Secondary Logo

Journal Logo


Pelvic Floor Muscle Training During Pregnancy to Prevent Urinary Incontinence

A Single-Blind Randomized Controlled Trial

Mørkved, Siv MSc, PT; Bø, Kari PhD, PT; Schei, Berit MD, PhD; Salvesen, Kjell Åsmund MD, PhD

Author Information
  • Free

Urinary incontinence is a chronic health complaint that severely reduces quality of life and has many sufferers reporting effects on their social, domestic, physical, occupational, and leisure activities.1–3 Aside from the personal and social costs to sufferers, the direct and indirect health care costs are substantial. The approximate annual cost of the condition in the United States has been estimated at $11.2 billion in the community and $5.2 billion in nursing homes.4 Urinary incontinence is defined by the International Continence Society as “the complaint of any involuntary leakage of urine.” Urinary incontinence is more common in women than in men and affects women of all ages. Prevalence rates in women between 15 and 64 years of age vary from 10% to 30%.4 However, prevalence estimates of urinary incontinence during pregnancy and after childbirth are even higher, varying between 20% and 67%, and 0.3% and 44%, respectively.5–7

Pregnancy and vaginal delivery are considered to be one of the main risk factors in the development of urinary incontinence because pregnancy and childbirth may cause damage to the fascias, ligaments, pelvic floor muscles, and nerves supporting and controlling the bladder neck and urethra.8–12

To prevent urinary incontinence, women have been encouraged to conduct pelvic floor muscle exercises during pregnancy and after childbirth.13 Pelvic floor muscle training after childbirth has been demonstrated to be effective in prevention and treatment of urinary incontinence,14–18 but the effect of pelvic floor muscle exercises during pregnancy on the prevention of urinary incontinence has been sparsely documented. We are aware of two published randomized controlled trials (RCTs) addressing pelvic floor muscle exercise during pregnancy.19,20 However, in Sampselle et al's study19 the dropout rate was high, and it is questionable whether this is a prevention study, whereas only severity of incontinence—and not the prevalence of incontinence at baseline and after intervention—was reported. In Reilly et al's study20 only women with existing bladder neck mobility were included.

At present there is insufficient evidence to determine whether pelvic floor muscle training during pregnancy is effective or ineffective at preventing urinary incontinence in childbearing women.

The aim of our study was to assess whether a 12-week intensive pelvic floor muscle training program during pregnancy, carried out in a primary care setting, could prevent urinary incontinence during pregnancy and after childbirth.


Nulliparous women attending the routine ultrasound control at The National Center for Fetal Medicine, Trondheim University Hospital, at 18 weeks' pregnancy were invited to participate in the study. They received a written invitation and information about the study as a supplement to the letter inviting them to the routine ultrasound control. The women were asked to return a signed consent form if they wanted to participate in the study. Women were eligible for the trial if they were nulliparous and 18 years or older, with a single live fetus at the routine ultrasound scan. Exclusion criteria were pregnancy complications, high risk for preterm labor, pain during pelvic floor muscle contractions, ongoing urinary tract infection, or diseases that could interfere with participation. In addition, women who lived too far from Trondheim to be able to attend weekly training groups were excluded. The procedures followed were in accordance with the ethical standards of the responsible regional committee on human experimentation and with the Helsinki Declaration. The Regional Medical Ethics Committee approved the study. The participants were not compensated financially.

Women were recruited to the trial from October 1998 to May 2000 and followed up until April 2001. They came from a nonselected population from a geographically well-defined area consisting of four municipalities surrounding and including the city of Trondheim. In this period, 1533 nulliparous women from this area attended a routine ultrasound scan at the National Center for Fetal Medicine in Trondheim.

Three hundred forty-two women gave their signed consent to participate in the trial. Forty-one women were excluded or withdrew before the first examination: 17 lived too far away to be able to attend the training groups, eight had pregnancy complications, four were not nulliparous, two had twin pregnancies, five were not able to meet for the first assessment, and five withdrew for unknown reasons (Figure 1).

Figure 1
Figure 1:
Trial profiles showing the flow of participants through each stage of the randomized trial comparing training and control group.Mørkved. Prevention of Incontinence. Obstet Gynecol 2003.

Three hundred one women were randomly allocated to a pelvic floor muscle training group or to a control group. Randomization was done in blocks of a maximum of 32 with the use of opaque, sealed envelopes. The professional staff involved in the training groups or the outcome assessments had no access to the randomization procedure. A secretary with no other involvement in the trial prepared the envelopes. All the envelopes were mixed thoroughly before they were stored in a larger envelope. Each participant drew and opened one envelope herself and was enrolled by the secretary in the secretary's office. The women were asked not to reveal any information about group allocation to the principal investigator (SM) doing the assessments. The principal investigator was not involved in the training of the women and was blinded to group allocation while making the assessments and plotting the data.

