Background: About a third of women have urinary incontinence and up to a 10th have faecal incontinence after childbirth. Pelvic floor muscle training is commonly recommended during pregnancy and after birth both for prevention and the treatment of incontinence.

Objectives: To determine the effect of pelvic floor muscle training compared to usual antenatal and postnatal care on incontinence.

Search methods: We searched the Cochrane Incontinence Group Specialised Register, which includes searches of CENTRAL, MEDLINE, MEDLINE in Process and handsearching (searched 7 February 2012) and the references of relevant articles.

Selection criteria: Randomised or quasi-randomised trials in pregnant or postnatal women. One arm of the trial needed to include pelvic floor muscle training (PFMT). Another arm was either no PFMT or usual antenatal or postnatal care.

Data collection and analysis: Trials were independently assessed for eligibility and methodological quality. Data were extracted then cross checked. Disagreements were resolved by discussion. Data were processed as described in the Cochrane Handbook for Systematic Reviews of Interventions. Three different populations of women were considered separately, women dry at randomisation (prevention); women wet at randomisation (treatment); and a mixed population of women who might be one or the other (prevention or treatment). Trials were further divided into those which started during pregnancy (antenatal); and those started after delivery (postnatal).

Main results: Twenty-two trials involving 8485 women (4231 PFMT, 4254 controls) met the inclusion criteria and contributed to the analysis.Pregnant women without prior urinary incontinence (prevention) who were randomised to intensive antenatal PFMT were less likely than women randomised to no PFMT or usual antenatal care to report urinary incontinence up to six months after delivery (about 30% less; risk ratio (RR) 0.71, 95% CI 0.54 to 0.95, combined result of 5 trials).Postnatal women with persistent urinary incontinence (treatment) three months after delivery and who received PFMT were less likely than women who did not receive treatment or received usual postnatal care to report urinary incontinence 12 months after delivery (about 40% less; RR 0.60, 95% CI 0.35 to 1.03, combined result of 3 trials). It seemed that the more intensive the programme the greater the treatment effect.The results of seven studies showed a statistically significant result favouring PFMT in a mixed population (women with and without incontinence symptoms) in late pregnancy (RR 0.74, 95% CI 0.58 to 0.94, random-effects model). Based on the trial data to date, the extent to which mixed prevention and treatment approaches to PFMT in the postnatal period are effective is less clear (that is, offering advice on PFMT to all pregnant or postpartum women whether they have incontinence symptoms or not). It is possible that mixed prevention and treatment approaches might be effective when the intervention is intensive enough.There was little evidence about long-term effects for either urinary or faecal incontinence.

Authors' conclusions: There is some evidence that for women having their first baby, PFMT can prevent urinary incontinence up to six months after delivery. There is support for the widespread recommendation that PFMT is an appropriate treatment for women with persistent postpartum urinary incontinence. It is possible that the effects of PFMT might be greater with targeted rather than mixed prevention and treatment approaches and in certain groups of women (for example primiparous women; women who had bladder neck hypermobility in early pregnancy, a large baby, or a forceps delivery). These and other uncertainties, particularly long-term effectiveness, require further testing.