See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/6616821

Systematic reviews of bladder training and

voiding programmes in adults: a synopsis of
findings from data analysis and outcomes using
metastudy techniques

ARTICLE in JOURNAL OF ADVANCED NURSING FEBRUARY 2007

Impact Factor: 1.69 DOI: 10.1111/j.1365-2648.2006.04097.x Source: PubMed

CITATIONS DOWNLOADS VIEWS

26 104 76

4 AUTHORS, INCLUDING:

Joan Ostaszkiewicz Jill Milne

Deakin University 19 PUBLICATIONS 203 CITATIONS
30 PUBLICATIONS 140 CITATIONS
SEE PROFILE

SEE PROFILE

Sheila Wallace
University of Aberdeen
39 PUBLICATIONS 1,180 CITATIONS

SEE PROFILE

Available from: Joan Ostaszkiewicz

Retrieved on: 21 July 2015
JAN REVIEW PAPER

Systematic reviews of bladder training and voiding programmes in

adults: a synopsis of findings from data analysis and outcomes using
metastudy techniques
Brenda Roe1, Joan Ostaszkiewicz2, Jill Milne3 & Sheila Wallace4

Accepted for publication 31 July 2006

1
Brenda Roe PhD RN RHV FRSH R O E B . , O S T A S Z K I E W I C Z J . , M I L N E J . & W A L L A C E S . ( 2 0 0 7 ) Systematic
Professor of Health Sciences reviews of bladder training and voiding programmes in adults: a synopsis of findings
Faculty of Health and Applied Social from data analysis and outcomes using metastudy techniques. Journal of Advanced
Sciences,
Nursing 57(1), 1531
Liverpool John Moores University,
doi: 10.1111/j.1365-2648.2006.04097.x
Liverpool, UK

2
Joan Ostaszkiewicz MSN RN Abstract
Coordinator Aged Care and Rehabilitation Title. Systematic reviews of bladder training and voiding programmes in adults: a
Nursing Research synopsis of findings from data analysis and outcomes using metastudy techniques
The Peter James Centre, Aim. This paper reports a comparison of the data analysis and outcomes from
Deakin University, four Cochrane systematic reviews on bladder training and voiding programmes
Melbourne, Australia
for the management of urinary incontinence using metastudy descriptive tech-
3 niques to inform clinical practice, generate new ideas and identify future research
Jill Milne PhD RN
Postdoctoral Fellow directions.
School of Nursing, Background. Bladder training is used for cognitively and physically able adults to
University of Alberta, regain continence by increasing the time interval between voids. Prompted void-
Edmonton, Canada ing, habit retraining and timed voiding, collectively known as voiding pro-
grammes, are generally used for people with cognitive and physical impairments
4
Sheila Wallace BSc MSc in institutional settings. Bladder training and voiding programmes feature as
Search Coordinator
common clinical practice for the management of urinary incontinence.
Cochrane Incontinence Reviews Group,
Health Services Research Unit,
Methods. A synopsis of four Cochrane systematic reviews that included rand-
University of Aberdeen, omized controlled trials on bladder training, prompted voiding, habit retraining
Aberdeen, UK and timed voiding was undertaken using metastudy techniques for the synthesis of
qualitative research, and has provided a discursive comparison and contrast of the
Correspondence to Brenda Roe: meta-data analysis and outcomes of these reviews.
e-mail: b.h.roe@ljmu.ac.uk Results. Frequency of incontinence was the most common and constant outcome
measure of effectiveness in the reviews. Limited data were available on other
health outcomes, change in dependency status, quality of life and cost-effective-
ness. The systematic review on bladder training included different types of urinary
incontinence, whereas those on voiding programmes did not differentiate the type
of incontinence. There is evidence on the effectiveness of bladder training but
long-term follow up studies are needed. Evidence on the effectiveness of voiding
programmes is limited and not available for many outcomes.
Conclusion. Future research needs to consider the theory underpinning interven-
tions for bladder training and voiding programmes for urinary incontinence and
should incorporate recognized quality research designs, established outcomes and
long-term follow up. It is unclear whether health outcomes for people with

