Separation of uncertainty and variability in quantitative microbial risk assessment models

Quantitative risk assessment (QRA) modelling is increasingly used in food microbiology as a tool to evaluate health risks and to support the management of safe food production. Depending on the hazard and the process analysed, a QRA model may involve complex calculations: probability distributions are derived for the model parameters and the model is evaluated using specific risk analysis software. Second-order modelling, involving the separation of uncertainty and variability of model parameters, is considered of increasing importance in several fields of risk analysis. However, it is commonly neglected in microbial risk assessment studies. In this paper the relevance of second-order modelling in microbial risk assessment is illustrated by a simple example of a risk assessment of growth of B. cereus in pasteurised milk. It shows that the prediction of the outbreak size may depend on the way that uncertainty and variability are separated, and that a major outbreak may be overlooked if the distinction between uncertainty and variability is neglected.