Responsiveness and clinical utility of the geriatric self-efficacy index for urinary incontinence

Abstract

Objectives: To report on the responsiveness testing and clinical utility of the 12-item Geriatric Self-Efficacy Index for Urinary Incontinence (GSE-UI).

Design: Prospective cohort study.

Setting: Six urinary incontinence (UI) outpatient clinics in Quebec, Canada.

Participants: Community-dwelling incontinent adults aged 65 and older.

Measurements: The abridged 12-item GSE-UI, measuring older adults' level of confidence for preventing urine loss, was administered to all new consecutive incontinent patients 1 week before their initial clinic visit, at baseline, and 3 months posttreatment. At follow-up, a positive rating of improvement in UI was ascertained from patients and their physicians using the Patient's and Clinician's Global Impression of Improvement scales, respectively. Responsiveness of the GSE-UI was calculated using Guyatt's change index. Its clinical utility was determined using receiver operating curves.

Results: Eighty-nine of 228 eligible patients (39.0%) participated (mean age 72.6+5.8, range 65-90). At 3-month follow-up, 22.5% of patients were very much better, and 41.6% were a little or much better. Guyatt's change index was 2.6 for patients who changed by a clinically meaningful amount and 1.5 for patients having experienced any level of improvement. An improvement of 14 points on the 12-item GSE-UI had a sensitivity of 75.1% and a specificity of 78.2% for detecting clinically meaningful changes in UI status. Mean GSE-UI scores varied according to improvement status (P<.001) and correlated with changes in quality-of-life scores (r=0.7, P<.001) and reductions in UI episodes (r=0.4, P=.004).

Conclusion: The GSE-UI is responsive and clinically useful.