SEARCH

SEARCH BY CITATION

Keywords:

  • Bladder training;
  • faecal incontinence;
  • pelvic floor dysfunction;
  • pelvic floor muscle training;
  • pelvic organ prolapse;
  • randomised controlled trial;
  • urinary incontinence

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

Objective

To determine the long-term (12-year) effects of a conservative nurse-led intervention for postnatal urinary incontinence.

Design

Follow-up of a randomised controlled trial.

Setting

Community-based intervention in three centres (in the UK and New Zealand).

Population

A cohort of 747 women with urinary incontinence at 3 months after childbirth, of whom 471 (63%) were followed up after 12 years.

Methods

Women were randomly allocated to active conservative treatment after delivery (pelvic floor muscle training and bladder training), or to a control group receiving standard care.

Main outcome measures

Prevalence of urinary incontinence (primary outcome) and faecal incontinence, symptoms and signs of prolapse, and performance of pelvic floor muscle training at 12 years.

Results

The significant improvements relative to controls that had been found in urinary incontinence (60 versus 69%; risk difference, RD, −9.1%; 95% confidence interval, 95% CI, −17.3 to −1.0%) and faecal incontinence (4 versus 11%; RD −6.1%; 95% CI −10.8 to −1.6%) at 1 year did not persist for urinary incontinence (83 versus 80%; RD 2.1%; 95% CI −4.9 to 9.1%) or faecal incontinence (19 versus 15%; RD 4.3%; 95% CI −2.5 to 11.0%) at the 12–year follow up, irrespective of incontinence severity at trial entry. The prevalence of prolapse symptoms or objectively measured pelvic organ prolapse also did not differ between the groups. In the short term the intervention motivated more women to perform pelvic floor muscle training (83 versus 55%), but this fell in both groups by 12 years (52 versus 49%).

Conclusions

The moderate short-term benefits of a brief nurse-led conservative treatment for postnatal urinary incontinence did not persist. About four-fifths of women with urinary incontinence 3 months after childbirth still had this problem 12 years later.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

About 20–30% of women have postpartum urinary incontinence,[1] and 3–5% have faecal incontinence.[2] Controversy exists about how to manage these problems. A Cochrane review on the effects of pelvic floor muscle training (PFMT) for the prevention and treatment of incontinence in antenatal and postnatal women suggested that in those who are incontinent after delivery, PFMT was better than no treatment, and that more intensive exercising was best.[3] These conclusions were based on just three randomised controlled trials,[4-6] one of which was the first report of the current trial.[6] A further three trials investigated the effects of PFMT in antenatal women who were incontinent before delivery.[7-9] Overall, whether treatment was started during or after pregnancy, fewer women were incontinent in the PFMT groups (229/735, 31%) than in the control groups (362/806, 45%) at up to 1 year after delivery (relative risk, RR, 0.58; 95% confidence interval, 95% CI, 0.39 to 0.87)[3]; however, none of the other trials reported follow-up after the first year.

We previously reported that significant reductions relative to controls were found in urinary incontinence (60 versus 69%; risk difference, RD, −9.1%; 95% CI −17.3 to −1.0%) and faecal incontinence (4 versus 11%; RD −6.1%; 95% CI −10.8 to −1.6%) at 1 year,[6] but these did not persist at 6 years for urinary incontinence (76 versus 79%; RD −3.0%; 95% CI −10.2 to 4.1%) or faecal incontinence (12 versus 13%; RD −0.6%; 95% CI −6.4 to 5.1%).[10] At 1 year, the intervention had motivated more women to perform PFMT (83 versus 55%), but this fell to 50% in both groups by 6 years. Our objective was to report follow-up at 12 years, and we have included prolapse outcomes for the first time.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

Women were recruited from an original cohort of around 10 000 women who delivered in 1993–94 in three centres (Aberdeen, Birmingham, and Dunedin). Ethical approval was obtained in each centre for the original study and follow-up at 6 and 12 years. Information for the remainder of the cohort of women who were not randomised (non-trial women) has been published separately.[11]