 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd 15

B. Roe et al.
comorbidities, cognitive and physical impairments will improve if extensive
diagnostic and assessment investigations are undertaken.
Keywords: bladder training, metastudy, nursing, systematic review, urinary
incontinence, voiding programmes
Introduction
In this paper we present the findings on data analysis and
outcomes from four Cochrane systematic reviews on bladder
training, prompted voiding, habit retraining and timed
voiding. Metastudy techniques developed for the synthesis
of qualitative research comprise selection and appraisal of
primary research, meta theory, meta methods and meta data
analysis (Paterson et al. 2001) and provide a useful method
for the synopsis, descriptive comparison and contrast of
systematic reviews of randomized controlled trials (RCTs).
These techniques are useful for revisiting the theory under-
pinning interventions, their design, methods and outcomes
with a view to generating new ideas for research and future
directions for empirical study. The metastudy and findings for
selection and appraisal of primary research, meta theory and
meta methods of these systematic reviews have been reported
in an accompanying paper (Roe et al. 2007). The metastudy
technique for meta data analysis and the related findings on
outcomes are reported in this paper.
Background
Systematic reviews of RCTs that have investigated the value
of bladder training, prompted voiding, habit retraining and
timed voiding for the management of urinary incontinence in
adults are available in the Cochrane Library (Eustice et al.
2002, Ostaszkiewicz et al. 2004a, 2004b, Wallace et al.
2004). These behavioural interventions commonly form
bladder re-education strategies and voiding programmes
frequently used by nurses for the management of urinary
incontinence in community, clinic and institutional settings
(Hadley 1986, Kennedy 1992). The development of these
interventions dates back to the mid to late 1970s with
bladder training being the earliest and aimed at people who
are physically and cognitively able (Willington 1975, Frewen
1978). Habit retraining evolved at a similar time and aimed
to be delivered by motivated staff to people with cognitive
and physical disabilities (Clay 1978). Timed voiding (Castl-
eden & Duffin 1981) and prompted voiding (Hu 1989) are
also used for people with cognitive and physical disabilities
by care staff, and are common in institutional settings
(Hadley 1986, Kennedy 1992). Definitions of the terms used
16  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
for bladder training, prompted voiding, habit retraining and
timed voiding are available in the reviews themselves, and an
accompanying paper (Roe et al. 2007), which also provides a
synopsis based on meta study techniques for selection and
appraisal of primary research, and findings for meta-theory
and the meta-methods for the four systematic reviews.
Synopses are particularly useful in summarizing research
for clinicians, which they would not do for themselves by
reading directly the Cochrane Reviews, which can each be up
to 78 pages long (for example, Wallace et al. 2004). This
synopsis using metastudy techniques and the accompanying
paper (Roe et al. 2007) has allowed discursive comparison,
contrast and summary of the four Cochrane systematic
reviews and provides an accessible source for clinicians and
researchers. It is a requirement that clinical practice is based
upon sound evidence, and these interventions are the main
behavioural practices used internationally by nurses, as the
basis of every day care for the management of urinary
incontinence.
The study
Aim
The aim of the study was to provide a synopsis of the findings
from meta-data analysis and outcomes from the metastudy of
four Cochrane systematic reviews on behavioural interven-
tions which include, bladder training, prompted voiding,
habit retraining and timed voiding. The purpose was to
compare and contrast outcomes in relation to the interven-
tions and study designs to inform clinical practice, generate
new ideas and future directions for research.
Methodology
Methods for the synthesis of qualitative research are
evolving (Thorne et al. 2004) and a useful framework and
techniques have been developed for metastudy, which allow
discursive comparison and contrast (Paterson et al. 2001).
While the four systematic reviews in this paper are
quantitative research, RCTs in particular, such metastudy
techniques allow a useful synopsis to be undertaken. In this
paper, the references to each of the trials and their related
JAN: REVIEW PAPER
publications adhere to the Cochrane Library convention of
giving a single name and year. Findings on meta-data
analysis and outcomes are the focus of this paper. Selection
and appraisal of primary research, meta-theory and meta-
method of interventions have been presented elsewhere (Roe
et al. 2007).
Meta-data analysis methods
The meta-data analysis strategy adhered to that developed by
Paterson et al. (2001) and involved identifying the analysis
methods of the 22 trials that were eligible for inclusion in the
four systematic reviews, the development of a filing and
coding system, tabulation, categorization of the data and a
discussion and interpretation of the findings. Data were
categorized according to the comparisons that were
addressed in the four systematic reviews and the associated
outcomes. Outcomes were grouped as follows:
Frequency of incontinence.
Severity of incontinence.
Other objective outcomes.
Subjective outcomes.
In addition to examining the reported results within each
systematic review, we examined the specific units of measure
that were used to evaluate outcomes. We report data for
those outcomes wherein trialists reported statistically signi-
ficant differences and/or where this was demonstrated in the
relevant systematic review metaview, as well areas where
there is insufficient data. Where possible, data were com-
bined.
Findings
Frequency of incontinence
Methods employed by trialists to calculate the effectiveness of
the intervention on the frequency of incontinence varied
between trials and included measures of:
Number of episodes of incontinence within a defined per-
iod of time (i.e. usually 24-hour duration) measured at
variable point estimates (i.e. at the end of treatment and
3 months later).
Number of individuals with reductions in the incidence of
daytime or night-time incontinence.
Number of individuals with no improvement of wet epi-
sodes.
Number of individuals improved or cured immediately
after treatment, 3 months later and at a mean of 32 years
follow-up.
Individual and group rates of incontinence (Table 1).
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Systematic reviews of bladder training and voiding programmes in adults
Bladder training
Of the 10 eligible trials on bladder training, statistically sig-
nificant benefits were associated with the intervention on four
reported outcomes:
number of individuals improved immediately after blad-
der training for other incontinence (Fantl 1991) with
bladder training compared with no bladder training (45/60
vs. 15/65; Relative Risks (RR) 315; 85% confidence
interval (CI),
number of incontinence episodes for individuals with stress
incontinence immediately after treatment with bladder
training combined with pelvic floor muscle exercises and
biofeedback (n = 42; mean 92; SD 115) compared with
pelvic floor muscle exercises and biofeedback alone
(n = 46, mean 87; SD 100) (pairwise comparison
P = 0003, as reported by trialists) (Wyman 1998),
number of incontinent episodes for individuals with other
incontinence immediately after treatment with bladder
training combined with pelvic floor muscle exercises and
biofeedback (n = 16; mean 58; SD 95) compared with
pelvic floor muscle exercises and biofeedback alone
(n = 18; mean 119; SD 127) (pairwise comparison
P = 0003, as reported by trialists) (Wyman 1998),
number of individuals cured of other incontinence
immediately after treatment with bladder training com-
bined with pelvic floor muscle exercises and biofeedback
compared with pelvic floor muscle exercises and biofeed-
back alone [Weighted Mean Difference (WMD) 249; 95%
CI 118 to 526) (Wyman 1998).
Prompted voiding
Of the six trials included in the systematic review on
prompted voiding, two reported statistically significant
reductions in the mean number of episodes of incontinence
per individual for those receiving prompted voiding com-
pared with those who did not receive prompted voiding
(WMD 025; 95% CI 075 to 025) and (WMD 200;
95% CI 263 to 137) (Hu 1989, Schnelle 1989). These
data were pooled to provide a total sample of 257
participants (Gp 1: 127 Gp 2: 130) which demonstrated
statistical significance favouring prompted voiding (WMD
093; CI 95% 132 to 053), however it was noted that this
was largely due to the relatively large estimated effect in the
trial conducted by Schnelle (1989).
Habit retraining
Of the three trials that comprised the systematic review on
habit retraining, each reported reductions in the incidence of
incontinence from baseline to outcome for individuals in the
treatment groups. Where comparable data were available, no
17
B. Roe et al.

