Environmental Engineering Reference
In-Depth Information
example,
case reports,
usually made by physicians, pro-
vide information about people suffering some adverse
health effect or death after exposure to a chemical.
Such information often involves accidental poisonings,
drug overdoses, homicides, or suicide attempts.
Most case reports are not reliable sources for es-
timating toxicity because the actual dosage and the
exposed person's health status are often not known.
Nevertheless, they can provide clues about environ-
mental hazards and suggest the need for laboratory
investigations.
Epidemiological studies
compare the health of peo-
ple exposed to a particular chemical (the
experimental
group
) with the health of a similar group of people not
exposed to the agent (the
control group
). The goal is to
determine whether the statistical association between
exposure to a toxic chemical and a health problem is
strong, moderate, weak, or undetectable.
Three factors can limit the usefulness of epidemio-
logical studies.
First,
in many cases, too few people
have been exposed to high enough levels of a toxic
agent to detect statistically significant differences.
Sec-
ond,
conclusively linking an observed effect with expo-
sure to a particular chemical is difficult because people
are exposed to many different toxic agents throughout
their lives and can vary in their sensitivity to such
chemicals.
Third,
we cannot use epidemiological stud-
ies to evaluate hazards from new technologies or
chemicals to which people have not yet been exposed.
that some animal testing is needed because the alter-
native methods cannot adequately mimic the complex
biochemical interactions taking place in a live animal.
Critical thinking: should animal testing be banned?
Acute toxicity tests are run to develop a dose-
response curve, which shows the effects of various
dosages of a toxic agent on a group of test organisms.
In
controlled experiments,
the effects of the chemical on a
test group
are compared with the responses of a
control
group
of organisms not exposed to the chemical. Care
is taken that organisms in each group are as identical
as possible in terms of age, health status, and genetic
makeup, and that all are exposed to the same environ-
mental conditions.
Fairly high dosages are used to reduce the number
of test animals needed, obtain results quickly, and
lower costs. Otherwise, tests would have to be run on
millions of laboratory animals for many years, and
manufacturers could not afford to test most chemicals.
For the same reasons, scientists often use mathe-
matical models to extrapolate the results of high-dose
exposures to low-dose levels. Then they extrapolate
the low-dose results from the test organisms to humans
to estimate LD50 values for acute toxicity (Table 14-1).
Tw o general types of dose-response curves exist
(Figure 14-10). With the
nonthreshold dose-response model
(Figure 14-10, left), any dosage of a toxic chemical or
ionizing radiation causes harm that increases with the
dosage. With the
threshold dose-response model
(Fig-
ure 14-10, right), a threshold dosage must be reached
before any detectable harmful effects occur, presum-
ably because the body can repair the damage caused
by low dosages of some substances.
Establishing which of these models applies at low
dosages is extremely difficult and controversial. To be
on the safe side, the nonthreshold dose-response model
often is assumed.
Science, Ethics, and Economics: Using
Laboratory Experiments to Estimate Toxicity
Exposing a population of live laboratory animals
(especially mice and rats) to known amounts of a
chemical is the most widely used method for
determining its toxicity.
The most widely used method for determining toxicity
is to expose a population of live laboratory animals
(especially mice and rats) to measured doses of a spe-
cific substance under controlled conditions. Animal
tests take 2-5 years and cost $200,000-2,000,000 per
substance tested. Such tests can be painful to the test
animals and can kill or harm them. The goal is to de-
velop data on the responses of the test animals to vari-
ous doses of a chemical, but estimating the effects of
low doses is difficult.
Animal welfare groups want to limit or ban the
use of test animals or ensure that they are treated in
the most humane manner possible. More humane
methods for carrying out toxicity tests are available.
They include computer simulations and using tissue
cultures of cells and bacteria, chicken egg membranes,
and measurements of changes in the electrical proper-
ties of individual animal cells.
These alternatives can greatly decrease the use of
animals for testing toxicity. Some scientists point out
Nonlinear
dose-response
Linear dose-
response
Threshold
level
Dose
No threshold
Dose
Threshold
Figure 14-10
Science:
two types of
dose-response curves.
The linear and nonlinear curves in the left graph apply if even
the smallest dosage of a chemical or ionizing radiation has a
harmful effect that increases with the dosage. The curve on the
right applies if a harmful effect occurs only when the dosage
exceeds a certain
threshold level.
Which model is better for a
specific harmful agent is uncertain because of the difficulty in
estimating the response to very low dosages.
