BETTER DOSE-RESPONSE CHARACTERIZATION

It is possible to better reflect the observed data at the lower doses and to not force overstatements of the low-dose risks.  It is possible to provide a better dose-response characterization than that provided by extrapolating non-threshold dose-response models like the restricted multistage model to low doses or that provided by the linearized multistage models upper-bound potency measures like q₁*.

Figure 12 illustrates the ability of classical measures of toxicity like the no-observed-adverse-effect-level (NOAEL) to reflect dose levels that are likely to be without appreciable risks of deleterious effects.  These classical measures are based on the idea of using the observed data to determine a dose level at which there is no detectable increase in the occurrence of adverse health effects (no increase compared to the probability of response at dose zero).  These measures have been the basis for "allowable daily intake" (U.S. FDA), the reference dose and reference concentration (U.S. EPA), the minimum risk level (U.S. ATSDR), and the tolerable intake (WHO).  For example, the reference dose is described by EPA as ...an estimate (with uncertainty spanning perhaps an order of magnitude) of a daily exposure to the human population (including sensitive subgroups) that is likely to be without appreciable risk of deleterious effects during a lifetime.

This classical approach to toxicity assessment and dose-response characterization has been extended to both cancer and noncancer risk assessment and is the motivational basis for benchmark doses and the margin-of-exposure characterizations.

>> Figure 12