Table 1 The frequency of incontinence

Relevant comparison Unit of measure Relevant trial Results Analysis

BT vs. no BT Number of individuals improved Fantl Gp 1: (60) 45 RR 315;

immediately after treatment for (1991) Gp 2: (65) 15 95% CI 198 to 502
other incontinence Statistical significance
favouring BT
Number of individuals cured Gp 1: (60) 7 RR 368;
immediately after treatment Gp 2: (63) 2 95% CI 079 to 1699
for other incontinence
Number of incontinent episodes Gp 1: (45)
per week for stress incontinence mean = 1000 (SD 1200)
at the end of treatment Gp 2: (43)
mean = 1900 (SD 1900)
Number of incontinent episodes Gp 1: (7)
per week for urge incontinence mean = 500 (SD 600)
at the end of treatment Gp 2: (7)
mean = 1800 (SD 1400)
Number of incontinent episodes Gp 1: (8)
per week for mixed incontinence mean = 700 (SD 800)
at the end of treatment Gp 2: (12)
mean = 2000 (SD 1200)
BT vs. other treatment Number of incontinent episodes for Wyman Gp 1: (46)
(i.e. behavioural, stress incontinence immediately after (1998) mean = 125 (SD 83)
physical or treatment ended Gp 2: (48)
psychological) mean = 87 (SD 1000)
Number of incontinent episodes for Gp 1: (19)
other incontinence immediately after mean = 62 (SD 91)
treatment ended Gp 2: (18)
mean = 119 (SD 127)
Number of incontinent episodes for Gp 1: (62)
other incontinence 3 months after mean = 100 (SD 120)
the end of treatment Gp 2: (65)
mean = 94 (SD 140)
Number of individuals cured of other Gp 1: (68) 12 RR 141
incontinence immediately after Gp 2: (64) 8 95% CI 062 to 323
treatment
Number of individuals cured of other Gp 1: (62) 10 RR 081
incontinence at 3 months Gp 2: (65) 13 95% CI 038 to 170
Number of individuals cured of other Gp 1: (45) 4 RR 462
incontinence at a mean 32 years Gp 2: (52) 1 95% CI 054 to 3987
follow-up
BT another treatment Number of incontinent episodes Wyman Gp 1: (42) The authors reported
(i.e. PFMT biofeedback) for stress incontinence immediately (1998) mean = 92 (SD 115) statistical significance
vs. that other treatment after treatment ended Gp 2: (46) favouring BT PFMT
alone mean = 87 (SD 100) biofeedback (P = 0003)
Number of incontinent episodes Gp 1: (16) The authors reported
for other incontinence immediately mean = 58 (SD 95) statistical significance
after treatment ended Gp 2: (18) favouring BT PFMT
mean = 119 (SD 127) biofeedback (P = 0003)
Number of incontinent episodes Gp 1: missing data This data could not be
for other incontinence 3 months Gp 2: (65) used in the standard
after the end of treatment mean = 94 (SD 140) metaview analysis due to
missing data

18  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd

JAN: REVIEW PAPER Systematic reviews of bladder training and voiding programmes in adults

Table 1 (Continued)

Relevant comparison Unit of measure Relevant trial Results Analysis

Number of individuals cured Gp 1: (19) 61 WMD 249;

of other incontinence Gp 2: (8) 64 95% CI 118 to 526
immediately after treatment Statistical significance
favouring BT
PFMT biofeedback
Number of individuals cured Gp 1: (60) 16 RR 133;
of other incontinence Gp 2: (65) 13 95% CI 070 to 253
at 3 months
Number of individuals cured Gp 1: (48) 8 RR 867;
at a mean 32 years follow-up Gp 2: (52) 1 95% CI 113 to 6675
PV vs. no PV Mean number of episodes Hu (1989) Gp 1: (64) Pooled data
of incontinence per mean = 165 (SD 161) WMD 093;
individual per 24 hours Gp 2: (67) CI 95% 132 to 053
mean = 190 (SD 129) Total sample
WMD 025; 95% Gp 1: 127
CI 075 to 025 Gp 2: 130
Schnelle (1989) Gp 1: (63) Statistical significance
mean = 210 (SD 160) favouring PV
Gp 2: (63)
mean = 410 (SD 200)
WMD 200; 95%
CI 263 to 137
Number of individuals Hu (1989) Gp 1: (65) 17 Pooled data
with no improvement Gp 2: (68) 24 OR 059; 95%
of wet episodes OR 065: 95% CI 031 to 114
CI 031 to 136 Total sample
Schnelle (1983) Gp 1: (11) 2 Gp 1: (82)
Gp 2: (10) 3 Gp 2: (84)
OR 054; 95% Favouring PV
CI 008 to 383 but not statistically
Surdy (1992) Gp 1: (6) 1 significant
Gp 2: (6) 3
OR 025; 95%
CI 003 to 252
PV other vs. Mean number of Ouslander Treatment mean = This data could not be
PV alone episodes of incontinence (1995) 66 (SD 37) vs. 77 used in the standard
(i.e. oxybutynin) (SD 37) metaview tables as it was
reported at the end of the
study period and not given
separately for the first phase
Number of wet checks Treatment This data could not be used
percentage = in the standard metaview
20% (SD 11) vs. tables as it was reported
24% (SD 11) at the end of the study
period and not given
separately for the first phase
HT other vs. usual care Mean number episodes Colling Gp 1: (51) This data could not be
of incontinence (1992) mean = 45 used in the standard
(SD missing data) metaview tables due to
Gp 2: (37) missing data
mean = 55
(SD missing data)

