Research: Can Massage Therapy Help Address Shoulder Pain

A meta-analysis examines the effectiveness of massage therapy on shoulder pain.

By Martha Brown Menard, Ph.D., LMT, November 15, 2019

Person holding sleeping infant fact-to-face smiling

The Study Question

Shoulder pain is among the most common musculoskeletal disorders. Massage therapy is widely used by physical therapists for the treatment of shoulder pain. This 2017 study is the first systematic review and meta-analysis to examine both the short-term and long-term effectiveness of massage therapy for shoulder pain. The same author previously published a companion systematic review and meta-analysis of the effectiveness of massage therapy for shoulder range of motion.

The Study Methods

Databases searched included PubMed, CINAHL, Embase, PsycINFO, RISS, NDSL, NANET, DBpia and KoreaMed, with languages limited to English and Korean. Keywords used for the search were shoulder pain, shoulder impingement syndrome, rotator cuff, bursitis, adhesive capsulitis, massage, therapeutic touch, reflexotherapy, reflexion, manual, manipulative, clinical trial, random and placebo. Exclusion criteria included studies of patients with infection, neoplasm, fracture, instability, dislocation, hemiplegia or postoperative shoulder pain. Studies without reported means and standard deviations or that did not evaluate an intervention were also excluded.

The Results

The search identified 985 studies. Of those, 71 articles were potentially eligible for inclusion; however, only 15 studies met the criteria. Eleven of the studies were randomized controlled trials. The meta-analysis of the 15 studies included a total of 635 participants, with 340 participants in the experimental group and 295 in the control group. The control group was further subdivided into active and inactive therapies. Five studies assessed the follow-up effect of massage. Studies employing both Western and Asian massage approaches were included.

The effect size for short-term efficacy was large: −1.08. In the subgroup analysis comparing inactive therapies to massage (11 studies), a large effect size was also observed: −1.12. Four studies compared massage to active therapies, with a large effect size also seen: −1.06. A moderate effect size for the follow-up effect of massage was: −0.47.

Based on the subgroup analysis, the effect size of massage therapy was larger compared to no treatment or placebo treatments. However, there were no significant differences in comparisons with the effect sizes of other active treatments such as physical therapy, rest intervention and acupuncture.

Limitations of the Study

This study is a good example of the limits of meta-analysis applied to a discipline as diverse as massage therapy, which relies on an individualized approach tailored to the client in order to be clinically effective.

The length of massage treatment varied considerably across studies, from five days to 12 weeks. The duration of sessions varied widely, as well, from 10 to 45 minutes.

The types of massage employed were not described in any depth other than labeling them as sports massage or deep-tissue massage, for example. Some of the included studies had small sample sizes and were considered low-quality trials. A systematic review of the literature is still needed to further identify types or techniques of massage that are most effective in relieving shoulder pain due to specific conditions that can be responsible for shoulder pain, such as rotator cuff syndrome or adhesive capsulitis.

Implications for Evidence-Informed Practice

Because the lifetime prevalence of shoulder pain has been estimated to be as high as 66 percent, it is a common complaint that massage therapists are likely to encounter. This meta-analysis supports the use of massage therapy as broadly defined, and it appears to be as effective as other types of active treatments, including physical therapy and acupuncture.


1.Yeun YR. “Effectiveness of massage therapy for shoulder pain: a systematic review and meta-analysis.” J Phys Ther Sci. 2017 May; 29(5): 936–940.