Page 164 - Materials Chemistry, Second Edition

L1644_C04.fm Page 136 Tuesday, October 21, 2003 3:13 PM

RfDs, RfCs, MRLs and ADIs are very often construed as rigid threshold limits,
above which toxicity is likely to occur. The truth, however, is that these values
actually represent levels of a potential toxicant that are highly unlikely to represent
any threat to human health over a particular or specified duration of daily exposures.
The more frequently these levels are exceeded and the greater the excess, the more
likely that some toxic manifestation will occur. Most definitely these guidance or
reference values are not threshold values for the onset of toxicology in any exposed
population. Health guidance values must be considered in the context of their
intended role as mere screening or trigger values, as which they serve as tools for
assisting in the determination of whether further evaluation of a given potential
exposure scenario is warranted.
For a general environmental health risk assessment of an unspecified emitted
pollutant it is necessary to divide the population into different groups depending
on their sensibility, for instance, babies, children, adults, adults above 65 years
of age, people with asthma, etc. A division is especially important for the
assessment of noncarcinogenic pollutants with the effect of chronic illness. In
the case of carcinogenic pollutants, it is generally sufficient to divide the popu-
lation into two groups of adults and children (until a certain age) in order to
consider the different physical conditions (e.g., breathing, surface of the body,
etc.) and the different life-styles (e.g., children playing outside). For substances
that induce a carcinogenic response, it is always conservatively assumed that
exposure to any amount of the carcinogen will create some likelihood of cancer;
that is, a plot of response vs. dose is required to go through the origin. Therefore,
for noncarcinogenic responses, it is usually assumed that a threshold dose exists,
below which no response will occur. As a result of these two assumptions, the
dose–response curves and the methods used to apply them are quite different for
carcinogenic and noncarcinogenic effects, as suggested in Figure 4.4 (Masters,
1991). Realize that the same chemical may be capable of causing both kinds of
responses.

4.6.1.1 Toxicological Information: Carcinogenic Effect

A controversial point in the discussion of carcinogenic effects is the estimation of
the dose–response functions for different pollutants. Animal tests administer various
doses to observe toxic effect; however, because the doses in these tests are higher
than those existing in the environment, the discussion deals with the possibility of
an extrapolation from the higher concentration in the animal test to the lower
concentration in the environment and the possible effects. It is necessary to take into
account that, even with extremely large numbers of animals in a bioassay, the lowest
risks that can be measured are usually to a small percentage. Because regulators
attempt to control human risk to several orders of magnitude below that, no actual
animal data will be anywhere near the range of highest interest.
Different mathematical models to extrapolate to the lowest level exist, but there
is still a lot of uncertainty. With the main aim of protecting human health, the USEPA
chooses the safer way to estimate the risk with an additional correction factor, which
means an overestimation of risk (Olsen et al., 2000).

© 2004 CRC Press LLC