EDITOR—Ernst and Assendelft's editorial on chiropractic for low back pain seems to have been written more in a spirit of professional aversion than in one of critical doubt.1 This impression is conditioned by previous commentaries by these authors in the popular press and the biomedical literature.

The question is, why? There is substantial scientific evidence that the manipulation that chiropractors (and indeed osteopaths and some physiotherapists) do for back pain is both effective and safe. This evidence has been reviewed by multidisciplinary panels of experts in both the United Kingdom and the United States, which has resulted in the production of two national clinical practice guidelines for acute back pain that totally disagree with these authors. The only randomised controlled trial of overall chiropractic management for back pain, 23in contrast to manipulation alone, is not mentioned in this editorial. Yet this trial (included erroneously by one of these authors in 1991in a review of manipulation trials) was ranked as high quality, was positive in its evidence for chiropractic management, and yet was subsequently condemned as seriously flawed by Ernst in a separate paper. This editorial is equally contradictory.

No one would dispute the need to research further the evidence for the effectiveness, cost effectiveness, and safety of manipulation and associated treatment approaches. The Medical Research Council is currently supporting a large randomised trial by a multidisciplinary research team led by the department of health sciences and clinical evaluation at the University of York. Many other studies are in progress. Nevertheless, the United Kingdom's current national clinical practice guideline and evidence review states: “Within the first 6weeks of acute or recurrent low back pain, manipulation provides better short-term improvement in pain and activity levels and higher patient satisfaction than the treatments to which it has been compared” and …