All women in both groups were individually instructed in pelvic floor anatomy and how to contract the pelvic floor muscles correctly by a physiotherapist before randomization. Correct contraction was assessed by vaginal palpation and observation of inward movement of the perineum during contraction,21 and feedback (knowledge of results and performance) was given.

The training group followed a specially designed exercise course including pelvic floor muscle and general exercises. They trained with a physiotherapist for 60 minutes once a week for a period of 12 weeks (between 20 and 36 pregnancy weeks). The physiotherapist encouraged the women to perform near maximal pelvic floor muscle contractions, and to hold the contraction 6–8 seconds. At the end of each contraction the women were asked to add three to four fast contractions. The resting period was about 6 seconds. Group training was performed in lying, sitting, kneeling, and standing positions with legs apart to emphasize specific strength training of the pelvic floor muscles and relaxation of other muscles. Body awareness, breathing, and relaxation exercises and strength training for the abdominal, back, and thigh muscles were performed to music between positions. In addition, the women were encouraged to use their preferred position and perform eight to 12 equally intensive pelvic floor muscle contractions twice per day at home. Motivation was strongly emphasized by the physiotherapists. The pelvic floor muscle training protocol has previously been published by Bø et al22 and is in accordance with the recommendations for general training to increase strength of skeletal muscles.23 Adherence to the training protocol was based on registrations in the women's personal training diaries (two sets of eight to 12 contractions per day) and the reports from the physiotherapists who led the group training (participation in six or more group training sessions). Five physiotherapists from three different outpatient physiotherapy clinics in a primary care setting were involved in leading a total of 14 training groups, each including ten to 15 women.

Women in the control group received the customary information given by their midwife or general practitioner. They were not discouraged from doing pelvic floor muscle exercise on their own.

Women were examined at 20 and 36 weeks' gestation and 3 months after childbirth. The primary outcome measure was self-reports of urinary incontinence. Women reporting urinary incontinence once per week or more during the last month were categorized as incontinent.

Secondary outcomes were recordings, in a home voiding diary, of episodes of involuntary leakage during the 3 days directly after the first assessment (20 weeks' pregnancy) and immediately before the second (36 weeks' pregnancy) and third (3 months postpartum) assessments.24 In addition, the women reported whether the urinary leakage had changed (better, unchanged, worse) from baseline registration at 20 weeks' to 36 weeks' gestation. Vaginal palpation and observation during contraction were used to assess the women's ability to perform pelvic floor muscle contraction.21 Pelvic floor muscle strength (vaginal squeeze pressure [cm H2O]) was measured by a vaginal balloon catheter (balloon size 6.7 × 1.7 cm) connected to a pressure transducer (Camtech Ltd., Sandvika, Norway). The method was found to be reliable and valid in a previous study.21

We aimed to recruit 290 women, giving 85% power (α = 5%) to detect a 50% difference in self-reported urinary incontinence between the two groups, assuming that 42% of the women would be incontinent without intervention. These figures were conservatively based on findings in a previous study.25

The principal analysis was done on an intention to treat basis. The missing last values were carried forward by their baseline values. Groups were compared with exact computation of the Pearson χ2 test if data were categoric. Relative risks and their 95% confidence intervals were calculated for comparisons of proportions (observed ratio of proportions [StatXact 5; Cytel Software Corp., Cambridge, MA]). Normality was evaluated by using the Shapiro-Wilk W test for normality, and the Mann-Whitney U test was used to compare distributions between groups when variables were not normally distributed (SPSS 10.7; SPSS Inc., Chicago, IL). The influence of covariates on the primary outcome variable was explored using logistic regression for odds ratio (SPSS 10). Additional subgroup analyses were carried out on groups according to onset of incontinence. Results are given as mean values with 95% confidence intervals. P values less than .05 were considered significant.


In all, 301 nulliparous women were randomized to a training group (n = 148) or a control group (n = 153) (Figure 1). The trial groups were comparable at baseline (Table 1). Seven women in the control group and five women in the training group withdrew after the first assessment. The reasons for withdrawal were diseases connected to pregnancy (n = 6) or personal (n = 6) (eg, changes in work situation, familiar causes, relocation).

