Reduction strategies in animal research: a review of scientific approaches at the intra-experimental, supra-experimental and extra-experimental levels.

When discussing animal use and considering alternatives to animals in biomedical research and testing, the number of animals required gets to the root of the matter on ethics and justification. In this paper, some reduction strategies are reviewed. Many articles and reports on reduction of animal use focus mostly on the experimental level, but other approaches are also possible. Reduction at the intraexperimental level probably offers the greatest scope for reduction, as the design and statistical analysis of individual experiments can often be improved. Supra-experimental reduction aims to reduce the number of animals by a change in the setting in which a series of experiments take place--for example, by improved education and training, reduction of breeding surpluses, critical analysis of test specifications, and re-use of animals. At the extra-experimental level, reduction is a spin-off of other developments, rather than the direct goal. Through improved research or production strategies, aimed at better quality, consistency and safety, reduction in the number of animals used can be substantial. A revised definition of reduction is proposed, which does not include the level of information needed, as in some cases reduction in the number of animals resulting in less information or data, is still acceptable.