Determining Health Effects of Pollutants 87Risk assessment assumes that we can extrapolate
(work backward) from the huge doses of chemicals and
the high rates of cancer they cause in rats to determine
the expected rates of cancer in humans exposed to lower
amounts of the same chemicals. However, even if we are
reasonably sure that exposure to high doses of a chemi-
cal causes the same effects for the same reasons in both
rats and humans, we cannot assume that these same
mechanisms work at low doses in humans. The body me-
tabolizes small and large doses of a chemical in different
ways. For example, enzymes in the liver may break down
carcinogens in small quantities, but an excessive amount
of carcinogen might overwhelm the liver enzymes.
In short, extrapolating from one species to another and
from high doses to low doses is uncertain and may over-
estimate or underestimate a toxicant’s danger. However,
animal carcinogen studies provide valuable information: A
toxicant that does not cause cancer in laboratory animals at
high doses is not likely to cause cancer in humans at levels
found in the environment or in occupational settings.
Scientists do not currently have a reliable way to de-
termine whether exposure to small amounts of a sub-
stance causes cancer in humans. However, the EPA is
planning to change how toxic chemicals are evaluated
and regulated. Toxicologists are developing methods to
provide direct evidence of the risk involved in exposure
to low doses of cancer-causing chemicals (Figure 4.13).
Epidemiological evidence, including studies of
human groups accidentally exposed to high levels of
suspected carcinogens, is also used to determine whether
chemicals are carcinogenic. For example, in 1989 epide-
miologists in Germany established a direct link between
cancer and a group of persistent organic pollutants called
dioxins (see Table 4.2). They observed the incidence of
cancer in workers exposed to high concentrations of di-
oxins during an accident at a chemical plant in 1953 and
found unexpectedly high levels of cancers in their diges-
tive and respiratory tracts.
Risk Assessment of Chemical Mixtures
Humans are frequently exposed to various combinations
of chemical compounds, in the air we breathe, the food
we eat, and the water we drink. For example, cigarette
smoke contains a mixture of chemicals, as does automo-
bile exhaust. Cigarette smoke is a mixture of air pollut-
ants that includes hydrocarbons, carbon dioxide, carbon
monoxide, particulate matter, nicotine, cyanide, and a
Courtesy Centers for Disease ControlMeasuring low doses of a toxicant (dioxin) in
Õ>ÊL`ÊÃiÀÕÊUÊ}ÕÀiÊ{°£ÎÊÊ
Scientists are developing increasingly sophisticated
methods of biomonitoring to analyze human tissues and
fluids. Photographed at the Centers for Disease Control and
Prevention’s Environmental Health Laboratory.small amount of radioactive materials that come from
the fertilizer used to grow the tobacco plants.
The vast majority of toxicology studies have been
performed on single chemicals rather than chemical
mixtures, and for good reason. Mixtures of chemicals
interact in a variety of ways, increasing the level of com-
plexity in risk assessment, a field already complicated by
many uncertainties. Moreover, there are simply too many
chemical mixtures to evaluate.
Chemical mixtures interact by additivity, synergy, or an-
tagonism. When a chemical mixture is additive, the effect is
exactly what you would expect, given the individual effects
of each component of the mixture. If a chemical with a
toxicity level of 1 is mixed with a different chemical, also
with a toxicity level of 1, the combined effect of exposure
to the mixture is 2. A chemical mixture that is synergistic has
a greater combined effect than expected; two chemicals,