 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd 19

B. Roe et al.

Table 1 (Continued)

Relevant comparison Unit of measure Relevant trial Results Analysis

Mean number of episodes Colling (2003) Gp 1: (32) WMD 022;

of incontinence
TV other vs. Number of individuals with
usual care reductions in the frequency
of daytime incontinence
Number of individuals with
reductions in the frequency
of night-time incontinence
Individual rates of
incontinence
Group rates of incontinence
mean = 400 (SD 263) 95% CI 032 to 075
Gp 2: (24)
mean = 343 (SD 259)
Jirovec & Templin Treatment Gps 1 & 2: (44) WMD 038;
(2001) mean = 037 (SD 028) 95% CI 085 to 009
Control Gp: (30)
mean = 049 (SD 036)
Tobin & Gp 1: (102) 40 RR 134;
Brocklehurst (1986) Gp 2: (89) 26 95% CI 090 to 201
Gp 1: (95) 39 RR 180;
Gp 2: (79) 18 95% CI 112 to 289
Smith (1992) End of phase 3 This data could not be
Gp 1: (Gp size not stated) used in the standard
38% metaview analysis
Gp 2: (Gp size not stated) due to missing data
missing data
End of phase 3 This data could not be
Gp 1: (Gp size not stated) used in the standard
21% metaview analysis
Gp 2: (Gp size not stated) due to missing data
85%

PFMT, pelvic floor muscle exercise training; BT, bladder training; PV, prompted voiding; TV, timed voiding.

statistically significant between group differences were noted treatment group compared with the control group (RR 180;
on this outcome. 95% CI 112 to 289) (Tobin & Brocklehurst 1986).

Timed voiding
Outcome data that could be used in a meta-view were
available for one of the two trials included in the systematic
review on timed voiding. There was a statistically significant
increase in the number of individuals with reductions in the
frequency of night-time incontinence for individuals in the
Severity of incontinence
Severity of incontinence was calculated as an outcome
measure by three trials (Colling et al. 1992, 2003, Tobin &
Brocklehurst 1986) (Table 2). Two of these trials described a
habit retraining protocol (Colling et al. 1992, 2003) whilst the

Table 2 The severity of incontinence

Relevant comparisons Unit of measure Relevant trial Results Analysis

HT other vs. Mean volume of incontinence Colling (1992) Gp 1: (51) This data could not be
usual care per individual per 24 hours mean = 600 cc (SD missing data) used in the standard
based on pad weighing Gp 2: (37) metaview analysis
mean = 650 cc (SD missing data) due to missing data
Colling (2003) Gp 1: (32) WMD 045;
mean = 292 cc (SD 202 cc) 95% CI 008 to 099
Gp 2: (24)
mean = 193 cc (SD 233 cc)
TV other vs. Number of individuals with Tobin & Gp 1: (65) 16 RR 101;
usual care improvement in volume of Brocklehurst Gp 2: (45) 11 95% CI 052 to 196
incontinence based on pad (1986)
weighing

TV, timed voiding; cc, cubic centimetres.

