Abstract
OBJECTIVE: To estimate the effect of change in weight and change in urinary incontinence (UI) frequency on changes in preference-based measures of health-related quality of life (HRQL) among overweight and obese women with UI participating in a weight loss trial. METHODS: We conducted a longitudinal cohort analysis of 338 overweight and obese women with UI enrolled in a randomized clinical trial comparing a behavioral weight loss intervention to an educational control condition. At baseline, 6, and 18 months, health utilities were estimated using the Health Utilities Index Mark 3 (HUI3), a transformation of the SF-36 to the preference-based SF-6D, and the estimated Quality of Well-Being (eQWB) score (a summary calculated from the SF-36 physical functioning, mental health, bodily pain, general health perceptions, and role limitations-physical subscale scores). Potential predictors of changes in these outcomes were examined using generalized estimating equations. RESULTS: In adjusted multivariable models, weight loss was associated with improvement in HUI3, SF-6D, and eQWB at 6 and 18 months (P < 0.05). Increases in physical activity also were independently associated with improvement in HUI3 (P = 0.01) and SF-6D (P = 0.006) scores at 18 months. In contrast, reduction in UI frequency did not predict improvements in HRQL at 6 or 18 months. CONCLUSION: Weight loss and increased physical activity, but not reduction in UI frequency, were strongly associated with improvements in health utilities measured by the HUI3, SF-6D, and eQWB. These findings provide important information that can be used to inform cost-utility analyses of weight loss interventions.