Research to strengthen the scientific basis for health risk assessment: a survey of the context and rationale for mechanistically based methods and models.

Assessment of health risks is an integral part of regulatory decision-making that occurs at the interface between science (e.g. facts) and policy (e.g. values). Because existing scientific knowledge and understanding are often inadequate to answer the most critical risk-related questions, regulatory agencies have developed sets of formalized 'science policies' to extrapolate from existing data to real-life events and situations. These science policies, as, for example, the use of default assumptions or exposure scenarios, can introduce significant uncertainties into the final risk estimate. We survey the rationale for research to reduce extrapolation-related uncertainties, focusing specifically on the need to develop mechanistically based methods and models, including test methods to identify and characterize health effects, integrated human exposure models, physiologically based pharmacokinetic (PBPK) models and biologically based dose-response (BBDR) models.