Chapter 8 
Alternatives to Animal Use in Testing 
Alternatives to using animals in testing serve the 
same purposes that using whole animals does— 
protecting and improving human health and com- 
fort. The technologies on which alternatives are 
based result primarily from biomedical and bio- 
chemical research. Several of them are reviewed 
in this chapter, though they are discussed in 
greater detail in chapter 6. Some alternatives that 
might eventually replace the tests covered in chap- 
ter 7 are also described here. 
Notable progress in the move to alternatives has 
been achieved in certain areas (78). For example, 
biochemical tests to diagnose pregnancy have re- 
placed those using rabbits, and the Limulus ame- 
bocyte lysate test, which relies on the coagulation 
of a small amount of blood from a horseshoe crab, 
has replaced rabbits in testing for the presence 
of bacterial endotoxins that would cause fever 
(25,117). Many companies have modified the widely 
used LD S0 test to use fewer animals (22) and have 
otherwise refined the methods used to test for tox- 
icity (100). Mammalian cell culture assays are used 
extensively in industrial laboratories for safety test- 
ing of medical devices (52,53) and pharmaceutical 
CONTINUED, BUT MODIFIED, 
It has been suggested that many more animals 
are used for testing than are needed (90) and that 
changes in experimental design or improved meth- 
ods of data analysis could substantially reduce the 
number of animals used. Each experiment has 
unique requirements (see ch. 7), and the ways in 
which the number of animals might be reduced 
will vary accordingly. 
Many of the methods discussed in chapter 6 for 
the modified use of animals in research are also 
applicable to testing, such as gathering more data 
from each animal or improving the analysis of re- 
sults by using random block design or covariance 
analysis. In random block design, animals with a 
particular characteristic, such as litter mates or 
animals of a certain size, are randomly assigned 
to different groups to balance whatever effect 
substances (1,84) and as immune response assays 
(97,98). 
The development of alternatives to animals in 
testing has accelerated in recent years with the 
establishment of programs having development 
and implementation of alternatives as their goal 
(see ch. 12). However, the barriers to adoption of 
these tests are more than the technical barrier of 
developing and validating a new technology . Test- 
ing is an integral part of many regulatory schemes 
and product liability law, and validation ultimately 
rests on acceptance by the scientific, regulatory, 
and legal communities. 
Public concern over animal use in testing appears 
to be increasing in tandem with public concern 
for product and drug safety. Ironically, the pub- 
lic's increasing concern for safety could lead to 
more testing. Yet it also provides an incentive to 
develop new techniques, particularly those that 
promise to be cheaper and faster than current 
whole -animal methods . A further irony is that de- 
veloping alternatives, as well as validating them, 
sometimes requires animal use. 
USE OF ANIMALS IN TESTING 
these variables might have. If the groups being dis- 
tributed are sufficiently large, the results can also 
be analyzed to determine the effect of the mask- 
ing variable (47). Covariance can be used to ana- 
lyze results when some of the experimental varia- 
bles are uncontrolled but known, thus estimating 
their effect on the results. 
As in research, the number of animals needed 
as controls can be reduced by using the same group 
as a control for several simultaneous experiments. 
A laboratory’s ability to do this will be limited bv 
its size and the amount of lead time available to 
allow testing to be coordinated . Another difficulty 
is that environmental conditions must be exactly 
the same and the tests must start and finish at ex- 
actly the same times. The reduction in animal use 
that simultaneous experiments brings about is 
175 