Table 1
Table 1:
Background and Outcome Variables Before Treatment in the Training and the Control Groups

In all, 120 (81%) of 148 women in the training group followed the training protocol. The remaining 28 women were introduced to the training program but participated in less than half of the group training sessions and did not return their personal training diaries after the 12-week training program.

At follow-up, significantly fewer women in the training group than in the control group reported urinary incontinence: 48 of 148 (32%) versus 74 of 153 (48%) at 36 weeks' pregnancy and 29 of 148 (20%) versus 49 of 153 (32%) at 3 months postpartum (Table 2).

Table 2
Table 2:
Women With Self-Reported Urinary Incontinence at 36 Weeks' Pregnancy and 3 Months After Delivery

We found a decreased risk of urinary incontinence among women in the training group, who were 33% less likely to report urinary incontinence at 36 weeks' pregnancy and 39% less likely to report urinary incontinence at 3 months postpartum relative to the control group. The relative risk estimates are shown in Table 2. Potential effects of differences between groups according to baseline characteristics (age, body mass index, pelvic floor muscle strength, physical activity, urinary incontinence at baseline) were analyzed by logistic regression for odds ratio and did not change the results of the relative risk estimates.

Analyses of numbers needed to treat showed that intensive pelvic floor muscle training during pregnancy prevents urinary incontinence in about one in six women during pregnancy, and one in eight women after delivery.

The number of episodes of involuntary leakage during the 3 days directly after the first assessment did not significantly differ between the two groups. At follow-up at 36 weeks' pregnancy the number of leakage episodes was significantly lower in the training group (25 of 148 versus 44 of 144, P = .014). At 3 months postpartum we found similar results (20 of 148 versus 34 of 144, P = .049). In addition, significantly more women in the intervention group reported a reduction in urinary leakage from 20 to 36 weeks' pregnancy (29 of 148 versus nine of 153, P = .002).

The pelvic floor muscle strength was significantly higher in the training group at 36 weeks' pregnancy (P = .008) and 3 months postpartum (P = .048) (Table 3).

Table 3
Table 3:
Pelvic Floor Muscle Strength (Mean and 95% Confidence Interval) at 36 Weeks' Pregnancy and 3 Months After Delivery, Measured by Vaginal Squeeze Pressure (cm H2O)

We stratified women according to continence status before the pregnancy and at 20 weeks' pregnancy. The results from these subgroup analyses were consistent with the overall results (Table 4).

Table 4
Table 4:
Subgroup Analyses Based on Groups Stratified According to Continence Status Before Pregnancy and at Baseline

Intention to treat analyses, analyses including only women attending both pre- and posttests, and per protocol analyses showed minor and insignificant differences in the results.

No negative side effects of the training were reported. There were no statistically significant differences in instrumental deliveries between the two groups. In the training group, 110 women had vaginal deliveries, nine deliveries using forceps, 17 deliveries using vacuum extraction, and 12 cesarean deliveries. In the control group, 107 women had vaginal deliveries, 19 deliveries using forceps, 13 deliveries using vacuum extraction, and 14 cesarean deliveries.


We found that intensive pelvic floor muscle training during pregnancy prevents urinary incontinence during pregnancy and after childbirth. We also found higher pelvic floor muscle strength at 36 weeks' pregnancy (immediately after the end of the supervised training period) and 3 months after delivery in the training group.

This was an RCT of nulliparous women, with blinding of the investigator, a low dropout rate, the use of a standardized training protocol following recommendations from exercise science,23 and high adherence to the training protocol. We have registered data related to the onset of incontinence and performed subgroup analyses, indicating effects on primary, secondary, and tertiary prevention. The results of the subgroup analyses were consistent with the overall results in favoring the training group, which may indicate that the intervention was effective irrespective of onset of incontinence. No negative side effects of the intervention were reported. This is the largest of only three published RCTs addressing the effect of intensive pelvic floor muscle training during pregnancy.

In the present trial, only self-report was used as outcome measurement and in the classification of continent and incontinent participants. Because we studied healthy pregnant women, we found it important to use diagnostic tests and outcome measurements causing minimal discomfort to the participants. To date there is no agreement about what are the most appropriate outcome measures for urinary incontinence. The Urodynamic Society and the standardization committee of the International Continence Society have recommended using measures of urinary leakage and self-reports to evaluate treatment effect.26 We refrained from using pad tests with standardized bladder volume, to avoid inducing urinary tract infections among the pregnant women, and decided to rely on self-reports only. In addition, results from a previous study25 showed that only 45% of the women with urinary incontinence agreed to participate in urodynamic assessment 8 weeks postpartum. This is not surprising, as urodynamics are invasive and 23% of women will complain of moderately severe discomfort.27 Lagro-Janssen et al28 have concluded that urodynamics are unnecessary in most women presenting with urinary incontinence in general practice, but other studies have focused on the need for urodynamic assessment in making a diagnosis and formulating a treatment plan.29,30

