Estimation of True Salmonella Prevalence Jointly in Cattle Herd and Animal Populations Using Bayesian Hierarchical Modeling

Abstract

The Finnish salmonella control program (FSCP) for beef production is based on both randomized and selective testing of herds and animals. Sampling of individual animals in abattoirs is randomized. Herds are selectively tested for salmonella on the basis of clinical symptoms and/or other factors. Risk assessment of FSCP is inherently difficult due to the complexity of the complete data set, especially if the detailed labeling of the testing types is lost. However, for a risk assessment of the whole production chain, methods for exploiting all available data should be considered. For this purpose, a hierarchical Bayesian model of true salmonella prevalence was constructed to combine information at different levels of aggregation: herds in geographical regions and individual animals arriving for slaughter. The conditional (municipality specific) probability of selection of a herd for testing was modeled given the underlying true infection status of the herd and information about the general sampling activity in the specific region. The model also accounted for the overall sensitivity of the sampling methods, both at the herd and at the animal level. In 1999, the 95% posterior probability intervals of true salmonella prevalence in the herd population, in individual cattle, and in slaughter animal populations were [0.54%, 1.4%] (mode 0.8%), [0.15%, 0.39%] (mode 0.2%), and [0.12%, 0.36%] (mode 0.2%), respectively. The results will be further exploited in other risk assessments focusing on the subsequent parts of the beef production chain, and in evaluation of the FSCP.

Antimicrobial Drug Resistance and Molecular Characterization ofSalmonellaIsolated from Domestic Animals, Humans, and Meat Products

“…A similar Salmonella prevalence in feedlot cattle had been reported before and been attributed to low hygiene in feedlots (Vanselow et al 2007;Khaitsa et al 2007a). Also, previous reports of Salmonella prevalence in range cattle (Ranta et al 2005) and dairy cattle (Sorensen et al 2003;Huston et al 2002) have been comparable to what is reported by this study, and have been consistently lower than in feedlot cattle. However, the isolation of Salmonella in sick or dead cattle (13.6%) and sick humans (41.2%) was indicative of its increasing role in causing disease in both groups of hosts (Besser et al 2000;Padungtod and Kaneene 2006).…”

Alternative Methods for Computing the Sensitivity of Complex Surveillance Systems

“…For surveillance systems where systematically collected data are available, the BBN formulation is a bridge to a fully Bayesian analysis using Markov chain Monte Carlo methods to obtain posterior estimates of prevalence based on prior distributions of model parameters. ( 22–24 ) The dependence on prior distributions makes such methods inherently controversial ( 25 ) but the scenario tree and BBN approaches have a similar credibility problem when there is uncertainty in detection probabilities and other parameters. In both approaches, many sources of uncertainty—e.g., in parameter values, bounds, and model structure—are ignored for the sake of exposition and mathematical tractability.…”

Bayesian Methods in Veterinary Clinical Epidemiology

“…Bayesian inference is increasingly used in many research disciplines, including veterinary epidemiology (Dohoo et al, 2009). A few examples in this field are applications for diagnostic test evaluation, disease mapping, and risk assessment (e.g., Toft et al, 2005;Best et al, 2005;Ranta et al, 2005). The growing popularity of Bayesian analysis is often motivated by its advantage in handling complicated empirical problems which are difficult within the classical frequentist approach (Dohoo et al, 2009).…”