20  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd

JAN: REVIEW PAPER
other focused on timed voiding (Tobin & Brocklehurst 1986).
Outcome data were based on the mean volume of incon-
tinence per individual and/or the number of individuals with
reductions in the volume of incontinence. Where comparable
data were available, no statistically significant between group
differences were noted on this outcome (Table 2).
Other objective outcomes
Other objective outcomes that were reported included, self-
initiated requests for toileting, the frequency of voiding,
health measures (adverse drug effects, altered skin integrity,
urinary tract infection) and health economic measures (cost
of implementation and cost savings) (Table 3).
Self-initiated request for toileting
Whilst this outcome was evaluated for each eligible trial on
prompted voiding (Schnelle 1983, Hu 1989, Schnelle 1989,
Surdy 1992, Ouslander et al. 1995), outcome data were
limited to one trial (Schnelle 1989). This reported a statis-
tically significantly higher number of self-initiated requests
for toileting for the group exposed to prompted voiding
compared with the group that did not receive prompted
voiding (OR 190: 95% CI 151 to 229) (Table 3).
Frequency of voiding
Evaluation of the frequency of voiding was limited to three
trials that focused on bladder training (Jarvis 1981, Fantl
1991, Colombo et al. 1995). Measures used to determine the
effectiveness of the intervention included (a) the number of
day-time and night-time micturitions per week for stress
incontinence, and for other incontinence immediately
following treatment and (b) the number of individuals cured
of daytime and night-time urinary frequency. Data were
based on self-completed urinary diaries in two of the three
trials. Between-group data were available for eight outcomes
and three of these supported a statistically significant differ-
ence in favour of bladder training:
An increase in the number of individuals cured of daytime
frequency with bladder training compared with oxybuty-
nin (RR 123 95% CI 083 to 181) (Colombo et al. 1995).
An increase in the number of individuals cured of night-time
frequency with bladder training compared with oxybutynin
(RR 224 95% CI 080 to 630) (Colombo et al. 1995).
An increase in the number of individuals cured of daytime
frequency with bladder training combined with flavoxate
hydrochloride 200 mg TDS imipramine 25 mg TDS
(RR 256; 95% CI 128 to 513) (Jarvis 1981) (Table 3).
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Systematic reviews of bladder training and voiding programmes in adults
Health measures
Adverse drug effects
Data were available on this outcome from two of the 10 trials
on bladder training (Jarvis 1981, Colombo et al. 1995), and
from one trial on prompted voiding (Ouslander et al.
1995)(Table 3). Colombo et al. (1995) reported a statistically
significant number of participants who experienced adverse
events as a consequence of oxybutynin compared with
bladder training alone (RR 003; 95% CI 000 to 044). Jarvis
(1981) also reported a higher number of patients experien-
cing adverse events in the group receiving medication (i.e.
flavoxate hydrochloride 200 mg, three times a day
(TDS) imipramine 25 mg TDS) compared with bladder
training alone, however this difference did not reach statis-
tical significance (RR 003; 95% CI 000 to 055) (Table 3).
In the single trial on the effects of oxybutynin in prompted
voiding, the top three side effects noted were dry mouth
(n = 22 active and n = 19 placebo), constipation (n = 16
active and n = 13 placebo) and incomplete bladder emptying
(n = 14 active and n = 16 placebo) (Ouslander et al. 1995).
Two participants dropped out due to urinary retention. The
trialists concluded that oxybutynin does not add to the
clinical effectiveness of prompted voiding in the majority of
nursing home residents with urge type urinary incontinence
although some individuals may be more responsive to
prompted voiding whilst on this drug (Ouslander et al.
1995) (Table 3).
Altered skin integrity
The incidence of skin rash and skin breakdown was evaluated
by one trial of habit retraining (Colling et al. 2003) (Table 3).
Research staff conducted tests using the Oregon Health Sci-
ences University skin breakdown and rashes evaluation tool
at weeks 3, 6, 9 and 12. The trialists reported significant
within group decreases in incidence of skin rash and skin
breakdown associated with treatment. No further analysis
could be applied to the data presented (Table 3).
Urinary tract infection
Baseline screening for and treatment of urinary tract infection
were consistent features of trials within each of the reviews.
One trial on habit retraining further evaluated the incidence
of urinary tract infection as an outcome (Colling et al. 2003)
(Table 3). Research staff collected clean-catch specimens of
urine at weeks 3, 6, 9 and 12, and sent them for culture and
sensitivity if positive on dipstick. The trialist reported that
there were no statistically significant differences. No further
analysis could be applied to the type of data presented
(Table 3).
21
B. Roe et al.

Table 3 Other objective outcomes

Objective outcomes Relevant comparisons Unit of measure Relevant trial Results Analysis

Self-initiated PV vs. no PV Number of Hu (1989, 1990) Gp 1: (65) This data could not
requests for self-initiated mean = 265 be used in the standard
toileting requests for (SD missing data) metaview analysis due
toileting vs. Gp 2: (68) to missing data
caregiver mean = 112
initiated toileting (SD missing data)
Schnelle (1983) Gp 1: (11) This data could not
mean = 200 be used in the standard
(SD missing data) metaview analysis due
Gp 2: (10) to missing data
mean = 023
(SD missing data)
Schnelle (1989, Gp 1: (63) Statistical
1990) mean = 270 significance
(SD 120) favouring PV
Gp 2: (63) OR 190: 95% CI 151
mean = 080 to 229
(SD 100)
Percentage of Surdy (1992) F (1, 7) = 1026, This data could not
independent P = 001 be used in the standard
requests for metaview analysis due
toileting to missing data
PV other vs. Number of Ouslander Treatment This data could not be
PV alone self-initiated (1995) mean = 06 used in the standard
(i.e. oxybutynin) requests for (SD 15) vs. metaview tables as it
toileting vs. 05 (SD 10) was reported at the end
caregiver initiated of the study period and
toileting not given separately for
the first phase
The frequency of BT vs. no BT Number of daytime Fantl (1991) Gp 1: (45) WMD 500; 95% CI
voiding micturitions per mean = 51 1179 to 179
week immediately (SD 1100)
after treatment for Gp 2: (43)
stress incontinence mean = 56
(SD 2000)
Number of Gp 1: (45) WMD 300; 95% CI
night-time mean = 500 514 to 086
micturitions per (SD 400)
week immediately Gp 2: (43)
after treatment for mean = 800
stress incontinence (SD 600)
Number of day-time Gp 1: (15) WMD 400; 95% CI
micturitions per mean = 5600 2238 to 1438
week immediately (SD 2000)
after treatment for Gp 2: (20)
other incontinence mean = 6000
(SD 3500)
Number of Gp 1: (15) WMD 100; 95% CI
night-time mean = 800 569 to 369
micturitions per (SD 700)
week immediately Gp 2: (20)
after treatment for mean = 900
other incontinence (SD 700)

22  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd

JAN: REVIEW PAPER Systematic reviews of bladder training and voiding programmes in adults

Table 3 (Continued)

Objective outcomes Relevant comparisons Unit of measure Relevant trial Results Analysis

