NEW YORK NURSE: July/August 2010
by Rona F. Levin, PhD, RN, Professor, Lienhard School of Nursing, Pace University
Clinical practitioners from many healthcare professions use complementary or alternative treatments to achieve positive health outcomes. Massage is one of these treatments. Nurses administer massage every day in one form or another to promote patient comfort. I learned to incorporate massage as part of my care routine 45 years ago in my fundamentals of nursing course. The question of how different types of massage may affect clinical outcomes in different populations, however, is one that needs to be asked and answered with high quality evidence. Wilkinson, Barnes and Story (2008) attempted to answer this question for a specific population of patients by conducting a systematic review to “assess evidence of the effectiveness of massage for patients with cancer”(p. 430). The review focused on four outcome measures: physical symptoms, psychological symptoms, quality of life and adverse side effects of treatment. The population of patients consisted of cancer patients receiving some type of massage in a healthcare setting.
The authors followed the process for a systematic review developed and used by the Cochrane Collaboration, one internationally accepted standard for conducting these types of evidence reviews. This means we can have confidence that the authors used sound methods for conducting their study. A systematic review is a rigorous, comprehensive, critical summary of evidence on a specific topic or question of interest. There are two major types of systematic reviews, narrative and quantitative. The former is often called an integrative review, while the latter is also called a meta-analysis (Levin, 2009). Because the studies that Wilkinson et al. (2008) reviewed were too dissimilar or heterogeneous to combine statistics from individual studies, the authors conducted an integrative review of the evidence.
To begin Wilkinson et al. (2008) searched top databases to find published articles on their topic. In addition, they searched for unpublished works. Including unpublished works in systematic reviews is important because using only published research may create a bias in the findings. This bias can occur because published works often support hypotheses and/or show results in support of an intervention. Articles with non-significant results, for example, are often not as desirable for many journals, but certainly are important sources of evidence to consider.
Although Wilkinson et al. (2008) initially retrieved 1,435 papers from their systematic search of literature, only ten met their criteria to be included in the review. The criteria for inclusion were:
After agreeing on the studies to be included in the review, at least two authors independently assessed each of the ten studies for quality. They used standardized criteria that included, among other things, how the patients were assigned to experimental and control groups in the randomized controlled trials and whether or not patients and practitioners knew to which group patients were assigned. Of the ten studies included, data was presented for only nine of the studies. Seven of the studies compared massage with no massage and two of the studies compared massage using aromatic oils with carrier oil massage.
Different types of massages were used in the studies and included: Swedish/Esalen massage, acupressure, back massage, slow stroke massage, Swedish back massage, and foot massage. In addition, the different types of massages were applied to different body parts. This heterogeneity makes comparisons difficult. Also, the quality of methods varied considerably among the ten studies reviewed.
Putting the results in the above context, the authors found that some studies demonstrated improvement in pain and nausea, but the results of trials investigating psychological symptoms as outcome measures were mixed. Some studies found improvements in anxiety and other psychological symptoms. Considering quality of life and depression, results were also mixed, with two studies finding positive results for these variables and two studies finding no difference between groups who received massage and those who did not. Of note are the side effects observed. One patient developed a skin rash in one study and another study noted a higher incidence of digestive problems in the essential oil massage intervention group compared with the no-massage group.
The authors concluded that because of the mixed results of the studies reviewed and because of the relatively poor quality of the studies, definitive conclusions cannot be reached at this time. More research needs to be done to answer the questions related to the benefits of massage therapy for cancer patients. Because of the widespread use of massage based on anecdotal evidence of positive outcomes, the authors caution practitioners to carefully observe patients for signs of dermatologic or digestive side effects when receiving massage. Because the digestive symptoms were noted in patients massaged with aromatic oils, perhaps this type of massage should be avoided until further evidence is accrued. Also, the ingredients of the lotions used in massage therapy need to be monitored. In daily practice, nurses often provide massage as part of routine care with rare adverse effects. The question that remains is: do the risks of providing routine massage without aromatic oils outweigh the benefits for most patients?
Levin, R. F. (2009a). Translating research evidence for WCET practice: Appraising a systematic review. World Council of Enterostomal Therapists Journal, 29(2), 39-39-40.
Wilkinson, S., Barnes, K., & Storey, L. (2008). Massage for symptom relief in patients with cancer: systematic review. Journal of Advanced Nursing 63(5), 430–439. doi: 10.1111/j.1365-2648.2008.04712.x