I read the recent randomized controlled trial with great interest by Roseen and colleagues entitled “Yoga Versus Education for Veterans with Chronic Low Back Pain: a Randomized Controlled Trial.”1 In this study, the authors focused on the effectiveness of yoga as an alternative and popular method in treating chronic low back pain within the scope of physiotherapy and rehabilitation. In the study, yoga intervention was compared with a training program. The fact that the study’s sample consisted of Veterans is the prominent uniqueness of the study. At the end of the study, the authors reported that yoga had no additional contribution in terms of pain and disability. Although this study contributes explicitly to the literature, some issues regarding the methodology should be discussed further.

First, the study mentions only rater blinding with a single-blind design. In randomized controlled trials, rater blinding is valuable because it reduces the possibility of measurement bias.2 However, the study’s primary outcomes, “the modified Roland Morris Disability Questionnaire and pain on the Defense & Veterans Pain Rating Scale,” are patient-reported outcome measures. In other words, these questionnaires provide the outcome score with the direct statements of the individuals. Therefore, it should be noted that rater blindness has no methodological contribution.

Second, as reported in PEDro or CONSORT, equal physical and individual characteristics of the two groups are a requirement to prove that the groups are homogeneous.3 Although the authors presented both groups’ mean and standard deviation in the study, they should have proved that the groups were not different from each other with statistical significance tests through the p-value. In addition, there were more than two measurement points in the study. Therefore, post hoc analyses may have helped show which measurements differed statistically. For instance, while Table 2 clearly presents the 12-week outcomes, the statistical significance of the difference between weeks 0 and 24 or 12 and 24 can be determined by post hoc analyses regarding revealing within-group dynamics.4

Third, the study sample consisted of individuals with chronic low back pain. However, there is no evidence of how participants received the diagnosis of chronic low back pain during recruitment. Specific or non-specific causes can cause low back pain. Generally, individuals with low back pain for at least 6 months are considered to have chronic low back pain.5 No exclusion criteria exist for individuals with low back pain for more than 6 months. There are also no exclusion criteria for pain caused by spinal stenosis, lumbar disk herniation, or rheumatologic causes. Finally, the generalizability of the score could be more credible since the study included a single sample of older adults over 65 years and adults under 65 years of age. I would welcome the authors’ comments to address these issues, which will further provide additional information about their study.