All women had urinary incontinence 3 months after their ‘index’ delivery, and were randomised by remote concealed computer allocation, stratified by delivery mode, parity, and frequency of incontinence at trial entry. They received either enhanced conservative management at home or standard care in the control group. The intervention was delivered by trained healthcare professionals (who were not physiotherapists), and comprised one-to-one instruction in PFMT, with bladder training if indicated, on three occasions at 5, 7 and 9 months after the birth. Study methods, interventions, and outcomes at 1 and 6 years were reported previously.[6, 10]

We contacted the women again at 12 years by postal questionnaire. The primary outcome was the prevalence of urinary incontinence at 12 years after randomisation. Secondary outcomes included practice of PFMT, faecal incontinence, and prolapse symptoms using the Pelvic Organ Prolapse Symptom Score (POP-SS, a seven-item instrument scored as 0–28, with higher scores representing more symptoms).[12] Information about prolapse was not obtained at trial entry or at 1 or 6 years. Data regarding obstetric history were collected from the women.

We were also able to examine a subsample of women to measure pelvic organ prolapse using the objective Pelvic Organ Prolapse Quantification system (POP–Q).[13] These women responded positively to an invitation in the questionnaire, indicating that they were willing to be examined. In total, 35% of the women who responded to the questionnaire were examined, although a few women changed their minds or could not find a convenient time to attend a clinic after volunteering.

Comparisons used the Chi-squared test for proportions or the Student's t-test for continuous outcomes, with results expressed as either risk differences (RD) or mean differences (MD), with 95% confidence intervals. Analysis was by original group assignment (intention-to-treat analysis) without substitution for missing data from non-respondents. We carried out subgroup analyses amongst women with less and more severe urinary incontinence, and with and without faecal incontinence, at trial entry.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

Of the 747 women recruited, 524 (70.1%) had responded 1 year later,[6] and 516 (69.1%) responded at 6 years.[10] At 12 years, the response was 471 women (63.1%), but allowing for women whom we were unable to approach (75 were known to have moved, requested no further contact, or died; Figure 1), the response rate from those contacted was 471/672 (70.1%). In all, 404 women responded at all three follow-up points (54.1%). There were no statistically significant differences at trial entry between the women recruited (n = 747) and those who responded at 12 years (n = 471) except for age (12-year responders were 1.5 years older, P < 0.001; Table 1). The randomised groups were, however, similar to each other at 12 years (Table 1).

Table 1. Baseline comparison of respondents at trial entry (3 months after index delivery) and at 12 years later
All respondentsIntervention respondentsControl respondents
Entry N = 74712 years N = 471Non-respondents at 12 years, N = 276 P Entry N = 37112 years N = 230Entry N = 37612 years N = 241
  1. Index delivery indicates the birth that preceded enrolment and randomisation to the trial, 3–4 months before trial entry.

  2. a

    Based on respondents’ totals of 735 at entry and 463 at 12 years.

  3. b

    Based on respondents’ totals of 734 at entry and 463 at 12 years.

  4. c

    Includes forceps, breech, and vacuum delivery.

  5. d

    Based on respondents’ totals of 708 at entry and 450 at 12 years.

Age at trial entry

(mean years, SD)

29.6(5.0)30.2(4.6)28.6(5.5)<0.00129.8(4.9)30.2(4.6)29.4(5.1)30.2(4.6)
Primiparity at trial entry (%) a 37.138.235.30.42436.738.837.637.7
Method of index delivery 3 months before trial entry (%) b
Spontaneous vaginal delivery78.577.580.10.41478.377.578.677.5
Assisted vaginal deliveryc 13.814.712.20.33113.714.513.814.8
Caesarean section7.87.87.70.9908.07.97.67.6
Severe urinary incontinence at trial entry (at least once a week %) 55.753.559.40.11457.157.854.349.4
Any faecal incontinence at trial entry (%) d 15.714.018.60.11516.315.315.112.7
image

Figure 1. Flow diagram of progress through phases of randomised trial of conservative treatment for postnatal incontinence.

Download figure to PowerPoint

The mean length of follow-up from index birth was 12.7 years (range 11–14 years). The mean age of the women at this time was 43.3 years (range 30–57 years). Most women had had two children (236, 50%), whereas 7.9% had had only one child, and a further 42% had three or more children. Over half of the women (254, 53.9%) had only experienced spontaneous vaginal deliveries for all of their births, 20 (4.2%) had only had caesarean sections, 131 (27.8%) had at least one forceps delivery, and 26 (5.5%) had at least one vacuum delivery. The remainder (40, 8.5%) had a mixture of spontaneous vaginal deliveries and caesarean sections.

