Key messages
– Some approaches for pelvic floor muscle (muscles supporting the bladder, bowel, and womb) training for women with involuntary (accidental) urine leakage are better than others, and some might be as good as each other.
– We were unsure about many of the approaches, often because there was only one study or several small studies investigating the question.
– More studies are needed that directly compare different approaches to pelvic floor muscle training, especially studies to find out whether different amounts (dose) of training are better than others — for example, exercising more or less often each week.
What is urinary incontinence?
The muscles of the pelvic floor form a sling between the legs from the pubic bone at the front of the pelvis to the tail bone at the back of the pelvis. This sling supports the bladder, bowel, and womb. If these muscles become weak, a woman may not be able to control the flow of urine (wee) from the bladder. This is called urinary incontinence. There are three main types of urinary incontinence. These are leakage with physical exertion (stress urinary incontinence), leakage with an urgent need to empty the bladder (urgency urinary incontinence), and both together (mixed urinary incontinence). This review covers all three types. Urinary incontinence seriously affects women's quality of life.
What is pelvic floor muscle training?
Improving the strength, endurance, and co-ordination of the pelvic floor muscles by training them can decrease urine leakage. Training involves using different approaches (exercises) or different amounts (different doses) of the same approach. Supervision of training can be done differently too (for example, one clinician teaching one woman or one clinician teaching a group of women; or delivering the exercises by the internet, mobile phone, or in a leaflet).
What did we want to find out?
We wanted to find out if:
– one type of exercise was better than another;
– higher exercise dose was better than lower exercise dose;
– more-intensive supervision was better than less-intensive supervision of exercise.
What did we do?
We looked for studies that compared one type of exercise versus another, higher versus lower exercise dose, or more-intensive versus less-intensive supervision. The studies included women with urinary incontinence, but not women with conditions of the nervous system or women who were pregnant or had recently given birth.
We compared and summarised the study results, and rated our confidence in the evidence based on factors such as study methods and sizes. We were most interested in women's quality of life after treatment.
What did we find?
We found 63 studies that involved 4920 women with urinary incontinence. The biggest study included 362 women and the smallest 11 women. While the studies were conducted in countries around the world, most were from middle- to high-income countries (meaning that women had reasonable access to good health care). Most studies lasted three months. Three studies had funding or non-monetary support from a commercial company.
Main results
We were mainly interested in women's quality of life. This was measured using questionnaires about how often the incontinence happened, how much urine leaked, and how much it affected or limited the women.
1. Type of exercise
– Combined training (e.g. using bridging exercises at the same time as pelvic floor muscle contraction) may be slightly better than pelvic floor muscle training alone.
– Pelvic floor muscle training may be better than indirect training (exercises that do not include contractions of the pelvic floor muscles).
- Pelvic floor muscle training combined with indirect training may make little to no difference compared to pelvic floor muscle training alone.
2. Exercise dose
– There was not enough evidence to comment on dose.
3. More-intensive versus less-intensive supervision
– There is probably little or no difference between one-to-one supervision and group classes.
– Delivering training instructions via technology (e.g. internet, mobile phone app) may be slightly better than leaflets.
– There was not enough evidence to comment on other methods of training delivery.
What are the limitations of the evidence?
We are confident that there is little to no difference between one-to-one supervision and group classes. We are less confident that combining indirect training with pelvic floor muscle training compared to pelvic floor muscle training only makes little to no difference, and delivering pelvic floor muscle training instructions via technology is slightly better than the leaflets, or that pelvic floor muscle training is better than indirect training.
Our confidence in the evidence for other comparisons of approaches to these exercises is low to very low, and the results of further research could differ from the results of this review.
We were unsure about the evidence as study methods were not well described, not all the studies provided data about everything that we were interested in or they did not report it in a way we could use, results were inconsistent across different studies, and many studies were very small.
How up to date is this evidence?
This evidence is up to date to 27 September 2023.
Although there is low- to moderate-certainty evidence that some approaches to PFMT are better than others, for some there was little or no difference. The 7th International Consultation on Incontinence recommends PFMT as first-line therapy for women with urinary incontinence. Direct PFMT (sets of repeated, isolated, voluntary pelvic floor muscle contractions) may result in a small improvement in incontinence quality of life compared to indirect training. In terms of improved quality of life, PFMT can be supervised individually or in a group because it probably makes little to no difference in achieving this outcome.
Many comparisons had low- or very low-certainty evidence, often because there was only one trial or several small trials with methodological limitations.
More, better designed and reported trials, directly comparing PFMT approaches are needed, especially trials investigating exercise dose.
Pelvic floor muscle training (PFMT) is a recommended treatment for female stress, urgency, and mixed urinary incontinence. Training varies in exercise type (pelvic floor muscles contracting with and without other muscles), dose, and delivery (e.g. amount and type of supervision).
To assess the effects of alternative approaches (exercise type, dose, and delivery) to pelvic floor muscle training (PFMT) in the management of urinary incontinence (stress, urgency, and mixed) in women.
