Cumulative risk assessment of pesticide residues in food.

There is increasing need to address the potential risks of combined exposures to multiple residues from pesticides in the diet. The available evidence suggests that the main concern is from dose addition of those compounds that act by the same mode of action. The possibility of synergy needs to be addressed on a case-by-case basis, where there is a biologically plausible hypothesis that it may occur at the levels of residues occurring in the diet. Cumulative risk assessment is a resource-intense activity and hence a tiered approach to both toxicological evaluation and intake estimation is recommended, and the European Food Safety Authority (EFSA) has recently published such a proposal. Where assessments have already been undertaken by some other authority, full advantage should be taken of these, subject of course to considerations of quality and relevance. Inclusion of compounds in a cumulative assessment group (CAG) should be based on defined criteria, which allow for refinement in a tiered approach. These criteria should include chemical structure, mechanism of pesticidal action, target organ and toxic mode of action. A number of methods are available for cumulating toxicity. These are all inter-related, but some are mathematically more complex than others. The most useful methods, in increasing levels of complexity and refinement, are the hazard index, the reference point index, the Relative Potency Factor method and physiologically based toxicokinetic modelling, although this last method would only be considered should a highly refined assessment be necessary. Four possible exposure scenarios are of relevance for cumulative risk assessment, acute and chronic exposure in the context of maximum residue level (MRL)-setting, and in relation to exposures from the actual use patterns, respectively. Each can be addressed either deterministically or probabilistically. Strategies for dealing with residues below the limit of detection, limit of quantification or limit of reporting need to be agreed. A number of probabilistic models are available, but some of there are geographically constrained due to the underlying datasets used in their construction. Guidance on probabilistic modelling needs to be finalised. Cumulative risk assessments have been performed in a number of countries, on organophosphate insecticides alone (USA) or together with carbamates (UK, DK, NL), triazines, chloroacetanilides, carbamates alone (USA), and all pesticides (DE). All identifiable assumptions and uncertainties should be tabulated and evaluated, at least qualitatively. Those likely to have a major impact on the outcome of the assessment should be examined quantitatively. In cumulative risk assessment, it is necessary, as in other risk assessments, for risk managers to consider what level of risk would be considered "acceptable", for example what percentile of the population should be below the reference value. Criteria for prioritising CAGs for cumulative risk assessment include frequency of detection in monitoring programmes, high usage, high exposure relative to the reference value, large number of compounds (e.g. five or more) in a group.