Breakthrough on assessing exposure of the European population to multiple pesticides A high level of food safety is of the utmost importance for European policy makers and the general public. In this perspective, it needs to be ensured that pesticides residues do not present an unacceptable risk to humans, hereby also taking into account possible effects of mixtures of pesticides. In order to address possible cumulative effects of pesticides residues, the European Commission has funded the ACROPOLIS project. Fourteen countries participated in the project. A tool was developed for food authorities, regulators and the industry to address multiple adverse effects (so-called ‘cocktail effects’) of exposure to groups of pesticides in pesticide risk management. Use of this tool can help to take these effects into account in the regulatory process and future legislation concerning pesticide safety evaluations.
The European Commission has designated this project as a successful cooperation between countries and stakeholders that has led to a partnership between the National Institute for Public Health and the Environment (RIVMNational Institute for Public Health and the Environment ) and the European Food Safety Authority (EFSA). The partnership is directed to making the model accessible to all countries and stakeholders in the European Union that have to assess the safety of pesticides.
The first project of the EFSA-RIVM partnership, aiming to make the Monte Carlo Risk Assessment (MCRA) software scalable for large cumulative assessment groups and extremely large data sets, has been finalised successfully. The software, developed by Wageningen UR department Biometris, now can handle datasets with up to at least one hundred compounds in a cumulative assessment group and at least four million concentration records reported with the EFSA Standard Sample Description format.
Calculations were performed following the EFSA guidance on the use of probabilistic methodology for dietary exposure to pesticide residues, simultaneously. A fake data set was tested and results of all simulations, including uncertainty analyses and resembling the number of European countries, were generated in six hours calculation time. EFSA has started to deliver real data for calculating cumulative exposure to pesticides affecting the nervous system or the thyroid gland.
The calculations are performed on high-performance parallel computing facilities in a protected governmental ICT environment. Four years of Dutch governmental policy to merge computing facilities of governmental institutes has resulted in this ICT environment, which is hosted by RIVM and the Royal Netherlands Meteorological Institute (KNMI).
The MCRA tool includes models for calculating the exposure to pesticides probabilistically. The result of a probabilistic calculation is an exposure distribution, describing the range of exposure levels within a population. The IT tool is also compatible with the European data collection infrastructure.
This tool will now be further tested with the real EFSA data sets for exposure assessments of multiple pesticide residues as grouped by EFSA. If the initial assessments are successful, EFSA hopes to start progressively incorporating tests on cumulative risk assessment into its consumer exposure assessments that are conducted for the EFSA annual report on pesticides residues in food. The experiences gained in these tests will be used for further optimising the tool to ensure its fitness for purpose for taking into account cumulative effects of pesticides for regulatory purposes in future. Until then, there is however still a lot of more work to do, both by risk assessors and risk managers including transparent discussions among all stakeholders.
The project is being closely coordinated with the European Commission, which has established a working group with Member States' experts, EFSA and RIVM to ensure that the project meets the needs of risk managers.
Access to the MCRA tool can be provided after registration on the MCRA website and demonstration material can be provided on request.