In many countries scientists planning research that may involve the use of animals are required by law to examine the possibilities for replacement, reduction or refinement (the Three Rs) of these experiments. In addition to the large number of literature databases, there are now many specialist databases specifically addressing the Three Rs. Information centres, with a mandate to assist scientists and lay people locate information on the Three Rs, have also been established. Email discussion lists and their archives constitute another, although less quality-controlled, source of information. Furthermore, guidelines for the care and use of animals in research have been produced both by regulatory bodies and scientific organisations. The growth of the internet has put an enormous amount of data into the public domain, and the problems of accessing relevant information are discussed. Suggestions are also given for search strategies when using these information sources.

This paper highlights the essential need for appropriate statistical design and randomisation in laboratory animal studies. Using an example of a 21 day weight gain study in mice, we show that without the use of an appropriate statistical design and randomisation, incorrect conclusions may have been drawn. We used an experimental design that allowed comparisons to be made between five treatments that were free from systematic error. Two alternative designs that are practically attractive, yet had no statistical basis, are also described in this paper and the potentially incorrect conclusions highlighted. The use of appropriate statistical design is ethical because it results in clear, unambiguous conclusions. Conclusions that may be biased or ambiguous will require verification by further research and this, in the long term, is contrary to the reduction element of the Three Rs.

Six drug regulatory reviewers and 11 pharmaceutical industry scientists were interviewed to explore their perspectives on the obstacles and opportunities for greater implementation of the Three Rs (replacement, reduction, refinement) in drug research and development. Participants generally supported the current level of animal use in the pharmaceutical industry and viewed in vitro methods as supporting, but not replacing, the use of animals. Obstacles to greater use of the Three Rs cited by participants included the lack of non-animal alternatives; requirements for statistical validity; reluctance by industry and regulators to depart from established patterns of animal use; the priority of commercial objectives ahead of the Three Rs; and concern that less animal testing could jeopardise human safety. Opportunities identified for the Three Rs included the development of better animal models including genetically modified (GM) animals; pursuit of more basic knowledge, notably drug action on gene expression; re-use of animals; greater use of pilot studies; using sufficient numbers of animals per test to avoid repeating inconclusive studies; regular review of animal data in regulatory requirements; and following the regulatory option of combining segments of reproductive toxicology studies into one study. In some areas, greater implementation of the Three Rs seemed well aligned with industry priorities, for example, phenotypic characterisation of GM animals and validation of alternative methods. In other areas, wider use of the Three Rs may require building consensus on areas of disagreement including the usefulness of death as an endpoint; the suitability of re-using animals; and whether GM animals and the use of pilot studies contribute to reduction.

Russell and Burch's Three Rs (replacement, reduction and refinement) remain the cornerstone for principles guiding humane experimental research. However, the concept of refinement has evolved considerably since its first inception and there have been numerous interpretations, some of which are regressive from the original definition. In this paper we examine the interpretations of refinement, and propose a harmonised progressive definition that is in line with changes in animal ethics and animal welfare science. Our definition should be applied to all aspects of refinement: those related to housing, husbandry and care, techniques used in scientific procedures, periprocedural care, health and welfare monitoring, and experimental design. We argue not only that the concept should include the avoidance or minimisation of adverse effects experienced at any time during the life of an animal destined for use in a laboratory, but also that it should be applied to the founder animals. Furthermore, we take a proactive stance and argue that refinement should include enhancing well-being through environmental enrichment. The acceptance and application of this new definition by legislative authorities and in guidelines would represent a significant step forward for animal welfare.