We previously reported that overweight and obese women in a weight loss intervention had significantly greater reductions in stress and total incontinence episodes than those in a control group at 6 months. We now demonstrate that decreased stress incontinence episode frequency was maintained at 12 months with a trend toward continued reduction in total incontinence episode frequency. However, at 18 months there were no statistically significant differences in the reduction of weekly incontinence episode frequency between the groups despite a substantial difference in weight loss and greater participant satisfaction with improvements in UI.
The lifestyle intervention appeared to be effective in reducing stress incontinence episodes at 6 and 12 months but had little effect on urge incontinence episodes. It has been suggested that obesity, specifically central adiposity, may contribute to incontinence by increasing intra-abdominal pressure, which in turn leads to increased bladder pressure and urethral mobility.8,9,11
Reducing this cascade via weight reduction might lead to greater reductions in stress incontinence than in urge episodes. Further study of the mechanisms related to the effects of weight reduction on incontinence may help explain this difference.
The effect of the lifestyle intervention on reductions in stress and total incontinence episodes, relative to the control group, was greater at 6 and 12 months than at 18 months. This finding may relate to the fact that there was a 5.8% difference in weight loss between the groups at months 6 and 12, but only a 3.9% difference at 18 months. Increasing the magnitude of the weight loss or preventing weight regain may increase the beneficial effects on incontinence. However, it is important to note that the lack of significant differences in incontinence between the groups at 18 months did not result from relapses in incontinence in the intervention group, but rather from increasingly large reductions in the control group. The improvements in UI reported by the control group may reflect regression to the mean, increased awareness of bladder habits or measurement error due to repeated self-report assessments. There was no evidence that use of the incontinence booklet or differences in baseline incontinence episodes between dropouts in the intervention and control groups explained these improvements in the control group.
While the voiding diaries were the primary outcome measure for this trial, the observation that women in the intervention group reported sustained benefits relative to controls in their perceived frequency and magnitude of leakage, and greater overall satisfaction with changes in incontinence through 18 months, is an important finding supporting the use of lifestyle intervention as a first line treatment for overweight and obese women with UI. Greater perceived improvements in incontinence in the intervention group may be due to the weight loss, and overall satisfaction with health and appearance. Alternatively these self-report items may capture information about incontinence that is not reflected in the voiding diary or pad weights.
The lifestyle intervention used in this study, which included diet, physical activity and behavioral strategies, resulted in an average weight loss from baseline of approximately 8% at 6 and 12 months, and 5% at 18-month followup. These weight losses were similar to those seen in Look AHEAD, a clinical trial of lifestyle intervention in overweight and obese participants with type 2 diabetes, and exceeded the results of several other trials, perhaps due to the inclusion of free meal replacement products, the high physical activity goal, the motivation afforded by the condition of incontinence and the ongoing contact provided throughout the 18 months of followup.22–24
Difficulties with the long-term maintenance of weight loss are typical of lifestyle interventions, perhaps reflecting decreasing adherence to lifestyle interventions over time, and support the need for further research to improve weight loss maintenance.
Strengths of this study include the randomized design, excellent retention, multidimensional approach to assessing impact on UI, conservative imputation approach used for missing data and the collection of outcomes by staff masked to treatment assignment. Limitations include the fact that the control group reported marked decreases in incontinence episodes over time that could not readily be explained. Urodynamic measures were conducted only on a subset of women at baseline and 6 months. Participants in this study had a BMI of 36 kg/m2, and may have been particularly motivated and adherent. Therefore, their results may not accurately reflect weight loss and changes in UI that would be realized in the general population.