¿ìè¶ÌÊÓƵ

During the 1980s, the debate over animal experiments has polarised,
sometimes drowning out the voices in the centre. Some participants are looking
for a semantic way out of their difficulties, and animal protection is
supplanting animal rights as the in vogue word to describe groups promoting
the wellbeing of laboratory animals. But whatever the terminology, the main
piece of legislation under debate in the US is the Animal Welfare Act, which
first made its appearance in 1966.

Compared to British legislation, it is remarkably weak. For a start
it excludes the most commonly used laboratory animals, rats and mice, as
well as cold-blooded vertebrates, all of which come under protection in
Britain. Furthermore, in Britain, legislation has moved from a system which
licensed researchers to carry out a wide range of experiments to one that
directly controls what they do, and which requires specific justification
of each project. Many other countries are moving in a similar direction.
‘A notable exception to this trend is the United States,’ says a working
party on animal experiments set up by Britain’s Institute of Medical Ethics.
‘There, legislation still specifically excludes the research practices
themselves’ (Lives in the Balance, Oxford University Press, 1991).

America’s Animal Welfare Act offers very few rules, regulations or standards
concerning the actual practice of research using animals, and those it does
can sometimes be ignored in the interests of science. For instance, painkillers
and anaesthetics can be omitted if principal investigators can convince
the committees that vet experiments that such agents would interfere with
the science. Moreover, the Office of Technology Assessment, the research
arm of Congress, concluded in 1986 that the enforcement agency has ‘neither
the staff nor the budget to secure compliance even with the limited provisions
of the act’.

When the act became law 26 years ago, it covered only dogs, cats, hamsters,
rabbits, guinea pigs and nonhuman primates held in laboratories. Subsequent
amendments broadened the act to cover all warm-blooded animals (excluding
fish, amphibia and reptiles covered by British law). However, the Department
of Agriculture (USDA), which is the enforcing agency for the act, interpreted
the law to mean that the Secretary of Agriculture could define warm-blooded
animal. The upshot was to exempt rats, birds, mice and agricultural animals
used in biomedical research from the act.

Last year, the USDA, prompted by animal welfare groups, reinterpreted
its obligations under the Animal Welfare Act and brought agricultural animals
used in biomedical research under its control. The scientific community
and animal welfare groups are now waiting for the Department of Agriculture
to issue the regulations that will spell out what this means in practice.
John Miller, now the director of the NIH’s office for animal welfare, told
a conference in 1990 that one issue to be settled concerns the conditions
for ‘survival’ surgery – in which anaesthetised animals are allowed to recover
consciousness. Surgery on smaller farm animals can take place in aseptic
facilities set up for dogs and nonhuman primates, but larger animals simply
won’t fit in the operating theatres. Sometimes, vets operate in the open;
should that be the case for biomedical research?

Recognition for rodents?

Many scientists and animal welfare groups believe that birds, and rats
and mice, which comprise roughly 80 per cent of the animals in laboratories
in the US, will also soon be included in the act. In January, the Humane
Society of the US and the Animal Legal Defense Fund obtained a judgement
in a federal court which stated: ‘The secretary’s construction of the statute
to exclude birds, rats and mice from the definition of animal is arbitrary
and capricious.’ But the USDA says it will appeal, and has insufficient
funds to inspect the vastly increased number of institutions that would
then have to register as research facilities.

Any institution using animals covered by the act must register that
use with the USDA. The laboratories are then subject to unannounced inspections.
In 1990, inspectors visited the 3339 registered research establishments
an average of 1.39 times, and suspended research or sent cease and desist
orders to 51 of them and warnings to a further 252. The enforcement of the
Animal Welfare Act is a bone of contention with animal welfare groups. Martin
Stephens, vice-president for laboratory animals at the Humane Society of
the US, says: ‘The current system is a good starting point for regulation
if it includes all species and is properly enforced.’

The OTA noted in 1986 that the USDA itself sees failure to register
as a problem. Miller believes this to be a relatively small problem and
says that in some cases it is because a small college is running animal
experiments for the first time and is unaware that it should register. The
only laboratories that do not have to register are other government agencies,
such as the NIH itself, NASA, the Department of Defense and the Department
of Energy. These institutions must by law monitor their own compliance with
the act. The OTA’s report estimates that about 1.6 million animals are used
each year in government laboratories, of which 90 per cent are rats and
mice.

