Integration of epidemiology with other lines of scientific evidence into pesticide risk assessment

Abstract This chapter will address how epidemiological data can be integrated and weighed with data from experimental animal studies and in vitro/mechanistic studies for a more realistic human risk assessment of pesticides. The chapter will address the following: 1. Individual epidemiological studies. There is a need to make a better use of epidemiological data in human health risk assessment to improve the understanding and characterization of risks from environmental and occupational exposures. 2. Assessment of pooled epidemiological evidence. This section discusses how results from different studies on the association between pesticide exposure and human health can be combined and summarized. The impact of evidence synthesis on risk assessment is particularly highlighted. 3. How to evaluate epidemiological data for risk assessment. The diverse quality of epidemiologic studies can inform risk assessment in different ways. The incorporation of specific elements in future epidemiologic designs would allow a tiered risk assessment as current studies are not specifically designed for this purpose. 4. Other sources of scientific evidence for risk assessment. Classical risk assessment is based on animal regulatory studies, but other sources of evidence are available, such as in vitro or mechanistic studies that could provide biological support to the findings observed in animals or human studies. This would help to establishing causal relationships. 5. Principles for integrating evidence. Information available on human, animal (in vivo), and in vitro studies needs to be integrated and weighed properly following specific criteria accounting for the quality of data, biological plausibility, and concordance/discordance among studies. 6. Approaches used by international regulatory agencies to integrate epidemiological studies for risk assessment (IARC, US-EPA, and EFSA).