Ethics of Research involving animals
Reduction - continuation V
Barriers to implementing Refinement
12.23 We have observed above that implementation of Refinements is usually more straightforward than Reduction and Replacement and that many Refinements have been developed by scientists during the normal course of their work. There should in fact be fewer scientific barriers to Refinement. Where they do occur, it can be difficult to determine whether they are real or perceived and exactly what the nature of the barrier is. One common concern is that the provision of Refinement in the form of environmental enrichment may add unwanted variables that may reduce the validity of experimental data. For example, in toxicology it might be argued that giving wooden chews to animals affects their metabolism and interfers with the results. This may mean that more animals have to be used to generate statistically significant results. However, there are usually ways around such problems, for example, by using commercially available enrichments that have been fully characterised and standardised.
12.24 Other, non-scientific, barriers to the application of Refinement can result from:
- limited understanding of the concept of Refinement, why it is important and when and how to apply it;
- limited understanding of the species-specific needs of animals, causes of suffering and the impact of laboratory research and housing on the full lifetime experience of an animal;
- lack of specific information and guidance on practical Refinements, relating to what to do and how to do it;
- lack of resources, including time and funds; and
- lack of motivation and training.
12.25 All of these factors can significantly limit the implementation of Refinement, which, in view of its relative ease of application and its great potential for reducing suffering, is regrettable. Nevertheless, many establishments in the UK are very proactive with regard to Refinement and, taking animal husbandry as an example, have good, innovative environmental enrichment programmes. But there is also anecdotal evidence that some researchers argue that animals ‘do not do anything’, and therefore do not need anything to do. Such judgements point to limited knowledge of animal behaviour because the converse is often true. Animals that have nothing to do tend not to do anything.
12.26 Some problems in promoting Refinements arise from the fact that species-specific needs are not well characterised. Even when they are, as in our earlier example of rats, available knowledge is not always applied. For example, grid floors instead of solid floors may be used in some establishments without specific scientific justification and important resources such as substrate and nesting material are not universally provided. In a survey published in 2001, up to 25 percent of rats received no nesting material, up to 35 percent were not given anything to gnaw and over 50 percent were not provided with refuges, all of which play a significant role in relation to promoting the well-being of rodents.31 The effectiveness with which pain is managed is also inconsistent as practice with regard to recognising and alleviating animal pain varies across different establishments and abilities of different people.32
12.27 Another significant barrier to the implementation of Refinement is the relative dearth of detailed and accessible information on, and practical examples of, Refinement. Relevant information tends to be found in journals on research techniques and animal welfare which are rarely consulted by researchers whose primary interest is in their own specialised research field. Useful information is provided by the reports of the Joint Working Group on Refinement (JWGR),33 which provide practical advice on Refinement in husbandry of rabbits,34 mice,35 birds36 and dogs;37 and procedures including blood sampling,38 administration of substances,39 telemetry40 and the generation and care of GM mice.41 Lack of time and resources can also have implications for the implementation of enrichments and other Refinements, since such improvements require significant expenditure on staff and materials next page
Footnotes28 See Hendriksen CFM and Morton DB (Editors) (1999) Humane Endpoints in Animal Experiments for Biomedical Research.
Proceedings of the Int Conference, 22–5 Nov 1998, Zeist, The Netherlands. (London: Royal Society of Medicine).
29 Johannes S, Rosskopf-Streicher U, Hausleithner D et al. (1999) Use of clinical signs in efficacy testing of erysipelas vaccines, in
Humane Endpoints in Animal Experiments for Biomedical Research. Proceedings of the Int Conference, 22–5 Nov 1998, Zeist,
The Netherlands, Hendriksen CFM and Morton DB (Editors) (London: Royal Society of Medicine), pp102-105; Krug M and Cussler K (1999) Endotoxin in porcine vaccines: clinical signs and safety aspects, in Humane Endpoints in Animal Experiments
for Biomedical Research Proceedings of the Int Conference, 22–5 Nov 1998, Zeist, The Netherlands, Hendriksen CFM and Morton DB (Editors) (London: Royal Society of Medicine), pp114–17.
30 See also: Hendriksen CFM and Morton DB (Editors) (1999) Humane Endpoints in Animal Experiments for Biomedical Research. Proceedings of the Int Conference, 22–5 Nov 1998, Zeist, The Netherlands. (London: Royal Society of Medicine).