228 • Alternatives to Animal Use in Research, Testing, and Education 
cals sometimes refers information requests among 
member countries through its national corre- 
spondents. 
A recent U.S. -led project of the OECD, generally 
referred to as “Switchboard,” has also addressed 
the problems of obtaining information from other 
countries. Unpublished information may be re- 
quested through the Switchboard system for use 
in risk assessments or to otherwise protect health 
and the environment. A pilot system is to be run 
in which two requests per participating country 
COMPUTER 
Computers have two applications as an alterna- 
tive to using animals in research, testing, and edu- 
cation. First, they can be used to model or simu- 
late biological, chemical, and physical systems. In 
this way, a computer could be used as a direct 
replacement for some number of animals used in 
laboratories. This form of computer use is dis- 
cussed in chapters 6, 8, and 9. Second, computers 
are used to disseminate information that has been 
generated from prior use of animals in research 
and testing, thus avoiding the needless repetition 
of a procedure by other scientists. It is this role 
of computers as information disseminators that 
is discussed in the rest of this chapter. 
Advantages of Computers 
Biological testing (see ch. 7) can be described as 
the repetitive use of a standard biological test situ- 
ation, or protocol, employing different chemicals 
or different test parameters (e.g., species or bio- 
logical end points). Because the protocols in test- 
ing are more stereotyped and less varied than those 
in research, biological testing is more amenable 
than research to the institution of a computerized 
data retrieval system. In fact, testing emerged in 
the 1970s as the first discipline in which such a 
system was developed. 
If a comprehensive, computerized registry of bio- 
logical research or testing data were established, 
certain benefits might accrue. These benefits are 
predicated on the inclusion in the computerized 
registry of both control and experimental data, 
and of both positive and negative results. (Data 
per month would be referred to some combina- 
tion of government agencies, industry, academia, 
and research institutes that might have unpub- 
lished data relevant to the request. The mecha- 
nisms for referring requests on the national level 
and the enlistment of various organizations, either 
as requesters or responders, is the responsibility 
of Switchboard’s national focal point. This project 
will begin on a small scale and will be monitored. 
If appropriate, it could be expanded (18). 
SYSTEMS 
obtained from testing fall into two broad catego- 
ries: those derived from untreated (control) sub- 
jects, and those from treated (experimental) sub- 
jects. Data obtained from treated subjects may 
either show an effect from the treatment ("posi- 
tive results”) or no effect (“negative results”).) Fur- 
thermore, the advantages of such a registry de- 
pend on the acceptance by working scientists of 
the data contained in it— acceptance that seems 
possible only with the imprimatur of peer review 
of the data. The anticipated benefits of a computer- 
based registry of research or testing data include: 
• Decreased Use of Animals in Research or 
Testing. In some instances, an investigator 
would locate the exact data desired, possibly 
from a previously unpublished source, thus 
avoiding unintentional duplication of animal 
research or testing. Baseline data could per- 
mit the selection of a dose, a route of adminis- 
tration, or a strain of animal without the need 
for new animal experiments to establish these 
factors. Efficiencies could also include the use 
of fewer doses on smaller numbers of animals . 
Conceivably, the number of animals required 
for control groups could be reduced, although 
many experimental protocols require the use 
of concomitant control subjects, rather than 
of data from a pool of control subjects, in or- 
der to achieve statistical significance. 
• A Check for Genetic Drift. Certain experi- 
mental results can change over a span of many 
generations due to subtle, progressive changes 
in the underlying genetic constitution of the 
strain of animals ("genetic drift”). The regis- 
