A model to establish a "Performance Objective" (PO) for Campylobacter spp. in broiler carcasses at retail

This contribution presents an application of the "Food Safety Objective" (FSO) framework proposed by the ICMSF. A Performance Objective (PO) for Campylobacter spp. in broiler carcasses was calculated based on a hypothesized FSO of) log Campylobacter per serving and an the assumption that cross-contamination during food preparation is the most relevant infection route. Cross-contamination scenarios were taken from Luber et al. (2006) to model the transfer from raw poultry meat to ready-to-eat foods. By Monte-Carlo simulation techniques, it was estimated that in 95% of the cases a typical combination of two transfers (e.g. chicken to hand; hand to ready-to-eat food) result in less than 1% or -2 log) of the bacteria being transferred to ready-to-eat food. When several pathways are considered in parallel, the transfer would be ca. -1.8 og, or -0.8 log when a hand to mouth infection pathway is included. Assuming that there is no in- or decrease in Campylobacter concentrations during retail and storage, the PO would equate to 3.8 (or 2.8, if a hand to mouth infection pathway is included) log cfu/carcass or meat cut. This PO is compared to data obtained from an Austrian poultry abbatoir. Limitations and bent-fits of establishing PO's even in absence of an Appropriate Level Of Protection (ALOP) are discussed.