Antenatal Pelvic Floor Training: What It Prevents and When It Starts

Pelvic floor muscle training started in pregnancy by women who are not yet leaking is the only high certainty prevention finding in this field: they were about 29% less likely to report urinary incontinence at 3 to 6 months postpartum than women receiving usual care, 179 per 1,000 against 251 per 1,000, from 5 trials and 673 women. In late pregnancy the same comparison showed about 62% less likely, 160 per 1,000 against 421 per 1,000, graded moderate certainty¹.

I did nothing of the kind in either pregnancy, and my stress urinary incontinence began about six months after my second child. I cannot tell you that antenatal training would have prevented it, because a population effect is not a personal counterfactual, and I am wary of writing a regret narrative that dresses up an anecdote as evidence. What I can say is that in two pregnancies with a full complement of antenatal appointments, nobody at any point told me the figures in the paragraph above. Read red flags and when to stop and get checked first.

The finding worth leading with

Start early, and start before there is a symptom.

From the Cochrane review of pelvic floor muscle training for preventing and treating urinary and faecal incontinence in antenatal and postnatal women, comparing training started in continent women against usual care¹:

• Urinary incontinence in late pregnancy: risk ratio 0.38, 95% confidence interval 0.20 to 0.72. 421 per 1,000 in controls against 160 per 1,000 with training, about 62% less likely, from 6 trials and 624 women, moderate certainty.
• Urinary incontinence at 3 to 6 months postpartum: risk ratio 0.71, 95% confidence interval 0.54 to 0.95. 251 per 1,000 in controls against 179 per 1,000 with training, about 29% less likely, from 5 trials and 673 women, high certainty.
• Urinary incontinence at 6 to 12 months postpartum: risk ratio 1.20, 95% confidence interval 0.65 to 2.21, no evidence of benefit, from 1 trial and 44 women, low certainty.

The 3 to 6 month result is the one to hold onto, because high certainty gradings are genuinely rare in pelvic health and this is the only prevention finding in the set that carries one. The 6 to 12 month row is worth reading too, and this site prints it rather than omitting it: one trial, 44 women, no evidence of benefit. That does not overturn the earlier finding, but it does mean nobody should claim a demonstrated effect lasting a year.

The current version of that review is the 2020 update. A “Woodley 2025” citation circulates in some places; it is unverified and this site does not use it.

The control rate is the other half of the story

Look at 421 per 1,000 in late pregnancy again.

That is the group receiving usual care, and it means roughly two in five of them reported urinary incontinence in late pregnancy. Leaking during pregnancy is not a marginal event, and framing it as something that happens to unlucky women misdescribes the base rate substantially.

The postnatal base rate is similar in shape. A systematic review of postpartum urinary incontinence found a pooled prevalence in the first 3 months after birth of 33% for any incontinence, with a 95% confidence interval of 32 to 36, 12% weekly and 3% daily, and 31% after vaginal delivery against 15% after caesarean section². The World Health Organization repeats the same framing: approximately one third of women experience urinary incontinence in the first three months after childbirth, gradually decreasing during the first postpartum year³.

Common, and treatable, and the second word is the one that keeps getting dropped. Why symptoms get dismissed and how to be heard covers that mechanism.

What NICE actually triggers on, precisely

This is the precision point most summaries get wrong, and it is worth getting right because it determines whether you can ask for something specific.

NICE NG210 recommendation 1.3.12 says to consider a 3 month programme of supervised pelvic floor muscle training from week 20 of pregnancy, for pregnant women who have a first-degree relative with pelvic floor dysfunction⁴. That is one criterion, not a general category of women at risk. The same recommendation has a postnatal limb with a broader trigger list, covering assisted vaginal birth with forceps or vacuum, an occipito-posterior vaginal birth, or injury to the anal sphincter, and that broader list applies to the postnatal limb only.

Alongside that, recommendations 1.3.9 to 1.3.11 encourage women of all ages, including those pregnant or recently delivered, to do pelvic floor muscle training and to continue it throughout life⁴. So there is a general encouragement and one specific supervised-programme trigger, and conflating them is how a reader ends up believing they have been denied something they were never eligible for.

If you are already leaking during pregnancy, a different recommendation applies. NG210 recommendation 1.6.14 says to offer a programme of supervised pelvic floor muscle training for at least 3 months to women, including pregnant women, with stress or mixed urinary incontinence⁴, and NG123 recommendation 1.4.4 says the same for women aged 18 and over⁵. Note the verb change from consider to offer.

Where the evidence does not go

Three places, and this site names all three rather than blurring them.

