The 3Rs refer to a study published in 1959. At the annual meeting of the former American Association of Laboratory Animal Science in Washington D.C., the late Major Charles W. Hume, the founder of the Universities Federation of Animal Welfare (UFAW), presented a study by two English scientists, William Russell, described as a brilliant zoologist, psychologist and classics scholar, and Rex Burch, a micro- biologist. They had carried out a systematic study of the ethical aspects and "the development and progress of humane techniques in the laboratory".
Russell and Burch based their study on the philosophical concept of humanity - in the sense of "humane-ness" - v.s. inhumanity in the context of animal experiments. They stated that true humanity, which distinguishes humans from all other species, was the capability for social cooperation, intimately linked to a compassionate and empathetic attitude towards other species. Although they assumed that biologists treated their animals as humanely as possible within the boundaries of then-current experimental protocols, Russell and Burch observed that some procedures were inhumane per se. By analysing and documenting the relative "humanity" or "inhumanity" of biological experiments, the authors hoped to promote the development of humane experimental techniques and reduce the amount of pain and fear inflicted upon laboratory animals. Russell and Burch stated clearly that they limited their definitions of "humane" and "inhumane" strictly to the experimental procedures under analysis; no criticism or moral judgement of the experimenters involved was intended or implied.
How does one measure "inhumanity"? Russell and Burch used the criteria of pain, or more generally, "distress" experienced by the animals. Physiological and endocrine parameters offered objective measurements of stress. Any set-up using negative reinforcement (punishment) as motivation to train or condition responses qualified as a source of stress and fear. The behaviour of the animals towards the experimenters and each other served as a further indicator of well-being or distress (e.g. incidents of asocial behaviour such as biting and scratching, the need for restraints during experimental procedures vs. tameness, etc.).
A final area examined by Russell and Burch was the environment, both social and physical, of the experimental animals. All laboratory animals are alike in this respect: their environment is man-made. The size and distribution of their social groups, and their physical surroundings, tended to be optimized for the convenience of the experimenters, rather than to meet the needs of the animals.
The first step in the Russell and Burch study was to gather data on animal experimentation: what numbers of which species were used for what types of experiments. Their sources included the L.A.B. survey of 1952 (Laboratory Animal Bureau of the British Medical Research Council), which provided the following information: animal species used, laboratory type, purpose of research and the number of animals. Russell and Burch then analyzed each experimental procedure for its degree of inhumanity. They examined the incidence of inhumanity and graded the severity of the distress experienced by the animals from mild to severe. Especially severe forms of distress, such as potentially lethal experiments or operations with accompanying post-operative pain, were classified separately. The authors then combined the three variables to identify those procedures that were the most inhumane. This analysis provided the systematic basis upon which to launch a programme for humane procedures in experimental biology, termed the three R's: Replace, Reduce, Refine.
Replacement was the most radical proposal: The use of nonsentient organisms rather than higher animals for experiments. Microorganisms, metazoan parasites, and higher plants were suggested as possible alternatives. Experiments using these materials were labelled "absolute replacement", since no higher animals were required at any stage. In vitro techniques with cell cultures from animal tissues were defined as "relative replacement" procedures, since the experiments themselves were conducted on non-sentient material, but still depended upon animal materials.
Reduction meant obtaining the best quality and most precise information with the smallest possible number of animals. Experiments that were well-designed and well-conducted would deliver reliable results, and eliminate the need for endless repetition of the same tests. This included close cooperation with statisticians and establishing in advance the required level of statistical significance. Similar gains could be achieved by careful dose-level selections in dose-response studies.
Refinement was the most subtle approach. It referred to all changes in protocols that reduced the incidence or severity of distress experienced by laboratory animals.
Russell, W.M.S. and Burch, R.L., The Principles of Humane Experimental Technique. Methuen, London, 1959.
Reprinted by UFAW, 1992: 8 Hamilton Close, South Mimms, Potters Bar, Herts EN6 3QD England. ISBN 0 900767 78 2