BT vs. other Number of individuals Colombo Gp 1: (29) 20 RR 123 95% CI

treatment cured of daytime (1995) Gp 2: (32) 18 083 to 181
(i.e. oxybutynin) frequency
Number of individuals Gp 1: (18) 11 RR 224 95% CI
cured of night-time Gp 2: (11) 3 080 to 630
frequency
BT vs. other Number of individuals Jarvis (1981) Gp 1: (21) 17 RR 256; 95% CI
treatment (i.e. cured of daytime Gp 2: (19) 6 128 to 513
Flavoxate frequency Statistical
hydrochloride significance
200 mg TDS favouring BT
Imipramine
25 mg TDS)
Number of individuals Gp 1 (25) 19 RR 146; 95% CI
cured of night-time Gp 2: (25) 15 094 to 226
frequency
Health measures: BT vs. other Number of Colombo (1995) Gp 1: (37) 0 RR 003; 95% CI
adverse drug effects treatment individuals Gp 2: (39) 18 000 to 044
(i.e. oxybutynin) experiencing side Statistically
effects significance
favouring BT
BT vs. other treatment Jarvis (1981) Gp 1: (25) 0 RR 003; 95% CI
(i.e. Flavoxate Gp 2: (25) 14 000 to 055
Hydrochloride
200 mg
TDS Imipramine
25 mg TDS)
PV other vs. PV Ouslander (1995) Treatment with drug No further analysis
alone dry mouth n = could be applied to
(i.e. oxybutynin) 22 vs. 19, constipation this data
n = 16 vs. 13,
incomplete bladder
emptying n = 14 vs. 16
Skin integrity HT other vs. Incidence of skin Colling (2003) Author reported No further analysis
usual care rash & skin significant within could be applied to
breakdown group decrease in inci- this data
dence of skin rash and
skin breakdown asso-
ciated with treatment
Urinary tract HT other vs. Incidence of urinary Colling (2003) No significant No further analysis
infections usual care tract infections difference reported. could be applied to
this data
Health economic PV vs. no PV Average dollar cost Surdy (1992) $US 851 per patient No further analysis
measures: cost of implementing and reductions in other could be applied to
implementation/ intervention costs to value of $US this data
cost saving 2357
Average reduction in Hu (1989) Equal to one hour of
dollars per person time per patient per
based on 1 episode day
of incontinence
HT other vs. Average dollar cost Colling (2003) $US 23000 per
usual care savings based on a person per year
reduction of one incon-
tinent episode per day

BT, bladder training; PV, prompted voiding.

 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd 23

B. Roe et al.
Health economic measures
Cost of implementation/cost savings
The costs of implementing and/or the cost savings associated
with interventions were reported by two trials for prompted
voiding (Hu 1989, Surdy 1992) and one trial for habit
retraining (Colling et al. 2003) (Table 3). Both Hu (1989) and
Surdy (1992) included labour and supply costs. Hu (1989)
additionally undertook time and motion recordings of care
and training activities. This was supported further by inter-
rater reliability testing. On this basis Hu (1989) reported that
prompted voiding was reported to be equal to the cost of
1 hour of staff time per patient per day. Surdy (1992) esti-
mated the cost at $US 851 per patient however this was
offset by a cost saving of $US 2357 per patient. Based on a
single trial (Colling et al. 2003), the average cost savings
associated with habit retraining in a community based home
environment, were $US 23000 per person per year. Factors
that were considered in this calculation were confined to the
cost of supplies. Because of the nature of the data presented
on this outcome, no further analyses could be undertaken
(Table 3).
Subjective outcomes
Subjective outcomes that were evaluated were: (a) partici-
pants perception of improvement or cure of urge incontin-
ence, (b) participants perception of improvement of other
incontinence, (c) quality of life (QOL) of participants with
genuine stress incontinence, (d) quality of life of participants
with other incontinence, (e) number of individuals with
other incontinence who were satisfied, (e) caregivers
perception of burden/stress and (f) level of behavioural
capability (Table 4). Data collection instruments were:
The Incontinence Impact Questionnaire (Fantl 1991,
Wyman 1998).
The Urogenital Distress Inventory (Wyman 1998).
The Center for Epidemiological Studies Depression Scale
(Wyman 1998).
Visual Analogue Scales (Fantl 1991, Wyman 1998).
An Economic Burden Scale (Colling et al. 2003).
A Global Role Strain Scale (Colling et al. 2003).
A Preparedness for Caregiving Scale (Colling et al. 2003).
The Hadley, Wood, McCracken (HWM) Behavioral
Capabilities Scale for Older Adults (Colling et al. 2003).
Data on the perceptions of participants and QOL param-
eters were available in five trials on bladder training (Jarvis
1981, Fantl 1991, Colombo et al. 1995, Lagro-Janssen 1992,
Wyman 1998) and in one trial on habit retraining (Colling
et al. 2003) (Table 4). Data collection instruments that were
24  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
used to assess subjective outcomes in the review on bladder
training were designed for self-completion whilst the instru-
ments used in the trial on habit retraining were targeted to
community-based home caregivers.
Fantl (1991) reported statistically significant improve-
ments in the QOL score for participants with other
incontinence in the bladder training group compared with
controls. The only other trial that demonstrated a statisti-
cally significant result on subjective outcomes related to
satisfaction immediately after treatment (Wyman 1998).
Data favoured the control group which received pelvic floor
muscle exercises and biofeedback compared with bladder
training alone (RR 082; 95% CI 069 to 097). The
reviewer noted that this difference was not sustained at
3 months (Table 4).
Data reported on caregiver perception of burden/stress and
preparedness for caregiving were suggestive of an improve-
ment for participants in the intervention phase however no
further analyses could be applied to these data (Colling et al.
2003). Similarly, further analysis on the level of behavioural
capability was unable to be performed on the type of data
presented (Table 4).
Discussion
As reported in each of the relevant systematic reviews, the
analysis of results from each systematic review was limited by
a lack of comparable data and by variability in the quality of
available data. Similarly, the diversity of protocols and
methods (Roe et al. 2007) meant that combining data of
similar outcomes was not undertaken. Comparison across
reviews is cautioned on the basis of differences between
population groups in terms of their age, residential status and
health status and continence status. The limited data that
were available, however, highlight a number of pertinent
issues for further consideration.
It is noteworthy that the review on bladder training
differed from those on systematic voiding programmes in
that data were stratified for the four main diagnostic
groupings of urge incontinence, genuine stress incontinence,
mixed urinary incontinence and other incontinence. This
differentiation was not evident in trials on prompted voiding,
habit retraining and timed voiding with the exception of
timed voiding for individuals with urge incontinence (Tobin
& Brocklehurst 1986). Whilst arguably, choice of appro-
priate management is limited in the absence of an appropriate
diagnosis, experts debate the extent to which intensive
assessment and diagnostic investigations will improve the
health outcomes and QOL of individuals with comorbidities
and impairments, and urodynamic investigations are not
JAN: REVIEW PAPER Systematic reviews of bladder training and voiding programmes in adults