Pelvic floor exercises

Table 2 describes women's reported practice of pelvic floor muscle exercises at 12 years. The differences in the number of women performing any exercises, performing exercises every day, and number of contractions (which had been present at 1 year) were not evident at 6 or 12 years (Table 2).[6, 10]

Table 2. Reported use of pelvic floor muscle training (PFMT) at 12 years after index delivery
InterventionControlDifference (95% CI) 2P a
  1. a

    Two-sided P value.

All participants N = 230 N = 241
Performing any PFMT, n/N (%)118/228 (52)118/241 (49)

2.8% (−6.3 to 11.8%)

P = 0.545

Performing PFMT every day, n/N (%)15/227 (7)20/241 (8)

−1.7% (−6.4 to 3.1%)

P = 0.485

Number of contractions per day, mean (SD) N 9.6 (13.7) 2149.9 (15.3) 212

−0.3 (−3.1 to 2.5)

P = 0.836

Urinary incontinence

The difference in any urinary incontinence between the groups seen at 1 year (60 versus 69%) were no longer apparent by 12 years (83 versus 80%; P = 0.555; Table 3), although the prevalence had increased in both groups.[6] There was also no difference in the prevalence of more severe incontinence, defined as at least once per week, nor in the proportions using pads (Table 3). More women in the intervention group had already had an operation for urinary incontinence, but this finding is based on small numbers (14 versus 5; 6.5 versus 2.2%; P = 0.028). Finally, there was no evidence that women with more severe incontinence at trial entry were more likely to benefit at 12 years than those with minimal incontinence, although the prevalence at 12 years was marginally lower in the latter group (Table 3). No adverse effects related to the active intervention were reported at any time. Notably, four-fifths of the incontinent women were also incontinent 12 years later.

Table 3. Urinary incontinence at 12 years after index delivery
Group   
OutcomeIntervention N = 230Control N = 241Difference (95% CI) 2P a
  1. a

    Two-sided P value.

  2. b

    Women's rating of how much of a problem urinary incontinence was for them: from 0 (not at all) to 10 (a great deal). If no UI, ‘problem rating’ re-coded to 0.

Any urinary incontinence, n (%)190 (83)194 (80)

2.1% (−4.9 to 9.1%)

P = 0.555

At least once per week, n (%)106 (46)104 (43)

2.9% (−6 to 11.9%)

P = 0.522

Using pads, n (%)45 (20)35 (15)

5% (−1.7 to 11.8%)

P = 0.145

Overall rating of severity of urinary incontinenceb

Mean (SD)

6.03 (13.8)6.04 (15.7)

−0.007 (−2.69 to 2.68)

P = 0.996

Operation for urinary incontinence, n (%)

N = 217

14 (6.5)

N = 226

5 (2.2)

4.2%

(0.4 to 8%)

P = 0.028

Participants with urinary incontinence less than once per week at trial entry N = 97 N = 122
Any urinary incontinence at 12 years, n (%)74 (76)96 (79)

−2.4% (−13.6 to 8.8%)

P = 0.673

Participants with urinary incontinence at least once per week at trial entry N = 133 N = 119
Any urinary incontinence at 12 years, n (%)116 (87)98 (82)

4.9% (−4 to 13.8%)

P = 0.284

Faecal incontinence

Some women with urinary incontinence at trial entry also had faecal incontinence. The lower prevalence of faecal incontinence in the intervention group (4 versus 11%; P = 0.012) at 1 year was not sustained at 12 years (19 versus 15%; P = 0.215; Table 4). The pattern was similar for more severe faecal incontinence (Table 4), and amongst subgroups characterised by whether or not they had faecal incontinence at trial entry (Table 4). A small number of women had received treatment for faecal incontinence (pads or plugs, constipatory drugs, surgery, or physiotherapy): nine in the intervention group and seven in the controls, and 12 of the 16 still had some residual faecal incontinence. Notably, two-fifths of the women who did have concomitant urinary and faecal incontinence at trial entry had persistent faecal incontinence at 12 years.

