Rev. sci. tech. Off. int. Epiz., 2014, 33 (1), 265-272 Scientific uses of animals: harm–benefit analysis and complementary approaches to implementing the Three Rs G. Griffin (1)* , J. MacArthur Clark (2) , J. Zurlo (3) & M. Ritskes-Hoitinga (4) (1) Canadian Council on Animal Care, 1510-130 Albert St, Ottawa, K1P 5G4, Canada (2) Animals in Science Regulation Unit, Home Office, Marsham Street, London SW1P 4DF, United Kingdom (3) The Johns Hopkins Center for Alternatives to Animal Testing, Bloomberg School of Public Health Sciences, 615 N. Wolfe St. W7032, Baltimore, MD 21205, United States of America (4) SYRCLE (SYstematic Review Centre for Laboratory animal Experimentation), 231 CDL, Radboud University Medical Center, P.O. Box 9101, NL6500 HB Nijmegen, the Netherlands *Corresponding author: ggriffin@ccac.ca Summary The principles of humane experimental technique, first described by Russell and Burch in 1959, focus on minimising suffering to animals used for scientific purposes. Internationally, as these principles became embedded in the various systems of oversight for the use of animals in science, attention focused on how to minimise pain, distress and lasting harm to animals while maximising the benefits to be obtained from the work. Suffering can arise from the experimental procedures, but it can also arise from the manner in which the animals are housed and cared for. Increased attention is therefore being paid to the entire lifetime experience of an animal, in order to afford it as good a quality of life as possible. Russell and Burch were also concerned that animals should not be used if alternatives to such use were available, and that animals were not wasted through poor- quality science. This concept is being revisited through new efforts to ensure that experiments are well designed and properly reported in the literature, that all results – positive, negative or neutral – are made available to ensure a complete research record, and that animal models are properly evaluated through periodic systematic reviews. These efforts should ensure that animal use is truly reduced as far as possible and that the benefits derived through the use of animals truly outweigh the harms. Keywords Experimental design – Harm-benefit analysis – Meta-analysis – Publication standard – Reduction – Refinement – Systematic review – Three Rs. Introduction When Russell and Burch wrote the Principles of Humane Experimental Technique in 1959, espousing the tenet of the Three Rs (Replacement, Reduction and Refinement) (1), their primary focus was on minimising the harms experienced by experimental animals. In the interim fifty-plus years, the Three Rs have been interpreted and re-interpreted by many. While there are researchers who remain unclear about the definition of each of the Three Rs (2, 3), the general concept, as part of a framework for the ethics of animal experimentation, is now included in most legislated and non-legislated systems of oversight (4). Adherence to the Three Rs can ensure that animals are used only when absolutely necessary, that the numbers of animals used are kept to a minimum, and that suffering is minimised. Other issues concerning the ethical use of animals in science, not directly addressed by the Three Rs, include: considerations of whether the use of an animal is acceptable to the public; whether animal model data can actually be translated to the end goal (for example, human treatment, conservation, toxicity assessment); and whether the suffering likely to be experienced by the animal during the project is too great, irrespective of any potential gains. While these concerns cannot be addressed directly by considering the Three Rs, there may be potential for ‘Three