EVOLUTION OF SCIENCE-BASED UNCERTAINTY FACTORS IN NONCANCER RISK ASSESSMENT

''Safe'' doses estimated with standard methods by agencies around the world, whether ADIs, ECNCs, MRLs, TDCs, TCs, TIs, RfCs, or RfDs, shoul d be considered as accurate-but imprecise-estimations of doses or conc entrations believed to be without risk to populations of humans (inclu ding sensitive subgroups). Restated, these ''safe'' doses are thought to be below population thresholds for adverse effect, but the degree t o which they underestimate the population threshold is generally not k nown. Part of this imprecision comes from the use of 10-fold default u ncertainty factors. We show research and case studies drawn from a lar ge sample of U.S. EPA and Health Canada risk values where uncertainty factors other than a default value of 10-fold were used in the estimat ion of a RfD, RfC, TDI or TC. Percentages for the use of these ''data- derived'' factors vary between 3.6% and 47%. In five case studies, we explicitly review the types of data that have been used to support a c hange in the default value, why the data support a different UF, and w hat assumptions have been satisfied, replaced, or how the uncertainty was reduced.