Bardia, A., Barton, D.L., Prokop, L.J., Bauer, B.A., & Moynihan, T.J. (2006). Efficacy of complementary and alternative medicine therapies in relieving cancer pain: A systematic review. Journal of Clinical Oncology, 24, 5457–5464.

To evaluate the efficacy of various complementary and alternative medicine (CAM) therapies to reduce cancer pain

The type of article is systematic review.

• Databases searched were MEDLINE, EMBASE, CINAHL, Allied and Complementary Medicine (AMED), and the Cochrane Library up to August 2005. Investigators also searched reference lists from articles to identify relevant studies.
• Search keywords were cancer, pain and alternative medicine, and neoplasm, as well as terms for major individual CAM therapies.
• Studies were included in the review if they were randomized clinical trials (RCTs) that had a CAM intervention for cancer pain.
• Exclusion criteria were not cited.

The initial search identified 101 articles, of which investigators excluded 85. Investigators included an additional two articles, which were found through manual scans of reference lists. Investigators appraised articles by using the Jadad scale.

• The final sample included 18 studies reporting on 1,499 patients.
• The average sample included less than 100 patients; median sample size was 54 patients.

• Acupuncture: Investigators evaluated three trials, of which one was a high-quality study that used auricular acupuncture to treat 90 patients. Compared to the placebo group, the treatment group had a significant decrease in pain intensity that lasted two months.
• Mind-body interventions: Investigators assessed five trials. Two of these were trials of intermediate quality in which support groups were effective in decreasing pain. Interventions included group supportive psychotherapy, hypnosis, and support groups. Two trials concluded that relaxation and imagery were effective in reducing pain. Researchers obtained no greater relief by adding cognitive behavioral therapy to relaxation or imagery.
• Music: Assessment revealed that the three trials were of poor quality. Results were mixed. Effects measured were of very short duration.
• Herbal mixtures: Two trials examined different herbal supplements. Both trials were of poor quality, and one did not report actual statistical results. Authors could draw no meaningful conclusions from this information.
• Massage therapy: The four trials involving massage therapy had mixed results in terms of efficacy. Most of these trials had very small sample sizes, and none found intervention effects that lasted longer than four weeks.
• Healing touch or Reiki: Two trials involved these interventions. Results did not warrant conclusions about effectiveness.

The result of this systematic review was that none of these interventions can be recommended as effective. The most promising therapies appear to be mind-body interventions. In particular, hypnosis and relaxation might have some effect in decreasing cancer pain. Effect may be limited by cognitive impairment caused by cancer or cancer treatment. Support groups may have some positive effects. Whether these benefits are due to increased patient awareness, with more frequent visits to care providers, better compliance with medication regimens, or as a result of group interaction and social or emotional support is unclear. One study found that benefit occurred only in those patients who were more distressed at baseline. It is possible that effect sizes of therapies may be more discernible among patients with higher pain ratings, demonstrating a floor effect related to the symptom of pain.

• Most trials were of low quality, with missing data and incomplete reporting. This fact made extracting sufficient usable information difficult.
• Few studies in the analysis examined a specific intervention.

Current evidence does not support the efficacy of the cited CAM interventions in the management of pain of patients with cancer. Methodologically strong research that incorporates appropriate attentional and sham controls, sufficient sample sizes, and longer duration of follow-up is needed.