A risk assessment model for Escherichia coli O157:H7 in ground beef and beef cuts in Canada: Evaluating the effects of interventions

Abstract

A stochastic, quantitative risk assessment model was developed to evaluate the public health risks associated with consumption of ground beef and beef cuts contaminated with Escherichia coli O157:H7 in Canada. The objectives of this work were to evaluate the relative effects of pre-harvest and processing interventions on public health risks using a novel approach, and compare the baseline risks from consumption of ground beef, non-intact beef cuts, and intact beef cuts. Rather than considering efficacy of all interventions at primary production and processing as default values, the model incorporated findings from critical systematic review and meta-analysis of published literature. Public health risks, expressed as average probability of illness per serving, were reduced by 30.9%–72.1%, 44.0%–96.5%, and 95.1%–99.9%, for single pre-harvest interventions, single processing interventions excluding water spray chilling, and combinations of interventions, respectively, relative to a worst-case scenario where no pre-harvest or processing interventions were applied. Combinations of interventions applied at pre-harvest and throughout processing resulted in the greatest relative risk reductions through their effects on both prevalence and concentration of the pathogen in cattle faeces and on cattle carcasses. The use of systematic review methodology to critically assess the results of scientific studies before use of the data in risk modelling enhances the confidence in risk predictions and provides a more evidenced-based model for public health analyses. Analysis of conditions reflective of current practices in Canada indicated that risks from consumption of ground beef were approximately two to three orders of magnitude greater than those for beef cuts, suggesting that risk management measures should focus on the former product to maximize benefits to public health. Risks from consumption of non-intact beef cuts, that is, steaks or roasts that are tenderized, were an order of magnitude greater than those for intact beef cuts. The model provides a useful tool to compare relative efficacies of different intervention strategies to determine their potential impact on public health risks. This tool can be used to evaluate an essentially limitless combination of intervention scenarios and can be adapted to include interventions applied at different points along the farm-to-fork continuum as critically-reviewed data become available.