Human Health Risk Assessment - Dealing with Contaminated Sites: From Theory Towards Practical Application - page 236

exposure of the test organisms versus irregular exposure of human beings in real
cases). Extrapolation also relates to intraspecies variation, that is, the differences in
sensitivity between human individuals. Adjustment factors could cover, for exam-
ple, the influence of exposure to more than one contaminant and possible synergistic
effects, which causes the overall effect to be stronger than the sum of the separate
effects of each contaminant. Uncertainty relates to limitations in the quality of the
experimental data set such as the number of tests performed, effect measurement
errors, or the use of a LOAEL instead of a NAEL. There is a lot of confusion about
terminology in regard to these factors. Extrapolation factors as defined above are
often used from a wider angle, that is, as the term assessment factor above.
Another option is to include political and socio-economic factors in the derivation
of an overall assessment factor, as is done in ECETOC ( 1995 ).
Over the last few decades, there has been a lot of debate about the use of
appropriate assessment factors. Many researchers have criticized the poor scien-
tific foundation and the conservative nature of assessment factors (e.g., Slob 1999 ),
while others referred to the lack of uniform terminology and of their use.
In the early days of Human Health Risk Assessment, when the focus was mainly
on exposure through food and drugs use, an overall assessment factor of 100 was
often used such as by the U.S. Food and Drug Administration (Lehman and Fitzhugh
1954 ) and the Joint FAO/WHO Expert Committee on Food Additives (JECFA
1987 ). The factor of 100 was a quantitative representation of a qualitative analysis
of differences between test animals and humans. Since then, additional insight has
been gained into the true value of assessment factors. The Dutch Health Council, for
example, has tried to apply differences in body weight, represented by differences
in caloric demand.
The default assessment factor of 100 can be subdivided into two assessment fac-
tors of 10, one for interspecies extrapolation and one for intraspecies variation. Each
of those can be subdivided into two factors, accounting for toxicodynamic and tox-
icokinetic aspects. Using the chemical-specific information on toxidynamics and
toxicokinetics, modifications in the default sub-factors can be considered (ECHA
2008 ; International Programme on Chemical Safety 2005 ).
Finally, the Critical Exposure value is calculated as follows:

POD
AF

(5.5)

critical exposure value =

where POD is Point of Departure (e.g., the NOAEL, LOAEL or BMDL) and AF is
the assessment factor.
A more sophisticated approach towards assessment factors is the so-called
Benchmark dose concept in which the separate assessment factors (and the effect
levels such as the NOAEL or LOAEL) are expressed as a probability density func-
tion and the overall assessment factor is derived by Monte Carlo techniques (e.g.,
Filipsson et al. 2003 ; Slob and Pieters 1998 ).
The derivation of Toxicological Reference Values for threshold contaminants are
described in detail in Langley ( Chapter 12 of this topic).