterium avium (Ma) test on pig producer incentives to control Ma in finishing pigs. A possible Ma control system which includes a serodiagnostic test and a penalty on finishing pigs in herds detected with Ma infection was modelled. Using a dynamic optimization model and a grid search of deliveries of herds from pig producers to slaughterhouse, optimal control measures for pig producers and optimal penalty values for deliveries with increased Ma risk were identified for different sensitivity and specificity values. Results showed that higher sensitivity and lower specificity induced use of more intense control measures and resulted in higher pig producer costs and lower Ma seroprevalence. The minimal penalty value needed to comply with a threshold for Ma seroprevalence in finishing pigs at slaughter was lower at higher sensitivity and lower specificity. With imperfect specificity a larger sample size decreased pig producer incentives to control Ma seroprevalence, because the higher number of false positives resulted in an increased probability of rejecting a batch of finishing pigs irrespective of whether the pig producer applied control measures. We conclude that test sensitivity and specificity must be considered in incentive system design to induce pig producers to control Ma in finishing pigs with minimum negative effects.
Subject(s)
Animal Husbandry , Serologic Tests/methods , Swine Diseases/epidemiology , Swine Diseases/prevention & control , Tuberculosis/veterinary , Abattoirs , Animal Husbandry/economics , Animal Husbandry/methods , Animals , Models, Biological , Mycobacterium avium/isolation & purification , Prevalence , Sensitivity and Specificity , Seroepidemiologic Studies , Serologic Tests/veterinary , Swine , Swine Diseases/microbiology , Tuberculosis/epidemiology , Tuberculosis/microbiology , Tuberculosis/prevention & controlABSTRACT
The EU prescribes that food business operators must use food chain information to assist in food safety control. This study analyses usefulness of food chain information about antibiotic usage covering the 60-day period prior to delivery of pigs to slaughter in the control of antibiotic residues in pork. A dataset with 479 test results for antibiotic residues in tissue samples of finishing pigs delivered to a Dutch slaughter company was linked to information provided by pig producers about antibiotic usage in these finishing pigs. Results show that twice as many producers reported using antibiotics in the group of 82 producers with antibiotic residues (11.0%) compared to the group without antibiotic residues (5.5%) (p=0.0686). For 89% of consignments with a finishing pig with antibiotic residues, the producer reported 'did not use antibiotics'. Food chain information about antibiotic usage provided by Dutch pig producers was no guarantee for absence of antibiotic residues in delivered finishing pigs.
