Although urinary incontinence is a problem for up to one half of all adult women, the etiology remains controversial. Traditional theory and a few studies support a central role for vaginal delivery in the etiology of urinary incontinence. Conversely, other studies have failed to demonstrate an association. Advanced age, obesity, and smoking also have been identified as risk factors. Buchsbaum and colleagues investigated the role of vaginal delivery and familial factors in urinary incontinence.

The authors studied pairs of sisters recruited by advertising that did not identify incontinence as the focus of the study. Eligibility for the study depended on one sister being nulliparous and the other having had at least one vaginal delivery. All participants completed extensive questionnaires covering demographic, medical, surgical, and obstetric history, as well as use of tobacco and alcohol. Participants who reported at least one episode of incontinence within the previous four weeks were asked for additional information about the severity, duration, and type of incontinence.

Participants were offered a complete physical examination, including assessment of mobility, gait, and a cough stress test. Pelvic and rectovaginal examination was used to assess pelvic floor relaxation, status of pelvic organs, and the tone of the anal sphincter. Urethral mobility was assessed with a cotton-tipped swab test, and ultrasonography was used to estimate postvoid residual urine volume. Participants with signs or symptoms of urinary incontinence were offered multichannel urodynamic evaluation. The rate of incontinence and related variables was compared between the parous and nulliparous sisters.

Overall, 143 pairs of sisters completed the questionnaires, and 101 also completed the physical evaluation. The average age of participants was 61 years, and the average body mass index was about 28 kg per m². About 30 percent of the women reported hysterectomy, and a similar percentage reported hypertension. The parous and nulliparous participants did not differ significantly in any important variable. The parous sisters reported an average of three vaginal births. The reported rate of incontinence for nulliparous participants was 47.6 percent versus 49.7 percent for parous sisters. The difference between these rates was not significant. Rates of pure stress incontinence were identical (18.4 percent) in parous and nulliparous women. Rates of urge incontinence also were almost identical (8.1 for parous women versus 8.8 percent for nulliparous women). Measures of the severity of incontinence also failed to show any significant differences between parous and nulliparous women. Continence was concordant between 63 percent of the sister pairs (see accompanying table)

The authors conclude that incontinence is a common problem for postmenopausal women, but that no difference in prevalence or severity could be demonstrated between parous and nulliparous sisters. Conversely, familial factors appear to be highly significant predictors of urinary incontinence. The authors suggest that the current focus on delivery in the etiology of incontinence is inappropriate and that research and preventive efforts be directed toward understanding familial factors.

editor's note: Will this study inject some sanity into the current situation in which cesarean delivery is being advocated (and demanded by some patients) to prevent urinary incontinence later in life? By one report,¹ nearly two thirds of obstetricians support elective cesarean delivery for this reason. Combined with the current backlash against vaginal birth after cesarean delivery, the cesarean delivery rate seems set to spiral beyond the current 30 percent unless the debate can be refocused on scientific evidence of the net benefit for mothers. —a.d.w.