LAUSR

Abstract A methodology is presented to optimize a sampling plan for retail products to monitor the prevalence of foodborne pathogens in relation to disease burden. The optimization procedure links an exposure assessment, a quantitative measure for disease burden (i.e. Disability Adjusted Life Years), number of samples to be analyzed (based on uncertainty of the prevalence estimate) and costs for sample analysis. The methodology attributes DALY's on ‘pathogen-matrix level’ to ‘pathogen-food products’ using the exposure assessment. The subsequent procedure includes the number of samples that need to be analyzed per retail product such that the prevalences can be monitored within a preset uncertainty bound, and the costs per sample. The final optimization step sorts pathogen-product combinations using the costs per DALY criterion which results in a monitoring program with a maximum number of DALY's given a certain amount of money. An optimized sampling plan was established for four foodborne pathogens on meat products: Campylobacter in pork and poultry meat, Salmonella in pork, Toxoplasma in pork and Shiga-toxin producing Escherichia coli (STEC) O157 in beef, veal, and mutton/lamb. Results show that Campylobacter on poultry products and one Toxoplasma – pork combination consitute the top 10 in the proposed Dutch public health risk meat monitoring program. This optimized sampling plan monitors 98% of the total amount of DALY's attributed to the considered pathogen-animal species combinations. At the same time, the procedure gives insight in how the preset optimization criteria leads to the proposed set of pathogen-product combinations. An iterative implementation of updated model input (prevalences, DALY estimates) will lead to an up-to-date optimized risk based monitoring program.