Managing Conflicts of Interest in Practice Guidelines Panels-Reply.

ceived “a steroid injection in the parasacrococygeal region,” which makes it unclear whether SMT or the steroids were responsible for any observed effects. The statistical combination of such heterogeneous studies may be hard to justify from a clinical point of view. Furthermore, although the authors downgraded the quality of evidence from high to moderate due to inconsistency of results according to Grading of Recommendations Assessment, Development, and Evaluation (GRADE), risk of bias was not adequately taken into account. For example, for the outcome of pain in studies comparing SMT with other therapies, 50% of studies were considered to have a high risk of bias, which seems like low-quality evidence. Figure 2 of this systematic review2 also indicates that SMT was not superior to sham therapies regarding the outcome of pain and the quality of evidence was considered moderate at best. A more appropriate conclusion may be that there was moderate-quality evidence that SMT was nonsuperior to sham and low-quality evidence that it was superior to other therapies.