Agricultural pesticide exposure data bases for risk assessment.

For pesticide registration, data on toxicity and on occupational exposure are required. In the smaller countries, such as The Netherlands, only a small number of studies on exposure will be available for a proper risk assessment. Therefore, the primary data have to be obtained mainly from the literature. In this review the exposure data available in the formally published literature and from a few Dutch reports for mixing, loading, and application of pesticides are considered for the development of generic/surrogate data bases with which for a specific case (e.g., a new pesticide) the exposure can be estimated for use in a first step of risk assessment. The general aspects of the determination of exposure to pesticides under field conditions are considered, as well as the published proposals for generic data bases. It is concluded that the development of data bases is possible in principle, although in many cases, depending on the type of technique for mixing/loading and application, not enough data are available. Data bases on re-entry have not been proposed in the literature, although many studies on exposure after reentry have been reported. In view of the fact that in some cases the mixer/loader is not the same person as the applicator and considering the different hygienic approach that may be needed for diminishing exposure during these types of work, data on exposure during mixing and loading and on exposure during application were considered separately whenever possible or reasonable. Since data on exposure may vary considerably in daily agricultural practice and since the data sets are in most cases relatively small and not homogeneous, it is impossible to calculate statistically valid means or percentiles. Therefore, estimates have been made which are called indicative values. For risk assessment such estimates, to be used in a first approximation without further specific knowledge on chemical properties of the pesticide, need to be high percentiles, since all 'normal' situations have to be considered and a high degree of worker protection should be achieved. From the data, indicative 90th percentiles of exposure have been estimated and expressed as amount of formulated product per hour of mixing and loading and expressed in amount of spray liquid per hour application. These 90th percentiles are presented in Table 9. As can be seen from Table 9, for various application techniques no data on exposure were available in the open literature, or the number of published data was considered insufficient to obtain indicative surrogate levels that can be used in registration procedures.(ABSTRACT TRUNCATED AT 400 WORDS)