The objective of this study was to evaluate the impact of specific
modeling choices on risk estimates and model complexity in a
units that contains
2:1 cfu. As an illustrative example, consider food
QMRA. Specifically, we evaluated the impact of considering bacte- rial populations as concentration (per unit of volume or weight of food, e.g. cfu/g or log cfu/g) or as absolute numbers of bacteria per food unit (e.g. per batch, package, can or leaf, e.g. cfu/can).
We report the results of a specific comparison among three different approaches to model changes in bacterial prevalence and levels in contaminated food products. One method tracks the bacterial concentrations; a second method models changes in the number of bacteria in contaminated food units, and a third, novel method, evaluates the expected number of bacteria in contami- nated food units. To evaluate the potential impact of the choice in modeling approach on risk estimates, a simulation study was per- formed, evaluating randomly assembled, but plausible, sets of food pathways which included five distinct process steps.