Long-term prevalence and predictors of urinary incontinence among women in the Diabetes Prevention Program Outcomes Study

Abstract

Objectives: To examine the long-term prevalence and predictors of weekly urinary incontinence in the Diabetes Prevention Program Outcomes Study, a follow-up study of the Diabetes Prevention Program randomized clinical trial of overweight adults with impaired glucose tolerance.

Methods: This analysis included 1778 female participants of the Diabetes Prevention Program Outcomes Study who had been randomly assigned during the Diabetes Prevention Program to intensive lifestyle intervention (n = 582), metformin (n = 589) or placebo (n = 607). The study participants completed semi-annual assessments after the final Diabetes Prevention Program visit and for 6 years until October 2008.

Results: At the study entry, the prevalence of weekly urinary incontinence was lower in the intensive lifestyle intervention group compared with the metformin and placebo groups (44.2% vs 51.8%, 48.0% urinary incontinence/week, P = 0.04); during the 6-year follow-up period, these lower rates in intensive lifestyle intervention were maintained (46.7%, 53.1%, 49.9% urinary incontinence/week; P = 0.03). Statistically adjusting for urinary incontinence prevalence at the end of the Diabetes Prevention Program, the treatment arm no longer had a significant impact on urinary incontinence during the Diabetes Prevention Program Outcomes Study. Independent predictors of lower urinary incontinence during the Diabetes Prevention Program Outcomes Study included lower body mass index (odds ratio 0.988, 95% confidence interval 0.982-0.994) and greater physical activity (odds ratio 0.999, 95% confidence interval 0.998-1.000) at the Diabetes Prevention Program Outcomes Study entry, and greater reductions in body mass index (odds ratio 0.75, 95% confidence interval 0.60-0.94) and waist circumference (odds ratio 0.998, 95% confidence interval 0.996-1.0) during the Diabetes Prevention Program Outcomes Study. Diabetes was not significantly related to urinary incontinence.

Conclusions: Intensive lifestyle intervention has a modest positive and enduring impact on urinary incontinence, and should be considered for the long-term prevention and treatment of urinary incontinence in overweight/obese women with glucose intolerance.

Keywords: Diabetes Prevention Program Outcomes Study; lifestyle intervention; urinary incontinence; weight loss.

Figure 1. Prevalence, incidence, and resolution of UI in DPPOS by Treatment Group

Prevalence is defined as proportion of participants reporting “yes” to the question “in the past 7 days did you leak even a small amount of urine?” at each annual visit. Incidence is defined as proportion of participants reporting “yes” to the above question at the current visit and “no” in the previous visit. Resolution is defined as proportion of participants reporting “no” to the above question at the current visit and “yes” in the previous visit.

Figure 2. Predictors and Correlates of UI during DPPOS

Odds ratios are calculated as per unit increase in UI based on unit change in the predictors for the continuous predictors, or verse the reference group for the categorical predictors. Note that for BMI change, the unit was % reduction in BMI; and, for WC change, the unit was cm reduction in WC. Also note that X axis is logarithmic scale. References for the categorical variables are: Placebo for the treatment group; 01A visit for the visit year; Non-Hispanic White for the Race/Ethnicity groups; no children for Parity; never smoked for the smoking status at 01A visit. Abbreviations: OR, odds ratio; LCL, lower 95% confidence limit; UCL, upper 95% confidence limit; p, p value; FPG= Fasting plasma glucose; Waist circum= waist circumference; CRP= C-reactive protein; DPPOS= Diabetes Prevention Program Outcomes Study; DPP= Diabetes Prevention Program