Table 4 Subjective measures

Relevant comparisons Unit of measure Relevant trial Results Analysis

BT vs. no BT Participants perception of Lagro-Janssen Gp 1: (9) 8 RR 1700; 95% CI

improvement of urge incontin- (1992) Gp 2: (9) 0 113 to 25656
ence at 2 months
Participants perception of cure of Lagro-Janssen Gp 1: (9) 1 RR 300; 95% CI
urge incontinence at 2 months (1992) Gp 2: (9) 0 014 to 6516
QOL for participants with other Fantl, 1991 Gp 1: (39) mean = 025 Authors reported a
incontinence (SD 029) significant difference
Gp 2: (39) mean = 050 in QOL favouring
(SD 059) BT group
BT vs. other treatment Participants perception of cure of Colombo (1995) Gp 1: (37) 27 RR 099; 95% CI
(i.e. oxybutynin) urge incontinence at the end of Gp 2: (38) 28 075 to 130
treatment
Participants perception of cure of Gp 1: (27) 26 RR 160; 95% CI
urge incontinence 6 months after Gp 2: (28) 16 121 to 234
treatment ended
Participants perception of Gp 1: (37) 34 RR 113; 95% CI
improvement at the end of treat- Gp 2: (38) 31 094 to 135
ment
BT vs. other treatment Participants perception of cure of Jarvis (1981) Gp 1: (25) 21 RR 150; 95% CI
(i.e. flavoxate hydro- urge incontinence at the end of Gp 2 (25) 14) 102 to 221
chloride 200 mg treatment phase
TDS Imipramine Participants perception of cure of Gp 1: (25) 21 RR 150 95% CI
25 mg TDS) urge incontinence 2 months after Gp 2: (25) 14 102 to 221
treatment ended
BT vs. other treatment QOL of participants with GSI at Wyman (1998) Gp 1: (47) WMD 1800; 95%
(i.e. behavioural, the end of treatment Gp 2: (45) CI 138 to 3737
physical or Participants perception of Gp 1: (66) 43 RR 086; 95% CI
psychological) improvement of other incontin- Gp 2: (63) 48 068 to 107
ence at the end of treatment
Participants perception of Gp 1: (60) 37 RR 088; 95% CI
improvement of other incontin- Gp 2: (64) 45 068 to 113
ence 3 months after treatment
ended
QOL of participants with other Gp 1: (20) WMD -2800;
incontinence at the end of Gp 2: (18) 95% CI 6839 to
treatment 1239
QOL of participants with other Gp 1: (60) WMD 680; 95% CI
incontinence 3 months after the Gp 2: (64) 1223 to 2563
end of treatment
Number of individuals satisfied Gp 1: (66) 48 RR 082; 95% CI
immediately after treatment Gp 2: (63) 56 069 to 097 Statisti-
ended for other incontinence cally signifi-
cance  favouring
PFMT biofeed-
back
Number of individuals satisfied Gp 1: (60) 47 RR 095; 95% CI
2 months after treatment ended Gp 2: (64) 53 080 to 113
for other incontinence
BT another treatment QOL of participants with GSI at Wyman (1998) Gp 1: (44) mean WMD 1800; 95%
(i.e. PFMT biofeed- the end of treatment score = 6320 (SD 4920) CI 3658058
back) vs. that other Gp 2: (45)
treatment alone mean = 8120 (SD
3960)

 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd 25

B. Roe et al.

Table 4 (Continued)

Relevant comparisons Unit of measure Relevant trial Results Analysis

Participants perception of Gp 1: (61) 55 RR 118; 95% CI

improvement of other incontin- Gp 2: (63) 43 101 to 139
ence at the end of treatment
Participants perception Gp 1: (58) 44 RR 108; 95% CI
of improvement of other incon- Gp 2: (64) 45 087 to 134
tinence 3 months after treatment
ended
QOL of participants Gp 1: (17) WMD 4720; 95%
with other incontinence at the mean = 6760 (SD CI 8703 to 737
end of treatment 4850)
Gp 2: (18)
mean = 11480 (SD
7030)
QOL of participants Gp 1: (58) WMD 1220; 95%
with other incontinence mean = 7280 (SD CI 3045 to 605
3 months after the end of treat- 5040)
ment Gp 2: (64)
mean = 8500
(SD = 5240)
Number of individuals Gp 1: (61) 57 RR 105; 95% CI
satisfied immediately after treat- Gp 2: (63) 56 094 to 117
ment for other incontinence
Number of individuals Gp 1: (58) 51 RR 106; 95% CI
satisfied 2 months after treatment Gp 2: (64) 53 092 to 123
for other incontinence
HT other vs. usual Caregivers perception of burden/ Colling (2003) Caregivers involved in No further analysis
care stress HT reported less stress could be applied to
at outcome this data
Level of behavioural capability No significant changes No further analysis
reported could be applied to
this data

PFMT, pelvic floor muscle exercise training; QOL, quality of life; GSI, genuine stress incontinence; BT, bladder training.

