Sixty-six studies were included. The number of participants was unclear.
Non-pharmacological treatments.
There was evidence of effectiveness for:
pelvic floor muscle training compared with placebo (1 systematic review), RR of self-reported cure or improvement 1.53 (95% confidence interval, CI: 1.26, 1.87);
clinic-based pelvic floor exercise programme compared with a self-help booklet (1 RCT);
pelvic floor training plus bladder training versus either alone (1 systematic review plus 1 RCT);
nurse-administered home-based pelvic floor training plus biofeedback plus bladder training programme versus social visits from nurses (1 RCT); and
phenylpropanolamine versus pelvic floor training (1 RCT).
Electrical stimulation: there was evidence of effectiveness for electrical stimulation versus sham (1 RCT, n=68; NNT 5, 95% CI: 3, 42).
Vaginal cones: one systematic review found that vaginal cones increased the proportion who were subjectively cured in comparison with control interventions not involving pelvic floor muscles (RR 0.74, 95% CI: 0.59, 0.93), but found no difference between the treatments in objective outcomes.
Bladder training: there was evidence of effectiveness for bladder training versus no training (2 small RCTs); the OR for failure was 0.07 (95% CI: 0.03, 0.19).
Prompted voiding: there was some evidence of effectiveness for prompted voiding versus control (1 systematic review); the OR for no improvement in wet episodes was 0.59 (95% CI: 0.31, 1.14; WMD for episodes of incontinence in 24 hours -0.93, 95% CI: -1.32, -0.53).
Pharmacological treatments.
Anticholinergic drugs: there was evidence of effectiveness for anticholinergic drugs versus placebo for urge incontinence (1 systematic review); the OR for cure was 1.14 (95% CI: 1.29, 1.54). However, anticholinergic drugs increased dry mouth (tolterodine RR 3.02, 95% CI: 2.45, 3.71).
Adrenergic drugs: there was evidence of marginal effectiveness for adrenergic drugs versus placebo (1 systematic review); the RR for cure or improvement ranged from 1.55 to 1.96 across 3 specific drugs. However, adrenergic drugs non-statistically significantly increased adverse effects.
Surgical interventions.
Open retropubic colposuspension: there was evidence of effectiveness for open retropubic colposuspension versus bladder neck needle suspension, anterior vaginal repair and laparoscopic colposuspension. One systematic review showed lower failure rates than all three alternatives and fewer peri-operative complications than bladder neck needle suspension and anterior vaginal repair. However, open retropubic colposuspension increased new or recurrent prolapse compared with anterior vaginal repair.
Periurethral injections: there were no significant differences in cure or improvement between periurethral injection of autologous fat versus placebo (1 RCT, n=68; RR 0.98, 95% CI: 0.75, 1.29), but fat injection increased complications (NNH 5, 95% CI: 3, 11).
Bladder neck needle suspension: there was a lack of evidence.
Anterior repair: there was a lack of studies other than comparisons with open retropubic colposuspension and bladder needle suspension.
Laparoscopic colposuspension: no study was identified.
Suburethral sling procedures: there was a lack of studies other than comparisons with open retropubic colposuspension.
Other results were also reported.