Risk assessment of benzo[a]pyrene in heat-processed meat in Denmark: A probabilistic approach

Stylianos Georgiadis (Guest lecturer), Jakobsen, L. S. (Guest lecturer), Nielsen, B. F. (Guest lecturer), Maarten Nauta (Guest lecturer), Pires, S. M. (Guest lecturer)

Activity: Talks and presentationsConference presentations


Heat-processed meat may contain high concentrations of polycyclic aromatic hydrocarbons, a class of chemical components including benzo[a]pyrene (BaP) that is classified as group 1 human carcinogen by IARC. To minimize risk due to dietary exposure to these contaminants, food and health authorities may advise the population to limit certain preparation practices such as barbecuing. Assessment of the health risk given specific food preparation practices may direct food-safety strategies to those consumer-groups that are at higher risk. Our objective was to estimate the extra lifetime risk of cancer due to exposure to BaP through barbecued meat in Denmark, both for the overall population and for subgroups. The probabilistic risk model consists of two parts; exposure assessment and cancer risk estimation. The first part focuses on modelling human exposure to BaP, taking into account the variability among individuals and between population subgroups. Consumption of various meat types and concentration of BaP in barbecued meat types are described by probability distributions. In the second part, human exposure is converted to an equivalent animal exposure to allow the usage of an animal dose-response model. The variability in individuals' sensitivity to BaP and uncertainty in interspecies differences are quantified. The extra lifetime risk of cancer due to BaP is calculated through a dose-response model whose parameters are uncertain. To derive population-level results, the frequency of consumption of barbecued meat in Denmark is assessed, while for population subgroups, model outputs are presented as a function of the number of barbecued meat servings. An event-based Monte Carlo simulation scheme is built to perform the calculations and separate uncertainty and variability. The findings of this study highlight that uncertainties induced in the cancer risk assessment module overshadow any variations within population or between population subgroups.
Period4 Dec 2018
Event title2018 Society for Risk Analysis (SRA) Annual Meeting: The Many Faces of Risk
Event typeConference
LocationNew Orleans, United States, Louisiana
Degree of RecognitionInternational


  • Probabilistic risk assessment
  • Exposure modelling
  • Uncertainty
  • Variability
  • Monte Carlo simulation