Already leaking, started antenatally, as treatment. The Cochrane review found no clear effect at any time point in this population, with late pregnancy at a risk ratio of 0.70, a 95% confidence interval of 0.44 to 1.13, and very low certainty¹. That does not remove the guideline offer above, which stands on a wider evidence base for stress and mixed incontinence generally. It does mean the prevention numbers are not yours if you are already leaking.

Everyone, regardless of continence status. Recruiting antenatal women without selecting for continence gave a smaller effect: late pregnancy risk ratio 0.78, 95% confidence interval 0.64 to 0.94, from 11 trials and 3,307 women, moderate certainty; and 3 to 6 months postpartum risk ratio 0.73, 95% confidence interval 0.55 to 0.97, from 5 trials and 1,921 women, low certainty¹. The review’s authors conclude that early structured training in early pregnancy for continent women may prevent onset, and that population approaches may have a smaller effect.

Bowel leakage. Faecal incontinence outcomes were consistently null and low certainty across every comparison in the review¹, so this site makes no claim of a benefit there. Faecal incontinence and anal sphincter training and perineal tears and OASI recovery cover that territory.

What the World Health Organization says, and the mistake most summaries make

Read the direction of the recommendation carefully, because it is routinely reversed in summaries.

WHO recommendation 7, verbatim: for postpartum women, starting routine pelvic floor muscle training after childbirth for the prevention of postpartum urinary and faecal incontinence is not recommended. Its critical remark, also verbatim: while training started after childbirth is not recommended as a preventive measure, women with involuntary loss of small volumes of urine after childbirth should be advised of the potential benefits of training for treatment of urinary incontinence³.

WHO also records that training started early in pregnancy was outside the scope of its assessment, and notes separately that the Cochrane evidence suggests beginning early in pregnancy probably prevents urinary incontinence in late pregnancy and reduces the risk postpartum, particularly at three to six months.

So “WHO recommends postnatal pelvic floor training” is false, and “WHO says pelvic floor training does not work” is equally false. The accurate sentences are that WHO does not recommend routine postnatal training as prevention for women who are not leaking, and does advise that women who are leaking be told about its benefits as treatment. Postnatal pelvic floor recovery covers what that means in the first year.

Why supervision keeps appearing in these recommendations

Because the trials tested taught programmes, and because the movement is genuinely easy to get wrong.

NICE specifies that programmes should be supervised by a physiotherapist or other healthcare professional with the appropriate expertise in pelvic floor muscle training, with at least one review to assess progress during the programme and one at the end⁴. That structure is not decoration; it is what distinguishes a programme from an instruction.

The technique evidence supports it. In 47 women referred for urodynamic evaluation of urinary incontinence, only 49% achieved an ideal contraction after brief standardised verbal instruction and 25% performed a technique that could potentially promote incontinence⁶. That was a referred, symptomatic population in 1991 and not a claim about everyone: in 779 women attending community-based primary care practices, correct contraction on the first attempt ranged from 68.6% to 85.8% and 78% of those who got it wrong learned after brief instruction⁷. Most pregnant women reading this are closer to the second population than the first, which is genuinely reassuring, and the minority who are not are the ones with most to gain from having it checked. How to do a pelvic floor contraction correctly describes the movement without prescribing a dose, and pelvic floor exercise programmes explained covers how a real dose is set.

One counterweight applies in pregnancy as everywhere: if contracting makes symptoms worse, that is a reason to be reassessed rather than to persist. See hypertonic pelvic floor and downtraining.

Getting it, which is the harder part

The recommendation exists. Access to a supervised antenatal programme is another matter.

NHS England runs a perinatal pelvic health services programme, and the one defensible coverage fact this site holds is that 20 of 42 integrated care boards had a service in operation as at 16 May 2024, two months after the March 2024 universal coverage deadline⁸. No access or waiting-time data is published for those services at all, and this site does not substitute a general community physiotherapy figure, because that describes a different queue.

Where self-referral for pelvic health exists at all it is often restricted to pregnancy and the first year after birth, which makes pregnancy one of the few points where a direct route may be open to you. Getting referred to pelvic health physiotherapy sets out the routes across five health systems, and pelvic health physiotherapy waiting times is written around what is published and what is not.

The policy argument for doing better than this is on the record. The Independent Medicines and Medical Devices Safety Review recommended that pelvic floor education be encouraged in schools and certainly in antenatal classes, and that the NHS adopt the French model for universal postnatal pelvic floor rehabilitation, while recording in the same report that specialist pelvic floor physiotherapy cannot match the demand⁹. Pelvic floor physiotherapy sets out the treatment in full, and pelvic girdle pain in pregnancy covers the other main reason to be seen while pregnant.