Content uploaded by Cathryn Glazener
Author content
All content in this area was uploaded by Cathryn Glazener
Content may be subject to copyright.
doi:10.1136/bmj.38320.613461.82
2005;330;337-; originally published online 22 Dec 2004; BMJ
and P Don Wilson
Cathryn M A Glazener, G Peter Herbison, Christine MacArthur, Adrian Grant
incontinence: six year follow up
management of postnatal urinary and faecal
Randomised controlled trial of conservative
http://bmj.com/cgi/content/full/330/7487/337
Updated information and services can be found at:
These include:
Data supplement
http://bmj.com/cgi/content/full/bmj.38320.613461.82/DC1
"Figure"
References
http://bmj.com/cgi/content/full/330/7487/337#BIBL
This article cites 9 articles, 3 of which can be accessed free at:
Rapid responses
http://bmj.com/cgi/eletter-submit/330/7487/337
You can respond to this article at:
http://bmj.com/cgi/content/full/330/7487/337#responses
free at:
2 rapid responses have been posted to this article, which you can access for
service
Email alerting
box at the top right corner of the article
Receive free email alerts when new articles cite this article - sign up in the
Topic collections
(131 articles) Urology •
(783 articles) Pregnancy •
Articles on similar topics can be found in the following collections
Notes
http://www.bmjjournals.com/cgi/reprintform
To order reprints of this article go to:
http://bmj.bmjjournals.com/subscriptions/subscribe.shtml
go to: BMJTo subscribe to
on 27 June 2005 bmj.comDownloaded from
Papers
Randomised controlled trial of conservative management of
postnatal urinary and faecal incontinence: six year follow up
CathrynMAGlazener, G Peter Herbison, Christine MacAr thur, Adrian Grant, P Don Wilson
Abstract
Objective To determine the long term effects of a conservative
nurse-led intervention for postnatal urinary incontinence.
Design Randomised controlled trial.
Setting Community based intervention in three centres in the
United Kingdom and New Zealand.
Participants 747 women with urinary incontinence at three
months after childbirth, of whom 516 were followed up again at
6 years (69%).
Intervention Active conservative treatment (pelvic floor muscle
training and bladder training) at five, seven, and nine months
after delivery or standard care.
Main outcome measures Urinary and faecal incontinence,
performance of pelvic floor muscle training.
Results Of 2632 women with urinary incontinence, 747
participated in the original trial. The significant improvements
relative to controls in urinary (60% v 69%) and faecal (4% v
11%) incontinence at one year were not found at six year follow
up (76% v 79% (95% confidence interval for difference in
means –10.2% to 4.1%) for urinary incontinence, 12% v 13%
( − 6.4% to 5.1%) for faecal incontinence) ir respective of
subsequent obstetric events. In the short term the intervention
had motivated more women to perform pelvic floor muscle
training (83% v 55%) but this fell to 50% in both groups in the
long term. Both urinary and faecal incontinence increased in
prevalence in both groups during the study period.
Conclusions The moderate short term benefits of a brief
nurse-led conservative treatment of postnatal urinary
incontinence may not persist, even among women with no
further deliveries. About three quar ters of women with urinary
incontinence three months after childbirth still have this six
years later.
Introduction
About 20-30% of women have postpartum urinary inconti-
nence,
1
and 3-5% have faecal incontinence.
2
Controversy exists
about how to manage these problems. A Cochrane review
suggested that pelvic floor muscle training is better than no
treatment and that more intensive exercising is best.
3
The trials
included in the review were in general populations of older
women with stress incontinence with only two randomised trials
among postnatal women.
45
Pelvic floor muscle training is routinely taught in maternity
care, but its effectiveness in prevention of incontinence is
questionable.
6
A Cochrane review of prevention of incontinence
was inconclusive,
6
and trials targeting antenatal
78
and postpar-
tum
910
women had conflicting findings. Maximum follow up was
one year. These inconsistent findings, together with electrophysi-
ological evidence of postpartum nerve recovery,
11
suggested that
targeting women with persistent incontinence might be more
effective.
4
Our multi-centre, randomised controlled trial of nurse-led
enhanced conservative interventions in 747 women who
reported urinary incontinence three months after delivery
5
showed that at one year fewer women in the intervention group
had urinary incontinence (60% v 69%, P = 0.037) and fewer had
faecal incontinence (4% v 11%, P = 0.012). We examined the long
term outcomes to determine whether these differences persisted.
