Development and validation of a probabilistic second-order exposure assessment model for Escherichia coli O157:H7 contamination of beef trimmings from Irish meat plants

A second-order quantitative Monte Carlo simulation model was developed for Escherichia coli O157:117 contamination of beef trimmings in Irish abattoirs. The assessment considers initial contamination levels, cross-contamination and decontamination events during the cattle slaughter process. The mean simulated prevalence of E coli O157:H7 on trimmings was 2.36% and the mean simulated counts of E coli 01 57:H7 on contaminated trimmings was -2.69 log(10) CFU/g. A parallel validation survey provided some confidence in the model predictions. An uncertainty analysis indicated that microbial test sensitivity is a significant factor contributing to model uncertainty and requires further investigation while also indicating that risk reduction measures should be directed towards reducing the hide to carcass transfer (correlation coefficient 0.25) during dehiding and reducing the initial prevalence and counts on bovine hides (correlation coefficients 0.19 and 0.16, respectively). A characterisation of uncertainty and variability indicating that further research is required to reduce parameter uncertainty and to achieve better understanding of microbial transfer in meat plants. The model developed in this study highlights the need for further development of quantitative risk assessments in the food industry. (C) 2007 Elsevier Ltd. All rights reserved.