Quantitative risk assessment of Salmonella Enteritidis in shell eggs in Europe

QMRA, Bayesian Model, Monte Carlo Markov Chain, laying hens, Salmonella Enteritidis in eggs, sensitivity, sensitivity analysis
First published in the EFSA Journal
20. April 2010
19. April 2010
Scientific Report


Prevalences in terms of the number of eggs contaminated (internally and externally) with Salmonella Enteritidis per million are presented based on a 2 stage Bayesian model. The first stage estimates the average flock prevalence over laying period in the production system and the second stage estimates the proportion of contaminated eggs for an infected flock. These are combined to give the expected number of infected eggs per million using the hen data from 2 European Union Member States (MSs). Interpretation and extrapolation of the results has to be done with caution as they are sensitive to the egg data used which are most likely biased due to selection effects. Working with the lower and upper limits of the 95% credible intervals for the parameter estimates is strongly recommended.

It is recommended to repeat a similar exercise with the current model with data from more EU MSs. These should be collected in the format received from the 2 MSs analysed in this report. Further, it is important to have hen and egg data from the same flocks (with flock sizes) for at least some Member States.

European Food Safety Authority
didier.verloo [at] efsa.europa.eu
EFSA Journal 2010; 8(4):1588
Question Number
On request from