Methods
The women were recruited in 1993-4 in three centres (Aberdeen,
Birmingham, and Dunedin). All the women had urinary inconti-
nence three months after delivery and were randomised by
remote concealed computer allocation, stratified by method of
delivery, parity, and baseline frequency of incontinence, to
receive either enhanced conservative management or standard
care. Study methods, interventions, and one year outcomes have
been reported previously.
5
In the intervention group nurses
assessed urinary incontinence and gave advice on pelvic floor
muscle training exercises at five, seven, and nine months after the
index delivery, supplemented (for those with urge symptoms)
with bladder training at seven and nine months. Women in the
control group received standard postnatal management, which
usually included a brief description on pelvic floor exercises.
We contacted the women again by postal questionnaire at six
years. The primary outcome was the incidence of urinary incon-
tinence. Secondary outcomes included use of pelvic floor
exercises and faecal incontinence. We also collected data on
obstetric histories.
We compared data with
2
tests or Student’s t tests as appro-
priate and have expressed results as absolute or relative
differences with 95% confidence intervals. Analysis was by origi-
nal group assignment. We used pre-specified secondary stratified
analyses using logistic regression to explore possible differential
effects attributed to severity and type of urinary incontinence at
baseline, and whether or not women reported a fur ther delivery
by the six year follow up. We expected that any effect was more
likely to be sustained in the subgroups of women who had more
severe incontinence at baseline, had stress urinary incontinence,
and had no further deliveries.
Results
Of the 747 women recruited, 524 (70.1%) responded one year
later. At six years, we were unable to approach five women
Cite this article as: BMJ, doi:10.1136/bmj.38320.613461.82 (published 22 December 2004)
BMJ Online First bmj.com page 1 of 4
on 27 June 2005 bmj.comDownloaded from
known to have moved or who had died (figure), giving a
corrected response rate of 516/742 (69.5%). In all, 394
responded at both follow ups (52.7%). There were no significant
baseline differences between the women recruited (n = 747) and
those who responded at one year (n = 524) or six years (n = 516)
in age, parity, method of index delivery, and severity of urinary or
faecal incontinence (table 1). The randomised groups were also
similar.
The mean length of follow up from the index birth was 5.9
years (range 5.0-7.1). The mean age at this time was 35.9 years
(range 23-50). Parity was unchanged in 275/512 (54%). Thus,
237 women had had at least one further delivery (46.9% and
45.6% in the randomised groups), of whom 27 and 7,
respectively, had had two and three more births.
Table 2 describes women’s reported use of pelvic floor mus-
cle exercises. At six years, the differences in the number
performing exercises and the number of contractions (which
had been present at one year) had disappeared (table 2). The dif-
ference in urinary incontinence between the groups seen at one
year had also disappeared by six years (table 3). There was no
evidence that the effect at one year was more likely to persist
among women with more severe incontinence at baseline or
according to type of incontinence or further deliveries. No
women reported any adverse effects related to the active
intervention. Three quarters of the incontinent women were still
incontinent six years later.
Faecal incontinence
Some women with urinary incontinence at baseline also had fae-
cal incontinence. The reduction in faecal incontinence in the
intervention group (4% v 11%, P = 0.012) at one year was not
sustained at six years (12% v 13%, P = 0.932, table 4). The pattern
was similar for women with more severe faecal incontinence
(table 4) and among subgroups characterised by whether or not
they had faecal incontinence at entry into the trial (table 4).
Notably, two fifths of those women had persistent faecal inconti-
nence.
Incontinent women entering trial at baseline (n=747)
Intervention group (n=371)
Did not receive
intervention*
(n=55)
1993-4
1994-5
(1 year follow up)
1999-2000
(6 year follow up)
Received
intervention
(n=316)
Received
intervention
(n=376)
Control group (n=376)
Responded
(n=279, 75%)
No response
(n=92)
Responded
(n=263, 71%
of original
total)
No
response
(n=39)
No
response
(n=66)
Responded
(n=253, 67%
of original
total)
No
response
(n=61)
No
response
(n=39)
Responded
(n=245, 65%)
No response
(n=131)
Lost to follow
up
†
(n=2)
Lost to follow
up
†
(n=1)
Lost to follow
up
†
(n=2)
Lost to follow
up
†
(n=0)
*New pregnancy, too busy, moved away, family problems, incontinence resolved, non-stress or non-urge incontinence.
