Systematic review of spinal manipulation: A biased report

As Fellows of The Royal Society of Medicine, and also currently the President and Immediate Past President of the British Chiropractic Association, we wish to comment on the paper by Ernst and Canter (April 2006, JRSM1). The authors clearly demonstrate bias in the very carefully selected studies included in their review. It appears to us that these researchers wished to examine a number of rather obscure areas of investigation (infantile colic, asthma, dysmenorrhoea, etc.) that cannot fairly reflect the efficacy or efficiency of manipulation—and there are many good quality primary research papers available for critique that deal adequately with the major focus of mainstream manipulation.2,3 It would, of course, have been a more accurate and impartial review if they had concentrated on those elements where the outcomes are well documented—namely, low back pain syndromes.4,5 Indeed, this recent review included four `reviews of reviews' by the main author; this in itself would surely perpetuate any author bias, whether intentional or not. In fact, the Centre for Reviews and Dissemination at the University of York concluded that it was difficult to assess the methodological robustness of the reviews utilized or the quality or the results of the primary studies. It is obvious that the CRD will not be alone in their conclusion. However, our own profession will continue to strive for the highest standards in research, education and practice, where patients' best interests will always remain paramount. At the end of the day, no matter how robust the methodology of a critical review, the authors are inclined to their own bias: to conduct a review of reviews of one's own opinions, adds little to a meaningful literature base.