Skip to main content
. Author manuscript; available in PMC: 2010 May 27.
Published in final edited form as: N Engl J Med. 2009 Jan 29;360(5):481–490. doi: 10.1056/NEJMoa0806375

Table 2.

Body Weight and Frequency of Urinary-Incontinence Episodes at Baseline and at 6 Months According to Treatment Group.*

Outcome Weight-Loss Group
(N = 226)
Control Group
(N = 112)
P Value
Body weight
 Baseline — kg 98±17 95±16
 6 Mo — kg 90±17 94±17
 % Change (95% CI) −8.0 (−9.0 to −7.0) −1.6 (−2.7 to −0.4) <0.001
Urinary-incontinence episodes
 Any incontinence
  Baseline — no./wk 24±18 24±16
  6 Mo — no./wk 13±15 17±19
  % Change (95% CI) −47 (−54 to −40) −28 (−41 to −13) 0.01
 Stress incontinence
  Baseline — no./wk 9±11 10±10
  6 Mo — no./wk 4±7 7±9
  % Change (95% CI) −58 (−67 to −46) −33 (−50 to −9) 0.02
 Urge incontinence
  Baseline — no./wk 14±14 13±15
  6 Mo — no./wk 8±11 10±15
  % Change (95% CI) −42 (−51 to −32) −26 (−44 to −3) 0.14
*

Plus–minus values are means ±SD and were calculated with the use of multiply imputed data sets for body weight and frequency of urinary-incontinence episodes.

Percentage changes and P values for the comparison between the weight-loss group and the control group were calculated with the use of multiply imputed data sets and mixed linear regression models, with control for clinical site and correlation of outcomes in the intervention groups. The data sets for body weight were based on 221 women in the weight-loss group and 97 in the control group for whom data were available at baseline and 6 months.

Percentage changes and P values for the comparison between the weight-loss and the control groups were calculated with the use of multiply imputed data sets and negative binomial models, with control for clinical site and correlation of outcomes in the intervention group. The data sets for urinary incontinence were based on 214 women in the weight-loss group and 90 in the control group for whom data were available at baseline and 6 months.