In the present study we document a preventive effect of intensive pelvic floor muscle training during pregnancy. Results from previous trials addressing pelvic floor muscle training during pregnancy are inconsistent.19,20 In the study by Sampselle et al19 they did not report the number of women with urinary incontinence before and after intervention, and it is questionable whether this is a prevention study. Their results showed that when controlling for baseline urinary incontinence score, the analyses of covariance showed significantly less urinary incontinence in the training group at 35 weeks' pregnancy, 6 weeks postpartum, and 6 months postpartum. However, at the end of the 12-month period the difference had disappeared. The dropout rate was high, and no significant difference in pelvic floor muscle strength was found between the groups. Reilly et al20 used the same definition of urinary incontinence and reported results that correspond to those of the present study. Fewer women in the training group reported postpartum urinary incontinence: 19.2%, versus 32.7% in the control group. However, they studied only women in a high-risk group (with diagnosed bladder neck mobility) and found no difference in pelvic floor muscle strength between the groups after exercise.

In the present trial the participants came from a non-selected population of nulliparous women. All women were individually instructed in correct pelvic floor muscle contractions. The training group followed a 12-week specially designed pelvic floor muscle exercise course between the 20th and 36th weeks' pregnancy, including group training once per week and daily training at home. The training protocol followed recommendations from exercise science,23 highlighting intensity and frequency of training. Skilled physiotherapists were leading the training groups, gave instructions on the pelvic floor muscle exercises, and encouraged the women to perform intensive contractions. In addition, they emphasized the importance of adherence to the training protocol and motivated the women to follow the protocol. Also, results from previous studies suggest that the training protocol and a close follow-up by skilled physiotherapists are important.14,15,22 In the present study the intervention was carried out by five different physiotherapists in a primary care setting. This may indicate that the training program can easily be implemented as one part of a public health strategy to prevent urinary incontinence in childbearing women. However, there is still insufficient knowledge about whether pelvic floor muscle training during pregnancy can prevent urinary incontinence later in life. Long-term follow-up of the present trial and other comparable populations are needed.

In conclusion, our results show that a specially designed pelvic floor muscle exercise course during pregnancy prevents urinary incontinence during pregnancy and 3 months after delivery.


