ABSTRACT: BACKGROUND:Because of the limitations of existing clinical treatments for urinary incontinence, many women with incontinence are interested in complementary strategies for managing their symptoms. Yoga has been recommended as a behavioral self-management strategy for incontinence, but evidence of its feasibility, tolerability, and efficacy is lacking. OBJECTIVE:To evaluate the feasibility and tolerability of a group-based therapeutic yoga program for ambulatory middle-aged and older women with incontinence, and to examine preliminary changes in incontinence frequency as the primary efficacy outcome after 3 months. MATERIALS AND METHODS:Ambulatory women aged 50 years or older who reported at least daily stress-, urgency-, or mixed-type incontinence, were not already engaged in yoga, and were willing to temporarily forgo clinical incontinence treatments were recruited into a randomized trial in the San Francisco Bay area. Women were randomly assigned to take part in a program of twice-weekly group classes and once-weekly home practice focused on Iyengar-based yoga techniques selected by an expert yoga panel (yoga group), or a nonspecific muscle stretching and strengthening program designed to provide a rigorous time-and-attention control (control group) for 3 months. All participants also received written, evidence-based information about behavioral incontinence self-management techniques (pelvic floor exercises, bladder training) consistent with usual first-line care. Incontinence frequency and type were assessed by validated voiding diaries. Analysis of covariance models examined within- and between-group changes in incontinence frequency as the primary efficacy outcome over 3 months. RESULTS:Of the 56 women randomized (28 to yoga, 28 to control), the mean age was 65.4 (±8.1) years (range, 55-83 years), the mean baseline incontinence frequency was 3.5 (±2.0) episodes/d, and 37 women (66%) had urgency-predominant incontinence. A total of 50 women completed their assigned 3-month intervention program (89%), including 27 in the yoga and 23 in the control group (P = .19). Of those, 24 (89%) in the yoga and 20 (87%) in the control group attended at least 80% of group classes. Over 3 months, total incontinence frequency decreased by an average of 76% from baseline in the yoga and 56% in the control group (P = .07 for between-group difference). Stress incontinence frequency also decreased by an average of 61% in the yoga group and 35% in controls (P = .045 for between-group difference), but changes in urgency incontinence frequency did not differ significantly between groups. A total of 48 nonserious adverse events were reported, including 23 in the yoga and 25 in the control group, but none were directly attributable to yoga or control program practice. CONCLUSION:Findings demonstrate the feasibility of recruiting and retaining incontinent women across the aging spectrum into a therapeutic yoga program, and provide preliminary evidence of reduction in total and stress-type incontinence frequency after 3 months of yoga practice. When taught with attention to women's clinical needs, yoga may offer a potential community-based behavioral self-management strategy for incontinence to enhance clinical treatment, although future research should assess whether yoga offers unique benefits for incontinence above and beyond other physical activity-based interventions.