To test the validity of three patient global ratings, satisfaction, perception of improvement, and estimated percent improvement, for measuring outcomes of behavioral treatment for urinary incontinence.
This report is a secondary analysis of data from three randomized controlled trials testing behavioral interventions for incontinence. Participants were 359 community-dwelling women, aged 40-92 years, with stress, urge, or mixed urinary incontinence. All participants received an 8-week program of clinic-based or self-administered behavioral training. Subjective outcomes included a patient satisfaction question (PSQ), global perception of improvement (GPI), and estimated percent improvement (EPI). Convergent validity was tested by examining the relationship between each measure and reduction of incontinence (bladder diary), change on the incontinence impact questionnaire (IIQ), and desire for another treatment. Discriminant validity was explored by examining the relationship of the global ratings to five measures not expected to be related to outcome (age, race, BMI, education level, and change in perceived pain).
All three patient global ratings were significantly associated with each other (P < 0.0001), with diary measures of reduction of incontinence episodes (P < 0.0001), and change in the IIQ (P < 0.005), and inversely associated with desire for another treatment (P < 0.0001). All three patient ratings were not significantly associated with age, race, BMI, education level, or change in perceived pain.
Patient global ratings of satisfaction, perception of improvement, and estimated percent improvement have acceptable convergent and discriminant validity for measuring outcomes in studies of behavioral treatment for urinary incontinence.