Table 4. Faecal incontinence at 12 years after index delivery
InterventionControlDifference (95% CI) 2P a
  1. a

    Two-sided P value.

  2. b

    ‘Severe’ defined as faecal incontinence occurring sometimes, often, or always.

All participants N = 228 N = 240
Any faecal incontinence, n (%)43 (19)35 (15)

4.3% (−2.5 to 11%)

P = 0.215

Severe faecal incontinenceb, n (%)6 (3)6 (3)

0.1% (−2.7 to 3%)

P = 0.928

Participants with faecal incontinence at trial entry N = 34 N = 29
Any faecal incontinence, n (%)14 (41)12 (41)

−0.2% (−24.6 to 24.2%)

P = 0.987

Participants without faecal incontinence at trial entry N = 186 N = 198
Any faecal incontinence, n (%)26 (14)19 (10)

4.4% (−2.1 to 10.8%)

P = 0.183

Pelvic organ prolapse

There were no significant differences between the randomised groups at 12 years in terms of prolapse symptoms, severity of symptoms, or number of women having prolapse surgery (Table 5).

Table 5. Prolapse outcomes at 12 years after index delivery
InterventionControlDifference (95% CI) 2P a
  1. a

    Two-sided P value.

  2. b

    Pelvic Organ Prolapse Symptom Score (0, no symptoms; 28, maximum symptom score).[8]

  3. c

    Women's rating of how much their prolapse symptoms interfere with their everyday life, from 0 (not at all) to 10 (a great deal). If no prolapse symptoms were recorded on POP–SS, the ‘severity rating’ was recoded to 0.

  4. d

    Stage 2 classically defined as leading edge at –1 cm inside hymen or beyond on POP–Q.[9]

  5. e

    ’Significant prolapse’ defined as leading edge at hymen or beyond (≥ 0 cm).

All participants
POP–SSb, Mean (SD)

N = 230

3.14 (3.36)

N = 241

3.3 (3.47)

−0.16 (–0.77 to 0.46)

P = 0.622

Overall rating of severity of prolapse symptomsc, Mean (SD)

N = 215

0.33 (1.39)

N = 224

0.26 (1)

0.07 (–0.16 to 0.29)

P = 0.561

Surgery for prolapse, n/N (%)5/226 (2.2)7/237 (3)

−0.7% (–3.6 to 2.1%)

P = 0.615

Women examined at 12 years, n/N (%)76/230 (33)89/241 (37)
POP–Q stage ≥ 2d, n/N (%)49/76 (64)57/89 (64)

0.4% (−14.2 to 15.1%)

P = 0.954

POP–Q stage at hymen or beyonde, n/N (%)23/76 (30)26/89 (29)

1.0% (−12.9 to 15.0%)

P = 0.883

We were able to examine a subsample of 35% of the women in the trial (165 women) to identify objective signs of prolapse using the POP–Q system.[11, 13] These women were comparable to non-examined women from the trial on age, parity, proportion with severe urinary incontinence, or any faecal incontinence at trial entry, although fewer had a caesarean section as their index delivery (data not shown). They were significantly more likely to have reported symptoms of prolapse in their questionnaire at 12 years than the non-examined women from the trial (with a higher score for prolapse symptoms on the POP–SS: 3.72 versus 2.95; MD –0.77, 95% CI −1.41 to −0.12). They were similar in terms of prevalence of urinary or faecal incontinence (data not shown). Nevertheless, there were no significant differences between the trial groups in terms of objectively measured prolapse (Table 5), defined as: descent to 1 cm above the hymen or worse (classical definition of stage 2 or higher; 64.5 versus 64.0%; P = 0.954), or descent to the hymen or beyond as a marker of a clinically important prolapse (30.3 versus 29.2%; P = 0.883).

