This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
In a comparison of numerous clinical trials, some simplification is necessary for presentation. However, I do not believe that the article by Coxhead and colleagues was in any way misrepresented.The issue of blinded outcome assessments is never irrelevant in determining therapeutic efficacy, because both investigators and subjects are likely to have biases regarding therapy. I pointed out, both in the text and in Table 4, that blinding is difficult with regard to physical treatments. However, a third-party assessor who is unaware of treatment group can evaluate outcomes, and this should be less biased then self-assessment by a patient with full knowledge of his therapy. Ideally, the patients who did not receive a given therapy should receive sham treatments of a similar nature (as in drug trials), to create the impression of active intervention. Although I cited trials that employed manipulation and corsets in which this was done,
Deyo RA. Treatment of Low-Back Pain-Reply. JAMA. 1985;253(8):1122-1123. doi:10.1001/jama.1985.03350320042013