1. Norton P, MacDonald LD, Sedgwick PM, Stanton SL. Distress and delay associated with urinary incontinence, frequency, and urgency in women. BMJ 1988;297:1187–9.
2. Wyman JF, Harkins SW, Fantl JA. Psychosocial impact of urinary incontinence in the community-dwelling population. J Am Geriatr Soc 1990;38:282–8.
3. Nygaard I, DeLancey JOL, Arnsdorf L, Murphy E. Exercise and incontinence. Obstet Gynecol 1990;75:848–51.
4. Fantl J, Newman D, Colling J, DeLancey JOL, Keeys C, Loughery R, et al. Urinary incontinence in adults: Acute and chronic management. 2nd update. Clinical practice guideline 96-0682. Rockville, Maryland: Department of Health and Human Services, Public Health Service, Agency for Health Care Policy and Research, 1996.
5. Burgio KL, Locher JL, Zyczynski H, Hardin JM, Singh K. Urinary incontinence during pregnancy in a racially mixed sample: Characteristics and predisposing factors. Int Urogynecol J 1996;7:69–70.
6. Viktrup L, Lose G, Rolf M, Barfoed K. The frequency of urinary symptoms during pregnancy and puerperium in the primipara. Int Urogynecol J 1993;4:27–30.
7. Wilson PD, Herbison RM, Herbison GP. Obstetric practice and the prevalence of urinary incontinence three months after delivery. Br J Obstet Gynaecol 1996;103:154–61.
8. Allen RE, Hosker GL, Smith ARB, Warell DW. Pelvic floor damage and childbirth: A neurophysiological study. Br J Obstet Gynaecol 1990;97:770–9.
9. Snooks SJ, Swash M, Mathers SE, Henry MM. Effect of vaginal delivery on the pelvic floor: A five year follow-up. Br J Surg 1990;77:1358–60.
10. Iosif CS, Ingemarson I. Prevalence of stress urinary incontinence among women delivered by elective caesarean section. Int J Gynaecol Obstet 1982;20:87–9.
11. Meyer S, Bachelard O, DeGrandi P. Do bladder neck mobility and urethral sphincter function differ during pregnancy compared with during the non-pregnant state? Int Urogynecol J 1998;9:397–404.
12. Meyer S, Schreyer A, DeGrandi P, Hohlfeld P. The effects of birth on urinary continence mechanisms and other pelvic floor characteristics. Obstet Gynecol 1998;92:613–8.
13. Kegel AH. Progressive resistance exercise in the functional restoration of the perineal muscles. Am J Obstet Gynecol 1948;56:238–49.
14. Mørkved S, Bø K. The effect of postpartum pelvic floor muscle exercise in the prevention and treatment of urinary incontinence. Int Urogynecol J 1997;8:217–22.
15. Mørkved S, Bø K. Effect of postpartum pelvic floor muscle training in prevention and treatment of urinary incontinence: A one-year follow up. Br J Obstet Gynaecol 2000; 107:1022–8.
16. Wilson PD, Herbison GP. A randomised controlled trial of pelvic floor muscle exercises to treat postnatal urinary incontinence. Int Urogynecol J 1998;9:257–64.
17. Chiarelli P. Female urinary incontinence in Australia: Prevalence and prevention in postpartum women [dissertation]. Callaghan, Australia: The University of Newcastle, 2001.
18. Glazener CMA, Herbison GP, Wilson PD, MacArthur C, Lang GD, Gee H, et al. Conservative management of persistent postnatal urinary and faecal incontinence: Randomised controlled trial. BMJ 2001;323:593–6.
19. Sampselle CM, Miller JM, Mims BL, DeLancey JOL, Ashton-Miller JA, Antonakos CL. Effect of pelvic muscle exercise on transient incontinence during pregnancy and after birth. Obstet Gynecol 1998;91:406–12.
20. Reilly ETC, Freeman RM, Waterfield MR, Waterfield AE, Steggles P, Pedlar F. Prevention of postpartum stress incontinence in primigravidae with increased bladder neck mobility: A randomised controlled trial of antenatal pelvic floor exercises. Br J Obstet Gynaecol 2002;109:68–76.
21. Bø K, Hagen RH, Kvarstein B, Larsen S. Pelvic floor muscle exercise for the treatment of female stress urinary incontinence II. Validity of vaginal pressure measurements of pelvic floor muscle strength. The necessity of supplementary methods for control of correct contraction. Neurourol Urodyn 1990;9:479–87.
22. Bø K, Talseth T, Holme I. Single blind, randomised controlled trial of pelvic floor exercise, electrical stimulation, vaginal cones, and no treatment in management of genuine stress incontinence. BMJ 1999;318:487–93.
23. American College of Sports Medicine. Position stand. The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness in healthy adults. Med Sci Sports Exerc 1990;22:265–74.
24. Nygaard I, Holcomb R. Reproducibility of the seven-day voiding diary in women with stress urinary incontinence. Int Urogynecol J 2000;11:157.
25. Mørkved S, Bø K. Prevalence of urinary incontinence during pregnancy and postpartum. Int Urogynecol J 1999; 10:394–8.
26. Blaivas JG, Appell RA, Fantl JA, Leach G, McGuire EJ, Resnick N, et al. Standards of efficacy for evaluation of treatment outcomes in urinary incontinence: Recommendations of the urodynamic society. Neurourol Urodyn 1997;16:145–7.
27. Benness C, Manning J. Patient evaluation of urodynamic investigations. Neurourol Urodyn 1997;16:509–10.
28. Lagro-Janssen ALM, Debruyne FMJ, Van Weel C. Value of patient's case history in diagnosing urinary incontinence in general practice. Br J Urol 1991;67:569–72.
29. Cundiff G, Harris RL, Coates KW, Bump RC. Clinical predictors of urinary incontinence in women. Am J Obstet Gynecol 1997;177:262–7.
30. Clarke B. The role of urodynamic assessment in the diagnosis of lower urinary tract disorders. Int Urogynecol J 1997;8:196–200.

Cited By

This article has been cited 2 time(s).

Obstetrics & Gynecology
Risk of Stress Urinary Incontinence Twelve Years After the First Pregnancy and Delivery
Viktrup, L; Rortveit, G; Lose, G
Obstetrics & Gynecology, 108(2): 248-254.
PDF (275) | CrossRef
Obstetrics & Gynecology
Effect of Antenatal Pelvic Floor Muscle Training on Labor and Birth
Bø, K; Fleten, C; Nystad, W
Obstetrics & Gynecology, 113(6): 1279-1284.
PDF (186) | CrossRef
© 2003 The American College of Obstetricians and Gynecologists