"Unblinding" in randomized controlled drug trials for urinary incontinence: Implications for assessing outcomes when...

To determine whether women with urinary incontinence (UI) can identify their allocation in a randomized controlled trial (RCT) of tolterodine (TOL), and whether correct identification is associated with outcomes and adverse effects (AEs).

Exploratory analysis of a randomized, double-blind, placebo (PLC)-controlled trial of TOL 4 mg daily for 8 weeks in 743 women with urge-predominant mixed UI. Patient perception of their randomization was assessed at trial end. Main outcome measures were 7-day bladder diaries, patient perception of improvement, and UI-specific quality of life (QoL).

TOL produced a significant decrease in urge UI episodes compared to PLC (78% vs. 51%, P = 0.0001). Fifty-one percent of women correctly identified their randomization (58% on TOL vs. 37% on PLC, P < 0.001). Women who assumed they took TOL had better bladder diary outcomes than those who assumed they took PLC. Within each assumption group, patient perception outcomes were similar, regardless of actual randomization. QoL improved in all domains except general health for women on TOL. In women who assumed they took TOL, significant drug benefit was evident in three domains. Moderate-severe dry mouth was higher in those who assumed they took TOL (7.3% vs. 0%, P < 0.0001).

Greater than fifty percent of women in this RCT of antimuscarinic treatment were "unblinded" to their randomization. Patient assumption of randomization was associated with bladder diary and perception outcomes, specific QoL domains, and dry mouth. Efficacy of urge incontinence drugs should be considered in the context of patient assumptions, expectations, and "unblinding" by easily evident side effects.