Compared with the original cohort from which the women included in the trial were drawn, the women in the trial had slightly more prolapse symptoms at 12 years (with a mean score of 3.22 on POP–SS amongst women from the trial, compared with 2.58 amongst non-trial women; P < 0.001), and a higher proportion had objective evidence of prolapse at the hymen or beyond (30 versus 22%; P = 0.061).[11] Similarly, more women from the trial than non-trial women had urinary incontinence (81.5 versus 48.6%; P < 0.001) and faecal incontinence (16.6 versus 12.4%; P = 0.015) at 12 years.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

Main findings

Although at the 1–year follow-up the intervention group had lower rates of both urinary and faecal incontinence, these differences did not persist at 6 or 12 years. This was also true for the women who initially had more frequent urinary incontinence, and for those who initially had concomitant faecal incontinence.

There were no statistically significant differences in any prolapse outcomes according to the trial groups at 12 years, which was the first time that prolapse was assessed in this study. At least 30% of those examined had objectively measured prolapse at the hymen or beyond, but there was no evidence of difference between the randomised groups.

Four-fifths of those with urinary incontinence at trial entry still had urinary incontinence 12 years later. Two-fifths of women with faecal incontinence at trial entry also reported it 12 years later, and another one in ten reported it as a new symptom.

Strengths and limitations

A particular strength of this study was the long length of follow-up after randomisation, the inclusion of symptomatic assessment of prolapse symptoms using the POP–SS (which had been developed over the intervening time), and the opportunity to examine a subsample of women for objective evidence of pelvic organ prolapse. The women in the trial who volunteered to be examined did have some evidence of higher rates of pelvic floor dysfunction based on self-reported prolapse symptoms than those who were not examined, although the POP–SS values were low and the small difference (0.77 points) may not be clinically significant.[14]

The response rate of 63% at 12 years was slightly lower than at 1 and 6 years; however, this did not differ between the trial arms, and those responding were similar to non-responders in terms of age, parity, delivery mode, and urinary or faecal incontinence at trial entry (Table 1). The original trial was powered to detect a 10% difference in the rate of urinary incontinence. The confidence interval for the difference ranged from −10 to +10% (difference 2.1%; 95% CI −4.9 to +9.1%): this narrow confidence interval rules out anything but a small effect, which is unlikely to be of clinical significance. We are thus confident that the intervention did not have any long-term effect in reducing incontinence. The results for the other outcomes also supported this finding of no difference.

Women with faecal incontinence all had co-existing urinary incontinence. A further 7% of women had been continent for urine but incontinent for faeces at trial entry, but were not eligible for the trial because our inclusion criterion was presence of urinary incontinence. Our trial findings regarding faecal incontinence cannot therefore be generalised to women who do not also have urinary incontinence.

Women in the trial had a higher prevalence of urinary or faecal incontinence, or prolapse, at 12 years compared with women from the original cohort who were not randomised. This may suggest that women who chose to enter the trial because they had some urinary incontinence at 3 months after delivery, no matter how little, were more likely to have symptoms of pelvic floor dysfunction 12 years later.

Interpretation

The moderate short-term intervention effect (about one in 11 fewer women having urinary incontinence, and one in 16 fewer having faecal incontinence) did not persist in the long term. Similarly, the earlier difference in the performance of pelvic floor muscle exercises was not sustained, which is likely to account for the lack of difference in urinary, faecal, and potentially prolapse outcomes.

The effect might have persisted for longer if there had been continuing reinforcement or if it had been carried out by pelvic floor physiotherapists rather than trained but non-specialist nurses, but these possibilities would need to be tested by further controlled trials of different strategies to increase adherence to PFMT. These findings are also disappointing because PFMT and bladder training have the merit of being simple to teach and perform (although expensive in terms of teaching time by health professionals), and have few if any adverse effects.

The findings are, however, in line with Cochrane reviews of PFMT,[3, 15] and of bladder training for urinary incontinence and conservative treatment for faecal incontinence.[16, 17] Our results represent the longest follow-up of any trial so far.

Further information about the long-term follow-up of the cohort of women from whom the trial women were recruited is presented in other publications.[11, 18, 19] The progression and persistence of symptoms in this trial population is similar to that in the rest of the cohort from which the women were drawn. For example, at 3 months after delivery, 33.6% of women had urinary incontinence, compared with 52.7% 12 years later.[19] Faecal incontinence increased from 8.2 to 12.9%: of those who reported faecal incontinence at 3 months, 43% also reported it at 12 years.[18]

Conclusion and recommendations

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

The moderate short-term benefits of a brief nurse-led conservative treatment for postnatal urinary incontinence did not persist. There is therefore a need to identify conservative management strategies for urinary and faecal incontinence and prolapse that have longer term effects than those seen in this study, and then to test them rigorously by randomised controlled trials with long-term follow-up.

