Harder, H., Parlour, L., & Jenkins, V. (2012). Randomised controlled trials of yoga interventions for women with breast cancer: A systematic literature review. Supportive Care in Cancer, 20, 3055-3064.

STUDY PURPOSE: To examine physical and psychological benefits of yoga interventions in women with breast cancer

TYPE OF STUDY:  Systematic Review

DATABASES USED: MEDLINE, PsychINFO, the Cochrane Library, Embase, CINAHL, AMED, Web of Science, and Scopus

KEYWORDS: Yoga, breast cancer, and breast neoplasm

INCLUSION CRITERIA: Yoga intervention; women with breast cancer; a randomized controlled trial design (RCT); the studies were original full reports; and the studies were published in peer-reviewed journals.

EXCLUSION CRITERIA: Studies that investigated complementary and alternative medicines or exercise interventions; conference abstracts

TOTAL REFERENCES RETRIEVED: Eighteen RCTs met the inclusion criteria out of 274 initial data. The 274 initial articles returned were reduced to 132 after duplicates were removed. Further reductions occurred due to multiple publications of the same data or the same outcome measures; same studies or continuation of same studies also were removed from review.

EVALUATION METHOD AND COMMENTS ON LITERATURE USED: Physiotherapy Evidence Database (PEDro Scale) was used to rate methodological quality of RCTs. It is a 10-item scoring system that evaluates internal validity (random allocation; concealment of allocation; similarity of groups at baseline; blinding of participants, therapists, and assessors; adequate follow-up and undertaking an intention-to-treat analysis) and statistical information. A total score below 4 was considered to be of “poor” methodological quality; between 4 and 5 was considered to be of “fair” quality; 6 to 8 was considered to be of “good” quality; and 9 or 10 was considered to be of “excellent” quality. Two reviewers independently rated each study.

• FINAL NUMBER STUDIES INCLUDED: N = 18
• SAMPLE RANGE ACROSS STUDIES: The sample size range was 18–164 at baseline to 14–75 at follow-up.
• TOTAL PATIENTS INCLUDED IN REVIEW: N = 760
• KEY SAMPLE CHARACTERISTICS: The mean age (based on 10 studies adequately reporting age) was 52.7 years (mean age range = 45–62.9 years). Most studies investigated women with early or advanced stage disease; three included women with noninvasive breast cancers. Seventeen studies conducted repeated measures at a minimum of two time points (pre- and post-intervention) using an adequate baseline assessment performed before or after randomization (though not all studies reported change scores). Follow-up occurred over a range of one to six months.

PHASE OF CARE: Active antitumor treatment and transition phase after active treatment. Twelve studies were conducted during treatment; six were conducted post-treatment (two months to six years); two were conducted during mixed time periods during and after treatment (mean time since diagnosis or treatment = 1.7–6.5 years).     

APPLICATIONS: Elder care and palliative care

• The most common intervention was integrated yoga (consisting of postures, breathing, and meditation) and Iyengar and Hatha yoga (71 %) given via self-practice (83%) or in-class lessons. The duration was 4 to 12 weeks (median 8 weeks).
• The primary outcome variables of the yoga intervention were (1) mood and psychosocial functioning (depression, anxiety, stress, and psychological symptom distress); (2) health-related quality of life; (3) fatigue; and (4) biological changes and physical measures (i.e., wound healing, hospital stay, TNF-alpha, immunoglobulin, nausea and vomiting, and overweight).
• Overall, all 18 studies in the review reported positive effects from the yoga interventions, with the greatest impact on global QOL scores and emotional well-being. Few in the yoga program experienced improved cancer-related fatigue. Biological measures varied, and conclusions for this outcome cannot be drawn.
• Total quality rating scores for the RCTs was a median of 6, indicating that overall the quality was “good” (range 1 to 8); one study was rated methodologically poor (score 1). Low quality was found in the description of the randomization process (i.e., concealed allocation), blinding (i.e., blinding of assessors), and reporting of adequate follow-up (i.e., > 85% of subjects).
• Adherence was a major problem of the intervention.

• Qualities measures were used to evaluate studies. Overall study quality appears to be good, and the studies relatively consistently reported that yoga may be a useful practice.  
• However, long-term and specific objective effects of yoga interventions need to be further examined. Outcome variables in this review varied across studies. Only seven studies used validated depression measures, and only two studies used a validated anxiety measure. Among them, six studies reported positive effects from the intervention on depression and/or anxiety, whereas two studies reported no effect.
• The intervention program ranged from 6 to 26 weeks with up to three sessions of yoga per week and were generally well received and safe. Yet, more safety data are required to report that yoga is not harmful and is a credible intervention compared to conventional therapies. More economical and practical information also is needed to implement yoga.

This review does not specifically focus on depression and anxiety. Only studies with patients with breast cancer were included for this review. Thus, only several studies with depression or anxiety as outcome variables were included in the final review. None of the studies were found to have excellent design (e.g., small sample size and lack of long-term follow-up).

The intervention may be beneficial, yet its specific effect on depression and anxiety should be further examined. Also, the intense, duration, and practical issues (e.g., who provided the intervention, who paid the cost) should be considered. Nurses can conduct large-sample, long-term studies of the efficacy of yoga using instruments that measure change scores and calculating sufficient power to detect group differences.