How can exposure assessment for pesticides in epidemiological studies be improved? Insights from the IMPRESS project.

The IMPRoving Exposure aSSessment Methodologies for Epidemiological Studies on Pesticides (IMPRESS) project (http://www.impress-project.org/) aimed to further the understanding of the performance of pesticide exposure assessment methods (EAMs). To achieve this the IMPRESS project used two approaches to assess EAM performance, using existing and newly collected data from five studies from three different countries and use of published secondary data to undertake three meta-analyses for selected chronic health outcomes. Based on the findings of the IMPRESS project we provide in this paper insights on the overarching research question "How can exposure assessments for pesticides in epidemiological studies be improved"? Exposure assessment is a critical component of pesticide epidemiological studies. EAMs used and epidemiological practices employed need to reflect the changing nature and complexities of pesticide exposure in various occupational settings. To properly assess the association between exposure and selected health outcomes, the choice of EAM should provide a clear exposure contrast within the study population. Acquiring a practical understanding of the pesticide use practices is crucial to determine whether factors such as frequency or intensity of exposure have to be considered in planned analyses. Biomonitoring may be more beneficially applied intensively in a focussed exposure assessment analysis of a particular cohort, which can be used to determine the most relevant exposure factors within that cohort-specific context. Overall, improving pesticide exposure assessment in epidemiological studies requires a multi-disciplinary approach. A next step for the wider scientific community may be to consider the development of a decision tree to aid the selection of suitable EAMs. Such a decision tree would need to consider and be based on multiple parameters including, but not limited to, study type, health endpoint, socio-demographic context, farming system, pesticide used, and application methods.