Contribution to authorship

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

All authors in the ProLong Study Group contributed to the design, conduct, analysis or writing-up of the study. C.G., C.M., S.H., A.E., R.L., P.H., and D.W. contributed to the analysis and writing of the current report, and approved the final version.

Details of ethics approval

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

The 12–year follow-up study was approved by the Multicentre Research Ethics Committee (MREC), ref. no. 06/MRE10/25. C.G. was the guarantor.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

WellBeing of Women, grant number RG 819/06, Royal College of Obstetricians and Gynaecologists, London, UK; Health Research Council of New Zealand. All authors are independent of the funders.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

We are grateful to the women who have faithfully returned their questionnaires over the 12–year period, and those who volunteered for examination. Anne-Marie Rennie, AMcD, Jane Harvey, and Jane Cook provided nursing and administrative support for the original trial. Anne-Marie Rennie conducted the 6–year follow-up in Aberdeen and Birmingham, and Ros Herbison conducted the 6–year follow-up in Dunedin. N.D. carried out the 12–year follow-up in Dunedin; K.R., C.B., C.G., and Margaret MacNeil carried out the 12–year follow-up in Aberdeen, and C.M. and P.T.–H. carried out the 12–year follow-up in Birmingham. G.McP provided invaluable help with database management. The original trial was funded by a project grant from BirthRight (now WellBeing of Women). Supplementary funding from the New Zealand Lottery Grant Board, and the Health Services Research Unit, University of Aberdeen, enabled the 6–year follow-up to be carried out. Grants were obtained from Wellbeing of Women/RCOG for the UK follow-up at 12 years and from the Health Research Council of New Zealand for the follow-up in New Zealand.

Disclaimer

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information

The Health Services Research Unit is supported by a core grant from the Chief Scientist Office of the Scottish Government Health Directorates. The views and opinions expressed in this article are those of the authors and do not necessarily reflect those of the funders or the Chief Scientist Office.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information
Journal Club

Scenario

At your urogynaecology clinic, a patient presented with postpartum urinary incontinence 3 months after the delivery of her first baby. She asked, ‘am I going to have incontinence for the rest of my life? Is there anything that would help?’

Description of research

ParticipantsWomen who had urinary incontinence 3 months after the delivery of their children.
InterventionThree one-to-one pelvic floor muscle training (PFMT) sessions, with bladder training if indicated, at 5, 7, and 9 months after birth
ComparisonStandard care
Outcomes

Primary outcome: The prevalence of urinary incontinence at 12 years after randomisation.

Secondary outcomes: Practice of PFMT, faecal incontinence, the Pelvic Organ Prolapse Symptom Score (POP-SS), Prolapse Organ Prolapse Quantification (POP–Q).

Study designRandomised controlled trial

Discussion points

  • Was the intervention (e.g. intensity and duration) in this clinical trial the same as in other similar studies?
  • What is the purpose of stratification at randomisation in this clinical trial?
  • Was it necessary to stratify for mode of delivery and centre of care?
  • Was the primary outcome in this study appropriate? Would you select different outcome measures?
  • Was compliance satisfactory? How could this have affected the results?
  • Can you suggest any strategy to improve the response rate of a similar study in the future?
  • Take-home message: can you briefly summarise the results of this study? Has this study changed your practice?

EYL Leung

Women's Health Research Unit, Queen Mary,

University of London, London, UK

Join us at #BlueJC: You can collaborate with us to host a #BlueJC session. Find out more on our Journal Club page at www.BJOG.org. Follow @BJOGTweets to stay updated on scheduled #BlueJC sessions. For the background of #BlueJC, please refer to BJOG 2013 May;120:657–60. http://bit.ly/10VaiRZ

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and recommendations
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. Disclaimer
  14. References
  15. Supporting Information
FilenameFormatSizeDescription
bjo12473-sup-0001-suppinfo.pptapplication/mspowerpoint697KData S1. Powerpoint slides summarising the study.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.