We searched the Cochrane Incontinence Specialised Register (searched 27 September 2023; which contains CENTRAL, MEDLINE, ClinicalTrials.gov, and World Health Organization ICTRP), handsearched journals and conference proceedings, and reviewed reference lists of relevant articles.
Randomised, quasi-randomised, or cluster-randomised trials in female stress, urge, or mixed urinary incontinence where one trial arm included PFMT and another was an alternative approach to PFMT type, dose, or intervention delivery. We excluded studies with participants with neurological conditions or pregnant or recently postpartum.
Two review authors independently assessed trials for eligibility and methodological quality using the Cochrane RoB 1 tool. We extracted and cross-checked data and resolved disagreements by discussion. Data processing was as described in the Cochrane Handbook for Systematic Reviews of Interventions (Version 6). Synthesis was completed in intervention subgroups.
This is a review update. The analysis included 63 trials with 4920 women; the previous version included 21 trials with 1490 women. Samples sizes ranged from 11 to 362. Overall, study participants were mid-age (45 to 65 years) parous women with stress or stress-predominant mixed urinary incontinence (46 trials), who had no prior incontinence treatment or pelvic surgery, or appreciable pelvic floor dysfunction. Trials were conducted in countries around the world, mostly in middle- or high-income settings (53 trials). All trials had one or more arms using 'direct' PFMT, defined as repeated, isolated, voluntary pelvic floor muscle contractions.
Trials were categorised as comparisons of exercise type (27 trials, 3 subgroups), dose (11 trials, 5 subgroups, 1 with no data), and delivery (25 trials, 5 subgroups). Incontinence quality of life data are reported here as the primary outcome. Adverse event data were summarised narratively.
Comparison 1: exercise type
Co-ordinated training (body movements with concurrent pelvic floor muscle contraction) versus direct PFMT
Co-ordinated training may slightly improve quality of life (standardised mean difference (SMD) −0.22, 95% confidence interval (CI) −0.44 to −0.01; I2 = 81%; 8 trials, 356 women; low-certainty evidence).
Indirect training (exercises that are not contractions of the pelvic floor muscles) versus direct PFMT
Direct PFMT may moderately improve quality of life (SMD 0.70, 95% CI 0.38 to 1.02; I2 = 78%; 4 trials, 170 women; low-certainty evidence).
Indirect training combined with direct PFMT versus direct PFMT
Combining indirect training with direct PFMT may make little to no difference in quality of life (SMD −0.08, 95% CI −0.26 to 0.10; I2 = 33; 7 trials, 482 women; low-certainty evidence).
Comparison 2: exercise dose
PFMT with resistance device versus PFMT without resistance device
PFMT without a resistance device may slightly improve incontinence quality of life, but the evidence is very uncertain (SMD 0.22, 95% CI −0.04 to 0.48; I2 = 32%; 3 trials, 227 women; very low-certainty evidence).
Maximal pelvic floor muscle contractions versus submaximal pelvic floor muscle contractions
No data reported.
PFMT more days per week versus PFMT fewer days per week
PFMT more days per week may greatly improve incontinence quality of life (SMD −1.60, 95% CI −2.15 to −1.05; 1 trial, 68 women; low-certainty evidence).
PFMT in upright body positions versus PFMT when lying down
No data reported.
Comparison 3: exercise intervention delivery
PFMT supervised in clinic versus PFMT at home
Clinic supervision may slightly improve incontinence quality of life, but the evidence is very uncertain (SMD −0.30, 95% CI −0.65 to 0.05; I2 = 89%; 3 trials, 137 women; very low-certainty evidence).
More clinician contact for PFMT supervision versus less clinician contact
No usable data reported.
Individual supervision of PFMT versus group supervision
Individually supervised PFMT probably results in little to no difference in quality of life (SMD −0.18, 95% CI −0.35 to −0.01; I2 = 0%; 5 trials, 544 women; moderate-certainty evidence).
PFMT supervised in clinic versus supervision using e-health (mobile app communication with clinicians)
Clinic supervision may make little to no difference in incontinence quality of life, but the evidence is very uncertain (SMD −0.11, 95% CI −0.41 to 0.19; 1 trial, 173 women; very low-certainty evidence).
PFMT instruction delivered via e-health versus written instruction
E-health delivery may slightly improve quality of life (SMD −0.21, 95% CI −0.43 to 0.01; I2 = 25%; 3 studies, 318 women; low-certainty evidence).
Adverse events
Nine trials collected adverse event data; 66/1083 (6%) women had an adverse event. Almost all events were associated with use of an intravaginal or intrarectal training device. The adverse events were vaginal discharge, spotting, or discomfort.
Limitations in the evidence
Four main factors influenced our certainty in the evidence: 44 trials were at unclear or high risk of selection bias; data were sparse in some subgroups with few trials, trials that did not measure outcomes of interest, or did not report usable data; results were inconsistent; and many trials were small (imprecise).