Research establishments in the US are now in the process of complying
with regulations that emerged after the most recent amendments to the Animal
Welfare Act in 1985. Those amendments require that procedures involve the
minimum of discomfort, distress and pain; that the principle investigator
has considered alternatives to procedures that cause more than momentary
pain; that procedures are performed with the appropriate sedatives, analgesics
and anaesthetics, unless withholding such agents is justified by the science.
No muscle relaxants, such as curare, are allowed without anaesthetic. A
procedure should not call for an animal to recover from a second invasive
procedure, unless the science justifies a recovery. And animals that would
suffer chronic pain if they survived a procedure are to be painlessly put
down.

The amendments also called for improvements in housing for cats, regular
exercise for dogs and conditions for nonhuman primates that would promote
their psychological wellbeing.

When the regulations spelling out what these general provisions meant
first appeared in 1989, researchers were incensed, claiming that compliance
would cost about $1 billion and make some research too expensive. The regulations
went through various revisions, and were considerably softened. Laboratories
now have until 1994 to comply fully with the regulations and can decide
themselves how to meet the required standards.

Perhaps one of the most important provisions of the amendments was the
requirement that any place using animals covered by the act should set
up ‘institutional animal care and use committees’ to make sure its research
procedures live up to the general principles spelt out in the amendments.
Barbara Rich from the National Association for Biomedical Research, says:
‘We have a lot vested in these committees.’ Most institutions, in particular
big research centres such as the Dana Farber Cancer Institute, also produce
guidelines detailing how certain procedures should be carried out, and
the manuals serve as reference material for the committees.

It is these committees, which, in the US, carry the control of scientific
experiments on animals beyond the laboratory door. Not surprisingly, they
are a focus for both animal welfare groups and scientists. On the whole,
scientists wish to keep them because they provide a mechanism for what Miller
calls ‘enforced self-regulation’. He believes that they allow for a more
effective means of regulation than a system that relies on visits from external
inspectors looking for compliance with rigid laws.

Does self-policing work?

With the committees in place, Miller says that his office is receiving
an increasing number of official and unofficial notifications from the committees
saying they have noticed deviations from approved procedures during the
twice-yearly visits which the law stipulates they must make to laboratories.
Such ‘discrepancies’, he says can be, and are, corrected. Presumably, inspectors
from the USDA are meant as a safety net in case a committee doesn’t do its
job.

USDA regulations say that the committees should have a minimum of three
people, one of whom must be a vet familiar with laboratory animals; another
member must come from outside the institution and have no family ties inside
it. The NIH guidelines stipulate a minimum of five members, of which one
must be a vet, one a nonscientist within the institution and one from outside
it.

Much debate centres around the appropriate background for the representative
from outside the institution. Bruce Ewald, a vet who chairs the Animal Care
and Use Committee of Ciba-Geigy’s pharmaceutical division, says that he
believes the person should be a nonscientist who can raise questions that
won’t occur to other members. Ciba-Geigy’s external member is a local minister.
Welfare groups, and academics analysing the dynamic of these groups, worry
that the nonscientist will be intimidated by the scientists, and will frequently
be outvoted even if they do speak out against a particular procedure. Animal
welfare groups receive calls from lay members of these committees saying
that, in fact, this does happen. Some animal welfare advocates fear that
the committees simply rubber-stamp research. If, however, the committees
work as Congress intended, the burden of proof that a particular procedure
is warranted and that a particular species should be the subject lies with
the principal investigator. Probably, many committees do work as they should,
but it is difficult to tell from the outside.

Another issue is the unevenness that these committees bring to their
interpretation of the 1985 amendments. Where the trouble arises is in withholding
pain relief from animals if this is justified by the science, and in accepting
more than two recoveries from invasive surgery if justified by the science.
Stephens says: ‘The science is nearly always placed first. Eventually,
we are going to have to come to grips with the issue of determining the
benefit of the research compared with the pain stress and boredom suffered
by animals.’

