Knight, A. 2011. Weighing the costs and benefits of animal experiments. Alternatives to Animal Experimentation [ALTEX] Proceedings of WC8, 289-294.
Most regulations governing animal experimentation require that the harms expected to be incurred by animal subjects should be balanced against the likely benefits of the project. Too often, however, expected human benefits are based on unrealistic assumptions. To critically assess the human clinical and toxicological utility of animal experimentation, the published literature was comprehensively surveyed to locate relevant systematic reviews. In only two of 20 reviews did the authors conclude that animal models were either significantly useful in contributing to the development of human clinical interventions or substantially consistent with clinical outcomes. Included were reviews examining the clinical utility of invasive chimpanzee experiments, of highly cited animal experiments published in leading scientific journals, and of animal experiments approved by ethics committees at least partly on the basis of specific claims that they were likely to lead to concrete advances in human healthcare. Seven additional reviews failed to demonstrate reliable predictivity of human toxicities such as carcinogenicity and teratogenicity. ... Results in animal models were frequently equivocal or inconsistent with human outcomes. When considering costs and benefits overall, one cannot reasonably conclude that the human benefits exceed the costs incurred by animals subjected to scientific procedures. On the contrary, the evidence indicates that actual human benefit is rarely sufficient to justify such costs. Despite this, deficiencies in the implementation of regulatory and policy requirements to replace, reduce, and refine animal use remain marked and widespread. A range of policy initiatives are warranted to address these deficiencies. ... Disturbingly, only 48.7% (37/76) of these highly-cited animal studies published in leading scientific journals were of good methodological quality. Few included random allocation of animals to treatment groups or blinded assessment of outcomes, although it is well established that studies lacking randomization or blinding often over-estimate the magnitude of treatment effects (Poignet et al., 1992; Aronowski et al., 1996; Marshall et al., 2000). Accordingly, Hackam and Redelmeier cautioned patients and physicians about extrapolating the findings of even highly-cited animal research to the care of human disease.