†
Known to have moved or died.
Progress of women through phases of trial
Table 1 Baseline comparison of respondents at entry (three months after delivery) and at one year, and six years after index delivery
All respondents Intervention respondents Control respondents
Entry (n=747)
One year
(n=524)
Six years
(n=516) Entry (n=371)
One year
(n=279)
Six years
(n=263) Entry (n=376)
One year
(n=245)
Six years
(n=253)
Mean (SD) age at entry
(years)
29.6 (5.0) 30.0 (4.8) 30.0 (4.7) 29.8 (4.9) 30.2 (4.7) 30.1 (4.7) 29.4 (5.1) 29.7 (4.8) 29.8 (4.7)
Primiparity at entry* (%) 37.1 38.2 38.8 36.7 37.7 39.1 37.6 38.8 38.5
Method of index delivery† (%):
Standard vaginal 78.5 79.5 80.0 78.3 79.3 79.5 78.6 79.6 80.6
Assisted vaginal 13.8 13.0 13.5 13.7 13.4 13.6 13.8 12.5 13.4
Caesarean 7.8 7.6 6.5 8.0 7.2 7.0 7.6 7.9 6.1
Severe urinary incontinence
(at least once/week) (%)
56.1 52.1 54.3 57.7 53.8 57.8 54.5 46.8 50.6
Any faecal incontinence at
entry (%)‡
15.7 15.4 13.9 16.3 15.5 14.1 15.1 15.3 13.8
*Based on respondents’ totals of 735, 516, and 505 at entry, one year, and six years.
†Based on respondents’ totals of 734, 516, and 505 at entry, one year, and six years.
‡Based on respondents’ totals of 708, 500, and 488 at entry, one year, and six years.
Papers
page2of4 BMJ Online First bmj.com
on 27 June 2005 bmj.comDownloaded from
Discussion
Principal findings
Although at one year after delivery women who received active
pelvic floor muscle training (with bladder training if appropriate)
had lower rates of both urinary and faecal incontinence than
women who received standard care, these differences did not
persist to six years. There were no subgroups in which the inter-
vention was more successful long term, even among those who
had not had a further pregnancy.
Three quarters of those with urinary incontinence at baseline
still had urinary incontinence six years later. Two fifths of women
with faecal incontinence at baseline still reported it six years
later, and another one in 20 developed it as a new symptom.
Strengths and weaknesses
There was about a 30% non-response rate at both one and six
years. Those responding, however, were similar to non-
responders in terms of age, parity, delivery mode, and baseline
urinary or faecal incontinence. The results were the same when
we considered only those who responded at both times. The
study was powered to detect a 10% difference. Although it was
not possible to rule out a difference of this size (95% confidence
interval of the difference –10.2% to 4.1%), results for other out-
comes support the finding of no difference.
Women with faecal incontinence all had coexisting urinary
incontinence. A further 329/4818 (7%) were continent to urine
but incontinent to faeces at baseline but were not eligible for the
trial because our inclusion criterion was urinary incontinence.
Our findings regarding faecal incontinence cannot therefore be
generalised to women who do not also have urinary
incontinence.
Meaning of the findings
The moderate short term intervention effect (about 1 in 11 fewer
women having ur inary incontinence and 1 in 16 fewer having
faecal incontinence) was not present in the long term. Similarly,
the earlier difference in the performance of pelvic floor muscle
exercises had also disappeared, which could account for the lack
of difference in urinary and faecal outcomes. The effect might
have persisted for longer if there had been continual
reinforcement or if it had been carried out by physiotherapists
rather than nurses, but these suggestions would need to be tested
by controlled trial.
These findings are also disappointing because pelvic floor
muscle training 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, however, are in line with Cochrane
reviews of pelvic floor muscle training
3
and bladder training
12
for
urinary incontinence and conservative treatment for faecal
incontinence.