Some of the benefits, as listed in a report by the National Research
Council, the investigative arm of the US’s National Academy of Science,
include, among many, improved surgical techniques and vaccine development.
As Dame Mary Warnock, a philosopher who chaired Britain’s committee to review
the regulation of laboratory animals, said last year: ‘There is no alternative
but to hold the pain/benefit debate in public. It is not good to exaggerate
the possible therapeutic effect of research, but I can imagine circumstances
in which I would think it was appropriate to inflict pain on some other
sentient being, providing that one could demonstrate that the outcome would
be to have considerable impact on long-term pain of humans.’

While a philosophical debate about pain and benefit, which extends beyond
the scientific community, is clearly needed, the animal care and use committees
have to wrestle with case by case decisions, such as whether to allow injections
of an irritating mixture known as the complete Freund’s adjuvant. Injecting
an animal with this substance promotes a strong antibody response to a weak
antigen. That same strong response can cause severe inflammation at the
site of injection and kill tissue. Ewald says that the committees differ
in their views about how this procedure should be carried out.

To some extent, the committees’ views about this procedure rely on their
evaluation of the value of the science versus the pain of the animal. Stephens
says: ‘Pain is key to the act, but it is not dealt with effectively.’ The
amendments define pain simply: in the absence of knowledge to the contrary
a procedure which causes pain to a human would also cause pain to an animal.
Yet pain is allowed if justified by the science. Stephens would like to
see a system introduced like that in Britain which classifies pain more
precisely, and weighs the severity of pain against the value of the science.

Is pain under control?

According to figures from the USDA, pain is an issue in only 6 per cent
of procedures performed on laboratory animals. Not surprisingly, this figure
is contested. It also raises another point of contention between scientists
and animal welfare groups: that of the way in which researchers report their
activities. The 1985 amendments strengthened reporting requirements. Each
year, research institutions must tell the USDA how many animals of each
species covered by the act they use (rats and mice can be reported voluntarily).
They must also indicate how many procedures they carried out in each of
three categories: painless, relieved pain and unrelieved pain. A painful
procedure is defined as one that is more than momentary. An injection is
momentary, but if, for example, it delivers complete Freund’s adjuvant,
the animal is in pain afterwards. The act specifies that the appropriate
analgesics can be administered, but the immunologist may not think that
an analgesic is appropriate.

To clarify whether there is widespread underreporting of pain, animal
welfare groups would like to see the nature of procedures recorded. This
would form part of a ‘utilisation survey’, which would also include a breakdown
of animal use by sector, say product testing of pharmaceuticals or cosmetics
or biomedical research. In fact, what animal welfare groups would like to
see is very similar to what happens in Britain. It would also, they say,
help them to target their campaigns – something the research community is
well aware of.

Some researchers dispute the value of numbers. Barbara Rich, from the
National Association for Biomedical Research, says: ‘I have seen the British
numbers, they’re not very helpful, you could be deceived by apparent trends.’
She added: ‘They are expensive to collect.’ Rich says, however, that the
USDA’s figure that only 6 per cent of experiments are painful is accurate.

In a report entitled The Use of Laboratory Animals in Biomedical and
Behavioural Research, the National Research Council found the statistics
available on animals used in experiments imprecise. The report said that
although it would be expensive, collecting better numbers would make it
easier to assess the impact of policy on trends of animal use. The NIH is
now conducting a survey that would break down figures by sector, but it
will apply only to research funded by the institutes.

For now, debate seems set to continue between those who say you can
show anything with numbers and those like Henry Spira, a veteran human
and animal rights campaigner, who told the conference at Johns Hopkins:
‘If you don’t count the numbers, you run the risk of giving the impression
that the numbers don’t count.’

Helen Gavaghan is a freelance science and technology writer based in
Washington DC.

* * *

Testing times for laboratory animals

Alexis de Tocqueville, the 19th century French political writer who
commented extensively on American society, coined the phrase that the public
would rather believe a simple lie than a complex truth. The Johns Hopkins
Center for Alternatives to Animal Testing adopted this aphorism as a backdrop
to a meeting it held in April about the progress during the past 10 years
with in vitro toxicology. It served as an apt warning of the political,
ethical and scientific complexity that surrounds debate about the use of
alternatives to living animals when testing medical, cosmetic and household
products for the market.

