With a view to identifying main endocrine disruptors (ED) mixtures to which French consumers are exposed through food, their main diets were modelled using an adapted dimension reduction method. Seven specific diets could be modelled for adults while only one overall diet was considered for children aged 3-17 years. The knowledge of the contamination levels of 78 known or suspected endocrine disrupting compounds in the foods constituting these diets, collected in the frame of the second French Total Diet Study, made it possible to explore the mixtures of EDs to which consumers are exposed. We have thus shown that the ED substances most present in mass concentration are comparable for the whole population, whatever the diet considered. However, a second approach made it possible to highlight, for a given diet, the substances whose exposure is statistically higher than in the diet of the general population. Thus, significantly different ED mixtures could be established for each diet. For example, diets with a high proportion of animal-based foods induce significantly higher exposures to some persistent organic pollutants (e.g., PCDD/F, brominated flame retardants), whereas these exposures are lower for Mediterranean-type diet. On the other hand, the latter, richer in fruits and vegetables, is the one for which pesticides represent a specific signature.These results now pave the way for studying the specific effects of these cocktails of endocrine disruptors, each of which is representative of a type of chronic exposure linked to specific diets.
Keywords: Co-exposure assessment; Dietary patterns; Endocrine disruptors; Mixtures; Sparse and Unique Non-negative Matrix Factorization.
Copyright © 2024 The Authors. Published by Elsevier Ltd.. All rights reserved.