13
Our results represent the longest follow up of any
trial so far. There is a need to identify conservative strategies for
both urinary and faecal incontinence that have longer term
effects than those seen in this study and then to test them rigor-
ously by randomised controlled trials with long term follow up.
Table 2 Reported use of pelvic floor muscle training (PFMT) six years after
index delivery
Intervention Control
Difference (95% CI), P
value*
All participants 263 253 —
No (%) performing any
PFMT
131 (50) 126 (50) 0.0% (−8.6 to 8.6),
1.00
No (%) performing
PFMT daily
17 (6) 29 (12) −5.2% (−10.1 to –0.2),
0.060
Median (IQR) No of
contractions/day†
0.0 (0.0 to 15.0) 0.0 (0.0 to 15.0) 0 (0 to 0), 0.866
*For 2×2 tables P values are given with continuity correction.
†In 262 women in intervention group and 249 in control group. Medians are based on all
women—that is, counting those not doing PFMT as having frequency of contractions of 0.
Table 3 Urinary incontinence at six years after index delivery with subgroup
analyses. Figures are numbers (percentages) of women unless stated
otherwise
Intervention Control
Difference in % (95%
CI), P value*
No of women 263 253
Any urinary incontinence 201 (76) 201 (79) −3.0 (−10.2 to 4.1),
0.471
At least once/week 100 (38) 99 (39) −1.1 (−9.5 to 7.3), 0.867
Using pads 72 (28) 59 (24) 3.7 (−3.9 to 11.3), 0.395
Mean (SD) overall rating
of severity†
35.3 (25.1) 31.4 (23.8) 3.9 (−1.0 to 8.8), 0.120
Severity of incontinence at baseline:
At least once per week:
No of women 152 128
Any urinary
incontinence
122 (80) 102 (80) 0.6 (−8.8 to 10.0), 1.0
Less than once a week:
No of women 111 125
Any urinary
incontinence
79 (71) 99 (79) −8.0 (−19.1 to 3.0),
0.201
Type of incontinence at baseline:
Stress incontinence:
No of women 149 126
Any urinary
incontinence
111 (75) 97 (77) −2.5 (−12.6 to 7.7),
0.736
Urge incontinence:
No of women 41 39
Any urinary
incontinence
31 (76) 32 (82) −6.4 (−24.3 to 11.4),
0.667
Mixed incontinence:
No of women 71 80
Any urinary
incontinence
57 (80) 66 (83) −2.2 (−14.7 to 10.2),
0.888
Delivery after index birth:
Women without further deliveries:
No of women 138 137
Any urinary
incontinence
108 (78) 109 (80) −1.3 (−10.9 to 8.3),
0.907
Women with further deliveries:
No of women 122 115
Any urinary
incontinence
91 (75) 91 (79) −4.5 (−15.2 to 6.2),
0.501
*For 2×2 tables P values are given with continuity correction.
†Visual analogue scale 0=“no problem at all” to 100=“Can’t think of anything worse” (n=195
in intervention group, n=190 in control group).
Table 4 Faecal incontinence six years after index delivery. Figures are
numbers (percentages) of women unless stated otherwise
Intervention Control
Difference (95% CI), P
value*
All women 261 248
Any faecal incontinence 32 (12) 32 (13) −0.6% (−6.4 to 5.1), 0.932
Severe faecal incontinence† 15 (6) 8 (3) 2.5% (−1.1 to 6.1), 0.248
Faecal incontinence at baseline:
No of women 35 33
Any faecal incontinence 15 (43) 13 (39) 3.4% (−19.9 to 26.8),
0.965
No faecal incontinence at baseline:
No of women 213 203
Any faecal incontinence 12 (6) 15 (7) −1.8% (−6.5 to 3.0), 0.598
*
For 2×2 tables P values are given with continuity correction.
†Severe defined as occurring sometimes, often, or always.
Paper s
BMJ Online First bmj.com page 3 of 4
on 27 June 2005 bmj.comDownloaded from
We are grateful to Anne-Marie Rennie, Alison McDonald, Jane Harvey, and
Jane Cook, who provided nursing and administrative support to the origi-
nal trial. Anne-Marie Rennie conducted the six year follow up in Aberdeen
and Birmingham, and Ros Herbison in Dunedin. Magnus McGee provided
extra statistical help.
