In this study, patients receiving soft-tissue manipulation scored better on self-rated scales of pain, anxiety and function than controls. However, there is no way to know whether this was due to the nonspecific effects of being touched by a caregiver or to particular aspects of the intervention that were unique to massage therapy. Sham massage may have been a more appropriate control.

Another issue that weakens the conclusions of this paper is that of patient recruitment. Patients volunteering for a study of massage therapy may be predisposed to have faith in its tenets or have pre-existing expectations of its benefits. This is especially problematic in a study in which patients were not blinded to the type of treatment administered.

Preyde states that massage improved patient function. It would be more accurate to say that those receiving massage perceived their function to be improved. Unfortunately, this is a perception very prone to nonspecific provider influences.

For the reasons noted above, the self-rating scales used in this trial provide less than robust information. This concern is highlighted by the finding that lumbar range of motion was not different between groups. This was the only objective measure and the only one for which blinded evaluators were used. As such, this paper's most powerful findings indicate a lack of effect for massage therapy when compared with nonmassage controls.