The sensitivity of criteria in detecting unsafe food was studied for a hypothetical ready-to-eat food supporting growth of Listeria monocytogenes during storage. The effects of the prevalence of the pathogen and of the duration of the shelf-life on the probability for servings to be unsafe and on the probabilities of detecting unsafe food were evaluated. The results obtained in this example show that the probability for servings to be unsafe increases with the prevalence and the shelf-life while the probabilities of detecting unsafe food depend above all on the prevalence of the pathogen. The current criteria used for L. monocytogenes can then sometimes be relatively ineffective to prevent the consumption of unsafe food with regard to the shelf-life of foods. The exposure assessment approach described in this study could be used to establish shelf-lives consistent with a defined acceptable risk.