There is little doubt, according to Andrew Rowan, director of the Tufts
Center for Animal and Public Policy in Massachusetts, that in vitro toxicology
is the way of the future. The idea is to produce tests that reduce the number
or suffering of live animals in testing new products. And the challenge
for scientists is to convince the regulatory authorities that the alternatives
are as effective as animal tests for highlighting any damaging effects products
might have – the buzzword is validation.

In the US, there is a lot of activity in this field, says Rowan, but
the effort is not coordinated. Slowly, though, progress is being made.
Rowan is undertaking a study, called a National Agenda for Alternatives,
for the Environmental Protection Agency. His report is expected later this
year.

Rowan believes that responsibility for validation should lie with one
agency, but says: ‘There is little enthusiasm among the existing regulatory
authorities – EPA, Food and Drug Administration and the Consumer Products
Safety Committee – to assume responsibility for the task.’ An option,
he believes, would be to place responsibility for validation in the National
Toxicology Programme of the National Institute of Environmental Health Science.

Alternatives could not, in the current state of science, replace all
animal tests. The idea, though, is to follow the ‘three Rs’ – replacement,
refinement and reduction. Replacement is obvious, refinement means finding
the least painful protocols and reduction means developing alternatives
that give reliable results from fewer animals.

One of the most controversial issues in this arena is replacement of
the Draize eye irritancy test. This test involves placing possible irritants
into one eye of a rabbit. The other eye serves as a control. Rabbits do
not have tear ducts and cannot dilute the effect of an irritant. Possible
alternatives to this test already exist, but they have not yet been validated
by being repeated in many laboratories using a wide range of chemicals.

A document published by the OECD, the ‘club’ of the major industrialised
nations, suggests a way to speed up the validation process. If an alternative
test developed by one company seems to work when replicated in a few laboratories,
it could be sent to say 15 or 25 laboratories together with scores of samples
from central ‘chemical banks’. If all looks well, the test could then go
to 50 or 60 laboratories along with hundreds of chemical samples. Rowan
says: ‘The paper defines validation in a complex and long-winded way. In
common sense terms, I think it is saying when people believe its validated,
then it’s validated.’

Most progress, according to Bernard Robaire, a professor in the Department
of Dermatology at McGill University in Canada, has been in the field of
pharmacodynamics. Here, among other things, scientists need to know how
effectively a drug binds to a ‘receptor’ site in the body. As knowledge
increases, more and more work on the effectiveness of receptor binding can
be modelled on computers.

Perhaps least progress exists in the field of pharmacokinetics, where
scientists need to know how effectively the body absorbs, distributes, metabolises
and eliminates a drug. Bruce Ewald, who chairs Ciba-Geigy’s Animal Care
and Use Committee in the pharmaceutics division, explains: ‘A drug may work
on isolated receptors, but its action is different in the system. For example,
what does acidity in the stomach do to the drug, how is it excreted? That
could be important for a physician treating someone who also has a kidney
³¦´Ç³¾±è±ô²¹¾±²Ô³Ù.’

In toxicology, which includes testing for fatal doses, carcinogenesis,
the drug’s ability to cause genetic changes (muta-genicity), effects on
the reproductive system and on a fetus, progress is uneven. In each area,
scientists need to understand the biology of what happens in the body before
they can produce an alternative. In this context, Robaire believes that
it will be hardest to develop alternatives to test the effect on reproduction,
but says there has been great progress in mutagenesis.

Science, of course, is not the only arbiter of progress in this arena.
Bill Cottrell, from the Animal Welfare Institute in Washington, says that
product liability is a stumbling block. Cottrell’s institution is in the
animal welfare/protection camp, but seeks to raise a moderate voice in the
debate. At one meeting, he says, a lawyer explained that he tells clients
to generate as much animal data as possible. Even if the mounds of data
are more than are necessary for product safety, there is safety in numbers
in the American courtroom.

* * *

Australian researchers take on animal rights

In Australia, policies to ensure the welfare of animals used by humans
have developed in rather different ways from those adopted in Britain. Until
about 25 years ago, Australian researchers were subject to relatively lenient
regulations for the use of animals. Each state had its own animal protection
laws, most of them dating from the 1920s. They did not spell out any detailed
guidelines and were rarely, if ever, enforced. Only when the animals were
seriously maltreated was any inspection or prosecution called for, and then
only if someone complained. Responsibility for the humane treatment of animals
was left up to the individual researcher, farmer or animal technician. No
detailed applications had to be made for a licence to conduct scientific
experiments using animals, and there were no official inspectors to check
experimental procedures.

