The objective of this study was to evaluate the effectiveness of duloxetine in improving quality of life among women with stress and mixed urinary incontinence. The study included 451 women with self-reported stress incontinence episodes (>or=1/week) who were randomized to duloxetine (40 mg BID) or placebo in a double-blind, usual care design. Patients and physicians were allowed to titrate, augment, and/or discontinue treatment. Concomitant treatments were permitted. The primary outcome was the Incontinence Quality of Life Questionnaire (I-QOL) score, with assessments at 3, 6, and 9 months. Other measures included the Patient Global Impression of Improvement (PGI-I) and adverse events. The adjusted mean change in I-QOL total score was greater in the duloxetine group than in the placebo group and at a level comparable to that found in previous clinical trials, but the difference between placebo and duloxetine was not statistically significant in the intent-to-treat, last observation carried forward (LOCF) analysis. The difference approached statistical significance in favor of duloxetine at 3 months (p=0.07). PGI-I ratings did not demonstrate significant superiority for duloxetine in LOCF analysis; however, study completers taking duloxetine were significantly more likely to rate themselves as "better" (70.2%) than completers taking placebo (50.8%, p<0.05). Women utilized a variety of treatment methods including pelvic floor muscle training, estrogen, anticholinergic medication, weight reduction, and smoking cessation. In this study, while mean I-QOL change scores were numerically higher for the duloxetine group than mean change scores for the placebo group, this difference was not statistically significant. Among women who completed the study on study drug, a significantly greater proportion of duloxetine women versus placebo women rated their condition to be better.