tic studies from EPA’s Superfund program is surprising low. Bioavailability studies at complex hazardous waste sites could be instrumental in designing improved risk management at those sites.
Recently, EPA has evaluated research needs and prioritized research topics (EPA, 1999); bioavailability in human health risk assessment emerged as a high priority. For example, for soils the topic with the highest research priority was “Estimating Human Exposure and Delivered Dose.” This topic included focus points such as “evaluating the bioavailability of contaminants in various soil matrices,” “deriving dermal absorption factors for common soil contaminants,” and “developing biotransfer and bioaccumulation factors for contaminants to facilitate estimates of exposure via the food chain.” Despite this high priority, however, very little in the way of sponsored research on this topic is being funded by the agency. In fact, most of what is known about the potential oral bioavailability of contaminants from soil matrices, for example, comes not from agency-sponsored research projects, but rather from studies conducted by EPA Regions, states, and responsible parties on bioavailability of lead and arsenic from contaminated sites (e.g., EPA, 1996b; Casteel et al., 1997, 2001; Freeman et al., 1992, 1993, 1995; Roberts et al., 2002). These studies offer valuable observations regarding the absorption of contaminants from soils in specific situations, and some inferences on general behavior of absorption from soils might be gained from looking at these studies collectively. However, they are not an effective substitute for directed research because they have a different objective. The purpose of these studies was to obtain empirical measurements of relative bioavailability to support a human health risk assessment. For understandable reasons, this objective does not include an exploration of factors that might influence bioavailability processes, and therefore it is difficult to determine the extent to which these observations can be generalized or used to predict the results that might be obtained at different sites or under different conditions. Unless a greater commitment is made to fund bioavailability process studies from more of a research perspective, progress in developing information that can be utilized to advance human health risk assessments will be slow.
Bioavailability process considerations are not uniformly or widely embraced by scientists, regulators, or the public because of a lack of scientific and technical understanding. Explicit consideration of bioavailability processes and modeling in risk assessment would help to adjust cleanup goals by more accurately identifying that fraction of contaminant total mass that has the potential to enter receptors. Also, bioavailability process understanding would help guide the selection of appropriate remediation technologies. It is clear that more numerous validated tools and models are needed and that there should be reliance on an integrated suite of tools that lead to mechanistic understanding rather than on a single tool or