Derry, H.M., Jaremka, L.M., Bennett, J.M., Peng, J., Andridge, R., Shapiro, C., . . . Kiecolt-Glaser, J.K. (2014). Yoga and self-reported cognitive problems in breast cancer survivors: A randomized controlled trial. Psycho-Oncology. Advance online publication. 

To determine the effects of yoga on self-reported cognitive function

Certified yoga instructors led two 90-minute yoga group classes per week for 12 weeks. Each group contained 4–20 participants. Hatha yoga poses targeting relaxation, mindful breathing, chest opening, spinal extension, upper-body strength, and mobility were predetermined for each of the 24 sessions, and protocol drift was monitored. Pamphlets describing the poses from the classes and a yoga DVD were provided for home practice. Participants recorded weekly home and class time to determine dose effects. Participants who missed a class were called to improve adherence. Waitlist control participants continued regular activities and were directed not to begin yoga practice until after their final assessments. Measures were assessed before the behavioral intervention, immediately after the 12-week intervention, and three months after the intervention.

• N = 200  
• MEAN AGE = 51.6 years (SD = 9.2 years) 
• FEMALES (%): 100
• KEY DISEASE CHARACTERISTICS: Breast cancer survivors in stages 0-IIIA between two months and three years after the completion of primary and adjuvant treatment (except for antiestrogen therapy).
• OTHER KEY SAMPLE CHARACTERISTICS: The sample was an average of 17.3 (SD = 8.1) months since diagnosis and 10.9 ( SD = 7.9) months since treatment. The sample was highly educated; 69.5% of participants were college graduates or had postgraduate education. 81% of participants were postmenopausal.

• SITE: Single-site    
• SETTING TYPE: Multiple settings    
• LOCATION: Ohio State University Cancer Center in Columbus, Ohio, a yoga studio, and participants’ homes

• PHASE OF CARE: Late effects and survivorship

Randomized, waitlist-controlled trial

• Breast Cancer Prevention Trial (BCPT) Symptom Checklist–Cognitive Problems subscale
• Center for Epidemiological Studies–Depression (CES-D) scale
• Beck Anxiety Inventory (BAI)
• Pittsburgh Sleep Quality Index (PSQI)
• Medical Outcomes Study Short-Form survey (SF-36) Energy Subscale
• Community Healthy Activities Model Program for Seniors questionnaire (CHAMPS) (physical activity)
• Fasting blood samples for lipopolysaccharide-stimulated cytokine levels (IL-6, IL-1 beta, TNF-alpha)

At baseline, the average self-reported cognitive impairment was slight to moderate and did not differ between groups. Overall group and group-by-time effects were found (p < .05, both). Although no differences were found immediately after the 12-week intervention, the intervention group reported significantly less cognitive impairment than controls three months after the intervention ended (p < .01). However, these effects did not remain after controlling for symptom covariates (e.g., anxiety, depression, fatigue, sleep quality). Participants with more daily yoga practice (mean of 29 minutes) reported less cognitive impairment postintervention through three months (p = .011), and participants with less daily yoga practice (mean of 18 minutes) or no daily yoga practice did not report these changes. This dose response remained when controlling for symptom covariates. At three months, the intervention group reported more physical activity than controls (p = .032). Cytokine levels did not predict changes in self-reported cognitive impairment.

A 12-week, group Hatha yoga intervention provided by a trained instructor may improve self-reported cognitive impairment in breast cancer survivors. This effect may be mediated by improvements in co-occurring symptoms. Practicing yoga for at least 30 minutes per day may be required for a significant improvement over time in cognitive impairment.

• Risk of bias (no blinding)
• Risk of bias (no appropriate attentional control condition)
• Findings not generalizable
• Intervention expensive, impractical, or training needs
• Other limitations/explanation: Participants were not blinded to their group assignment. The nonspecific effects of social support or attention may have been responsible for improvements in symptoms given the lack of a control group. It was unknown what component of the intervention (e.g., physical activity, mindfulness) was active. The sample was mostly white and well-educated, limiting generalizability. The intervention would require training for facilitators to maintain fidelity. Cognitive function was measured only with self-report instruments. The duration of the intervention effect is unknown because of the lack of long-term (> 3 months) follow-up assessments.

A group Hatha yoga class delivered by a trained facilitator may improve cognitive impairment for breast cancer survivors. However, more research with a longer follow-up period is warranted to determine whether the intervention is effective, what component of the intervention is active, and whether yoga is practical for implementation in practice.