Testing Treatments: Better research for better healthcare
British Library Publishing Division, London, £12.95, 110pp
ISBN 0 7123 4909 X
There are numerous scientists who believe that more systematic reviews of animal studies would be desirable. These could contribute to validating animal models, understanding the limitations of animal studies, reducing publication bias, improving experimental design, and critically appraising the results. The ultimate benefit would be to derive the best information from those studies relevant to human medical advancement.
Much of the experience so far has come from systematic reviews of clinical studies. A book called Testing Treatments explains just why such reviews are so important.
The book gives an excellent overview of what constitutes good scientific evidence. It is written in plain English and avoids complex jargon. It is also a fascinating read, packed with revealing anecdotes and important historical lessons. The book describes how better research can ultimately improve clinical practice. The aim is to explain the difference between personal experience and more complex, but less biased, ways of distinguishing between what works and what does not. In other words, Testing Treatments supports good scientific practice and evidence-based medicine.
Whilst the focus is on clinical studies and their application to clinical practice, the importance of reviewing systematically the relevant evidence from animal research is also mentioned. The authors point out, for example, that had the results of animal experiments been reviewed, clinical trials of the drug nimodipine in stroke patients would never have been done. When the collated evidence of clinical trials involving nearly 8,000 patients was reviewed systematically in 1999, no overall beneficial effect of the drug was found. In the light of these results, the findings from animal experiments were systematically reviewed for the first time and it became clear that they were dubious at best.
The book is sceptical about some aspects of basic research, including animal research, and in particular about the volume of such research that is carried out. The authors claim that there is insufficient evidence of the benefits to justify the funding bias towards basic research. Unfortunately, at this stage the references become a bit thin and the narrative passes on too quickly for a satisfying analysis. Nonetheless, the book represents an important challenge for those of us in the scientific community who believe animal studies are important.