Predictors of yoga use among patients with breast cancer.

OBJECTIVE

Emerging research suggests that yoga may be beneficial for reducing symptoms and improving quality of life among breast cancer patients. However, very little is known about the characteristics of breast cancer patients who use yoga; thus, this study seeks to identify the sociodemographic and clinical characteristics of yoga users among this population.

DESIGN

A cross-sectional survey study was conducted.

SETTING

The study was conducted at an outpatient breast oncology clinic at a large university hospital.

PARTICIPANTS

Three hundred postmenopausal breast cancer patients currently receiving aromatase inhibitors were included in this study.

MAIN OUTCOME MEASUREMENT: Self-reported use of yoga following the cancer diagnosis was collected along with sociodemographic and clinical data. Multivariate logistic regression was used to identify independent predictors of yoga use among breast cancer patients.

RESULTS

Of 300 participants, 53 (17.7%) reported having used yoga following cancer diagnosis. White patients were significantly more likely to use yoga than nonwhite patients (P = .02). Higher education level, lower BMI (body mass index), part-time employment status, previous chemotherapy, and radiation therapy were all associated with greater yoga use (all P < .05). Controlling for other factors, greater yoga use was independently associated with higher education level (adjusted odds ratio [AOR] 2.72, 95% confidence interval [CI], 1.15-6.46), and lower BMI (AOR 0.25, 95% CI, 0.09-0.66).

CONCLUSION

Yoga use following breast cancer diagnosis was substantially higher for white patients and those with lower BMI and higher education levels. Considering its potential benefits for symptom management in cancer, more research is needed to understand the attitudes and barriers to yoga use among individuals with nonwhite race, lower education, and higher BMI level. Such investigation will help design yoga programs that are aligned to the needs of these populations.