Cepeda, M.S., Carr, D.B., Lau, J., & Alvarez, H. (2010). Music for pain relief. Cochrane Database of Systematic Reviews 2010(8).

To evaluate the effect of listening to music on acute or chronic cancer pain; to relate the effect of listening to music on analgesic requirements

• Databases searched were the Cochrane Pain, Palliative & Supportive Care Group Trials Register and the Cochrane Central Register of Controlled Trials (Sept. 14, 2004); MEDLINE (1966–Oct. 4, 2004); EMBASE (1980–Sept. 15, 2004); PsycINFO (1985–Sept. 24, 2004); and LILACS (1982–Sept. 8, 2004).
• Search keywords included free text, MeSH, and Emtree terms: music, music therapy, analges*, and pain. The report provides an extensive list of search terms.
• Studies were included if they
  • Were randomized controlled trials.
  • Evaluated the effect of music on acute, chronic neuropathic cancer pain or experimental pain in children or adults.
  • Involved evaluation of pain intensity, pain relief, global improvement, or opioid requirement.
• Studies were excluded if a music intervention was combined with any nonpharmaceutical approach to pain.

• Investigators retrieved 113 studies, 62 of which the investigators excluded.
• Authors assessed study quality by scoring subject assignment, study blinding, dropout rate, and intention-to-treat analysis.

• The final sample included 51 studies.
• The 51 studies comprised 1,867 patients who received a music intervention and 1,796 control patients, who did not.
• Eight studies evaluated the use of music to treat the pain of children and neonates.
• The range of patients in a sample was 11–233. Median sample size was 53.
• Samples included patients with chronic pain, acute pain, postoperative pain, and labor pain.

• Pain intensity: Meta-analysis involved 30 studies that used the same scale of pain intensity. The pooled estimate of effect was a 0.46 reduction on a 0–10 point scale (95% CI = –0.75 through –0.17, not significant). The studies comprised significant heterogeneity, and authors' analysis showed that one study had a major influence on overall effect size. When this study was removed, the effect of music was deleterious, being associated with a pain increase of 0.3.
• Studies of children: In the eight studies of children, four studies did not provide quantitative data. Overall, no effect on pain proved significant, though some positive results were reported.
• Pain relief: Four studies showed that 70% of those who used music had a greater probability of at least a 50% reduction in pain, relative risk (RR) – 1.70 (95% CI 1.21–2.37, p = 0.002).
• Opioid requirement: Thirteen studies evaluated differences in opioid requirements. These studies were of patients who had procedural or postoperative pain. The mean difference (MD) supported use of the music intervention: –1.29, (95% CI –2.22 through –0.37, p = 0.006)
• Authors judged that half the analyzed studies were of low quality. Authors noted a high level of heterogeneity in most subgroup analyses.

Music has limited utility in clinical practice for pain reduction: A music intervention was associated with minimal reduction in pain intensity. A music intervention was associated with a small reduction in opioid use for the treatment of acute pain. The reduction was smaller than that associated with the use of an NSAID or paracetamol.

The analysis provided little support for the effectiveness of music in the reduction of pain: The size of effects is small, and their clinical relevance is unclear. On the other hand, the analysis revealed no negative effects from the music intervention. Clinicians should be aware of the limited utility of music for pain management. Most of the studies in the analysis did not include patients with cancer; however, the highest effect sizes were in the setting of chronic pain, so a music intervention may be relevant to patients with cancer. Use of music along with other nonpharmaceutical and maximal pharmaceutical pain management may be helpful for some patients. Further research in this area would be useful.