Page 207 - Materials Chemistry, Second Edition

L1644_C05.fm Page 180 Monday, October 20, 2003 12:02 PM

the like. The net result is that the exposure assessment will be based on a number
of assumptions with varying degrees of uncertainty (U.S. Environmental Protection
Agency, 1992). Decision analysis literature has focused on the importance of explic-
itly incorporating and quantifying scientific uncertainty in risk assessment (Rose-
berry and Burmaster, 1991).
Several reasons lead to uncertainties concerning the validity and entirety of the
results of a risk assessment. These uncertainties can be regarded in different manners
and degrees depending on the methodology applied in the risk assessment process.
One source of high uncertainties is the application of models that simulate the
behavior of a pollutant in the environment and the uptake into the human body.
Computer models that attempt to describe natural processes are always simplifica-
tions of a complex reality. They require the exclusion of some variables that in fact
influence the results but cannot be regarded because of increased complexity or lack
of data. Moreover, many natural processes can only be approximated but not exactly
explained with mathematical correlations. Hence, a model is always affected with
uncertainties and gives only an imperfect description of the reality. Different models
for the same issue consider different uncertainties but disregard also different sources
of uncertainty.
On the other hand, because many parameters in a model cannot be treated as
fixed-point values, a range of values better represents them. This uncertainty of input
parameter can result from real variability, measurement and extrapolation errors as
well as the lack of knowledge regarding biological, chemical and physical processes.
Uncertainties that are related with lack of knowledge or measurement and extrapo-
lation errors can be reduced or eliminated with additional research and information.
However, real parameter variability, e.g., spatial and temporal variation in environ-
mental conditions or life-style differences, occurs always and cannot be eliminated.
It leads to a persisting uncertainty of the modeling results.
Risk assessment is subject to uncertainty and variability. Specifically, uncertainty
represents a lack of knowledge about factors affecting exposure or risk, whereas
variability arises from true heterogeneity across people, places, and time. In other
words, uncertainty can lead to inaccurate or biased estimates, whereas variability
can affect the precision of the estimates and the degree to which they can be
generalized.
Now let us consider a situation that relates to exposure, such as estimating the
average daily dose by one exposure route — inhalation of contaminated air. Suppose
that it is possible to measure an individual’s daily air inhalation consumption (and
concentration of the contaminant) exactly, thereby eliminating uncertainty in the
measured daily dose. The daily dose still has an inherent day-to-day variability
because of changes in the individual’s daily air inhalation or concentration of the
contaminants in air.
Clearly, it is impractical to measure the individual’s dose every day. For this
reason, the exposure assessor may estimate the average daily inhalation based on a
finite number of measurements, in an attempt to “average out” the day-to-day
variability. The individual has a true (but unknown) average daily dose, which has
not been estimated based on a sample of measurements. Because the individual’s
true average is unknown, it is uncertain how close the estimate is to the true value.

© 2004 CRC Press LLC