Contributors: All authors contributed to the design, analysis, and writing up
of the study. CMAG is guarantor.
Funding: Birthright (now WellBeing), Royal College of Obstetricians and
Gynaecologists, London; New Zealand Lottery Grant Board; Health
Services Research Unit, University of Aberdeen. The health services
research unit is funded by the Chief Scientist Office of the Scottish Execu-
tive Health Department.
Competing interests: None declared.
Ethical approval: Ethical approval was obtained in each centre for the origi-
nal and follow up studies from each local ethics committee.
1 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.
2 MacArthur C, Bick DE, Keighley MR. Faecal incontinence after childbirth. Br J Obstet
Gynaecol 1997;104:46-50.
3 Hay-Smith EJC, Bo K, Berghmans LCM, Hendriks HJM, de Bie RA, van Waalwijk van
Doorn ESC. Pelvic floor muscle training for urinary incontinence in women. Cochrane
Database Syst Rev 2001;(1): CD001407.
4 Wilson PD, Herbison GP. A randomized controlled tr ial of pelvic floor muscle exercises
to treat postnatal urinary incontinence. Int Urogynecol J 1998;9:257-64.
5 Glazener CMA, Herbison GP, Wilson PD, MacArthur C, Lang GD, Gee H, et al.
Conservative management of persistent postnatal urinary and faecal incontinence: a
randomised controlled trial. BMJ 2001;323:593-6.
6 Hay-Smith EJC, Herbison P, Mørkved S. Physical therapies for prevention of urinary
and faecal incontinence in adults. Cochrane Database Syst Rev 2002;(2):CD003191.
7 Hughes P, Jackson S, Smith P, Abrams P. Can antenatal pelvic floor exercises prevent
postnatal incontinence [abstract]. Neurourol Urodyn 2001;20:447-8.
8 Salvesen KA,.Morkved S. Randomised controlled trial of pelvic floor muscle training
during pregnanc y. BMJ 2004;329:378-80.
9 Sleep J, Grant A. Pelvic floor exercises in postnatal care. Midwifery 1987;3:158-64.
10 Chiarelli P, Cockburn J. Promoting urinary continence in women after delivery:
randomised controlled trial. BMJ 2002;324:1241.
11 Snooks SJ, Badenoch DF, Tiptaft RC, Swash M. Perineal nerve damage in genuine
stress urinary incontinence. An electrophysiological study. Br J Urol 1985;57:422-6.
12 Wallace SA, Roe B, Williams K, Palmer M. Bladder training for urinary incontinence in
adults. Cochrane Database Syst Rev 2004;(4):CD001308.
13 Norton C, Hosker G, Brazzelli M. Biofeedback and/or sphincter exercises for the treat-
ment of faecal incontinence in adults. Cochrane Database Syst Rev 2000;(2):CD002111.
(Accepted 23 November 2004)
doi 10.1136/bmj.38320.613461.82
Health Services Research Unit, University of Aberdeen Medical School,
Foresterhill, Aberdeen AB25 2ZD
Cathryn M A Glazener senior clinical research fellow
Adrian Grant director
Department of Preventive and Social Medicine, Dunedin School of Medicine,
University of Otago, PO Box 913, Dunedin, New Zealand
G Peter Herbison associate professor
Department of Women’s and Children’s Health, Dunedin School of Medicine,
University of Otago
P Don Wilson professor of obstetrics and gynaecology
Department of Public Health and Epidemiology, University of Birmingham, PO
Box 363, Edgbaston, Birmingham B15 2TT
Christine MacArthur professor of maternal and child epidemiology
Correspondence to: CMAGlazenerc.glazener@abdn.ac.uk
What is already known on this topic
Childbirth is a major cause of urinary and faecal
incontinence in women
A one year follow up study showed that women who were
randomised to active pelvic floor muscle training (with
bladder training if appropriate) were more often continent
than women in a control group and were also more likely
to be performing pelvic floor exercises
What this study adds
At six years after the index delivery, three quarters of the
women still had urinary incontinence and over 10% had
faecal incontinence
The benefits seen at one year were no longer apparent
Only half the women were still performing any pelvic floor
exercises, irrespective of initial group assignment
Papers
page4of4 BMJ Online First bmj.com
on 27 June 2005 bmj.comDownloaded from
