Comparisons of approaches to pelvic floor muscle training for urinary incontinence in women

Involuntary urine leakage (or incontinence) is a widespread condition experienced by about a quarter of women. Exercise for the pelvic floor muscles is often the first treatment women are offered. Improving the strength, endurance and co-ordination of the pelvic floor muscles can help decrease the urine leakage. This review included 21 studies in 1490 women and looked at whether one way of teaching, supervising or performing these exercises was better than another. Women who had regular and repeated contact with the person who taught them to do the exercises and monitored their progress were more likely to report they were improved after treatment. Further research is needed because there were problems interpreting the studies, which meant we could not draw any firm conclusions about many of the other possible ways of teaching, supervising or performing these exercises.

Authors' conclusions: 

This review found that the existing evidence was insufficient to make any strong recommendations about the best approach to pelvic floor muscle training. We suggest that women are offered reasonably frequent appointments during the training period, because the few data consistently showed that women receiving regular (e.g. weekly) supervision were more likely to report improvement than women doing pelvic floor muscle training with little or no supervision.

Read the full abstract...

Pelvic floor muscle training is the most commonly recommended physical therapy treatment for women with stress urinary incontinence. It is also sometimes recommended for mixed and, less commonly, urge urinary incontinence. The supervision and content of pelvic floor muscle training programmes are highly variable, and some programmes use additional strategies in an effort to increase adherence or training effects.


To compare the effects of different approaches to pelvic floor muscle training for women with urinary incontinence.

Search strategy: 

We searched the Cochrane Incontinence Group Specialised Trials Register, which contains trials identified from the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE and CINAHL, and handsearching of journals and conference proceedings (searched 17 May 2011), and the reference lists of relevant articles.

Selection criteria: 

Randomised trials or quasi-randomised trials in women with stress, urge or mixed urinary incontinence (based on symptoms, signs or urodynamics). One arm of the study included pelvic floor muscle training. Another arm was an alternative approach to pelvic floor muscle training, such as a different way of teaching, supervising or performing pelvic floor muscle training.

Data collection and analysis: 

We independently assessed trials for eligibility and methodological quality. We extracted then cross-checked data. We resolved disagreements by discussion. We processed data as described in the Cochrane Handbook for Systematic Reviews of Interventions (version 5.2.2). We subgrouped trials by intervention.

Main results: 

We screened 574 records for eligibility and included 21 trials in the review. The 21 trials randomised 1490 women and addressed 11 comparisons. These were: differences in training supervision (amount, individual versus group), in approach (one versus another, the effect of an additional component) and the exercise training (type of contraction, frequency of training). In women with stress urinary incontinence, 10% of those who received weekly or twice-weekly group supervision in addition to individual appointments with the therapist did not report improvement post-treatment compared to 43% of the group who had individual appointments only (risk ratio (RR) for no improvement 0.29, 95% confidence interval (CI) 0.15 to 0.55, four trials). Looking at this another way, 90% of those who had combined group and individual supervision reported improvement versus 57% of women receiving individual supervision only. While women receiving the combination of frequent group supervision and individual supervision of pelvic floor muscle training were more likely to report improvement, the confidence interval was wide, and more than half of the 'control' group (the women who did not get the additional weekly or twice-weekly group supervision) reported improvement. This finding, of subjective improvement in both active treatment groups, with more improvement reported by those receiving more health professional contact, was consistent throughout the review.

We feel there are several reasons why caution is needed when interpreting the results of the review: there were few data in any comparison; a number of trials were confounded by comparing two arms with multiple differences in the approaches to pelvic floor muscle training; there was a likelihood of a relationship between attention and reporting of more improvement in women who were not blind to treatment allocation; some trials chose interventions that were unlikely to have a muscle training effect; and some trials did not adequately describe their intervention.