It was in the late 1960s and early 1970s that Australia began to see
a rising tide of protest by the animal liberationists and organisations
such as the Humane Society and the RSPCA. ¿ìè¶ÌÊÓƵs who were using animals
for research responded by setting up their own guidelines and structures
for animal welfare. The aim was to get in first and avoid the introduction
of state or federal government laws.

The National Health and Medical Research Council worked swiftly and,
in 1969, drew up its code of practice for the care and use of animals for
scientific purposes. Most scientists reasoned that, if things had to change,
it was better for them to do so primarily under the control of scientists
rather than government bureaucrats. That way, the scientists felt, more
informed decisions would be made, and laboratory researchers would not be
plagued by surprise visits from government officers who had little understanding
of what they were trying to achieve. The code of practice was revised in
1979 and 1982 in conjunction with the CSIRO, the other major body concerned
with animal research in Australia, and in 1985 with additional input from
the Australian Agricultural Council.

In 1990 this code was revised again and it became the Australian Code
of Practice for the Use of Animals for Scientific Purposes. The code aims
to guarantee the ethical use and humane care of animals used for science
in the fields of medicine, biology, animal husbandry, industry and teaching.
It encompasses the use of animals for academic and applied research, teaching,
field studies, product testing, diagnosis and the production of biological
materials. It supplements the legislation relating to the use of animals
for scientific purposes enacted by several states. Its implementation is
overseen by Animal Experimentation and Ethics Committees (AEECs) at each
research institution and in the state of New South Wales by a state government
office as well.

The code of practice stipulates that an AEEC must have at least four
members, one of whom must be committed to furthering the welfare of animals
but not be employed by, or be otherwise associated with, the institution
concerned. In addition another independent person, who is not currently
or previously involved in animal research and who is ‘preferably’ not an
employee of the institution, must be included. Of the other two members,
one should be a person with qualifications in veterinary science and one
a person involved in animal research. Further members can be, and invariably
are, added to the AEEC. There is no requirement for the balance of membership
to be kept the same as in the initial structure for four members, and in
most committees the additional members are researchers actively involved
in animal research, and who therefore have a vested interest in it.

In addition to the Australia-wide code of practice, many states have
introduced legislation that protects the interests of animals used in research.
New South Wales, for example, has laws that require accreditation of institutions,
licensing of individuals and separate approvals by an AEEC of each scientific
undertaking which in any way uses animals. This legislation has been prompted
by continued pressure from public groups fighting for animal rights. Attitudes
to the use of animals have changed, and there is increasing public awareness
of the ongoing debate.

The substance of many of the guidelines and laws has come largely from
the scientific community, or has been strongly influenced by the national
code of practice. Thus guidelines which the scientists had originally drawn
up for themselves have become law in four states, though in the process
direct control has been removed from the scientists and their institutions.

Queensland has recently appointed its first state animal welfare officer,
directly following the election of a Labour government in that state after
many years of rule by the National Party. In February 1991, Queensland amended
its Animal Protection Act of 1925 to recognise the national code of practice.
The act in its original form allowed vivisection only if it was conducted
according to regulations authorised by the Home Secretary. As Queensland
has had no Home Secretary since 1935, all animal research in the state
had been technically illegal until the act was amended.

The Australian authorities have responded with impressive speed to changes
in attitudes to animal welfare which they have seen abroad, particularly
in Europe. Although antivivisection and animal rights movements are certainly
present in Australia and have influence, the pressures of both radical and
more middle-of-the-road groups have been anticipated at both institutional
and government levels. Consequently, radical protest has been less prevalent
and much less violent than in Britain. Without dramatic public protest,
Australia has improved standards of animal care and ethics within a relatively
short period of time. But these changes have concentrated on animals used
for scientific work, and further improvements are needed, particularly in
agricultural practices. And who knows what pressure and resistance those
changes will require or precipitate?

Lesley Rogers and Ruth Tremont

Lesley J. Rogers is associate professor and head of the Department of
Physiology, University of New England, New South Wales. Ruth Tremont, a
veterinary surgeon, is director of animal welfare at the University of
New England.