recommended for the frail elderly (Fonda et al. 2005). Tobin
and Brocklehurst (1986) reported that in a sample of 278
older adults, 78% of whom had cognitive impairment, only
three individuals were recommended for urodynamic proce-
dures and this was determined on the basis that the result was
likely to affect management and thus, be of direct benefit to
that individual.
At this stage, on the basis of the limited data, it is unclear if
certain types of incontinence are more responsive to system-
atic voiding programmes or if the value of these programmes
lies in none other than the fact that they provide dependent
individuals with regular voiding opportunities. This propo-
sition is not without precedence as other research has
identified immediate improvements associated with the
introduction of a prompted voiding protocol (Schnelle
1990). This observation led Schnelle to hypothesize that
responders were not developing new skills related to contin-
ence, instead they were responding to an environment that
offered increased opportunity for toileting assistance and
favourable reinforcement for continence (Schnelle 1990). If
the value of systematic voiding programmes pertains to the
fact that they provide individuals with otherwise limited
toileting opportunities, then clearly, there is little need for
intensive assessment procedures to identify the aetiological
type of incontinence.
The most common and consistent measure of effectiveness
was the frequency of incontinence. Although there were four
statistically significant measures that favoured bladder train-
ing compared with two trials with statistically significant
results in favour of prompted voiding, none on habit
retraining and one on timed voiding, it would be simplistic
at the very least to suggest that bladder training is more
effective than systematic voiding programmes. At the same
time, based on the sample of 1366 participants from 10
trials, the body of evidence on bladder training is greater
than that on prompted voiding (n = 479, 6 trials), habit

26  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd

JAN: REVIEW PAPER
retraining (n = 337, 3 trials) and timed voiding (n = 298, 2
trials).
Another measure of effectiveness was the frequency of
voiding; however, this was limited to the review on bladder
training. This is consistent with the aim of bladder training,
which is targeted to individuals with an overactive bladder,
urgency and/or urge incontinence and urinary frequency.
In terms of health outcomes, data were confined to the
incidence of adverse drug effects, skin breakdown and rash
and/or urinary tract infection although no further analysis
could be applied to these data. Other health factors that
could be considered in future trials could include; changes in
bladder capacity and post void residual volume. The first of
these two conditions is particularly relevant for individuals
undergoing bladder training. Assessment of a post void
residual is particularly relevant for individuals with poor
mobility and for those who often rely on another person for
toileting assistance.
The review on prompted voiding reported on three trials
(Schnelle 1983, 1989, Hu 1989) that evaluated the number
of self initiated requests for toileting. This is consistent with
the aim of prompted voiding which uses positive reinforce-
ment in an effort to increase individuals ability to discrim-
inate their continence status and initiate toileting. Whilst
individuals enrolled in the reviews on habit retraining and
timed voiding were similarly dependent, there was no
strategic effort made to alter this dependence. Instead,
efforts were directed toward accommodating this depend-
ence by identifying and mimicking the individuals unique
voiding pattern (habit retraining) or providing regular and
systematic voiding opportunities at times that are based on
the notion of a rhythmic bladder storage/voiding cycle
(timed voiding).
Cost data were evaluated by two trials on prompted voiding
(Hu 1989, Surdy 1992) and one trial on habit retraining
(Colling et al. 2003). It is noteworthy that the cost data on
prompted voiding are now more than 10 years old. No cost
data were available for bladder training or for timed voiding.
Evaluations of individuals perception of cure or improve-
ment were limited to the review on bladder training. Again,
this may relate to the higher cognitive status of individuals
involved in trials on bladder training compared to those who
were enrolled in trials on systematic voiding programmes.
Nevertheless, given the necessity for caregivers (i.e. either
informal or formal) to implement voiding programmes, their
perceptions, morale and QOL are of paramount importance.
The one trial that evaluated this outcome in caregivers of
community-dwelling and cognitively impaired older adults
stated that caregivers were less stressed at outcome (Colling
et al. 2003). It is also important to know the extent to which
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Systematic reviews of bladder training and voiding programmes in adults
these types of outcomes are related to supportive contact
from researchers compared with the intervention itself.
An important comparison that has not been addressed to
date is that of prompted voiding compared with either habit
retraining or timed voiding or both. In the absence of this
research it remains unclear which, if any, of the various types
of systematic voiding programmes are most effective and how
clinicians should choose between the three types. A lack of
outcome data also means that there is a lack of clarity about
the optimal methods for assessment, implementation and
evaluation. As noted previously, prompted voiding assess-
ment and evaluation procedures represent a function of the
intervention itself. These procedures involve intensive and
arguably, intrusive contact with individuals. Two trials on
habit retraining described the concurrent use of an electronic
data logger to obtain accurate data on individuals voiding
patterns thus mitigating some of the resource and privacy
issues associated with hourly or two hourly wet checks
conducted by other individuals.
More trials were included in the bladder training system-
atic review and there was an attempt to establish its
effectiveness for specific types of incontinence, while the
other systematic reviews lumped all urinary incontinence
together and did not differentiate between conditions. It is of
note that the more recent trials on bladder training, since the
1990s have more complex RCT designs (Fantl 1991, Wyman
1998 compared with Jarvis 1981). Such trials involve a
number of arms or interventions that have a cross over
element over time. They also combine a number of therapies
which reflects custom and clinical practice, but can ulti-
mately tease out the effectiveness of an individual therapy
under test, in this case bladder training. Based on the findings
and outcomes of Wyman (1998) combining therapies of
bladder training, pelvic floor muscle exercise training and
biofeedback was statistically significantly more effective
compared with PFMT and biofeedback alone, for stress
incontinence and other incontinence at the end of the
intervention. Trial designs and methods are evolving and are
becoming more complex combining therapies for both urge
and stress incontinence. These developments in design and
methods need to be considered when designing future studies.
A further observation from this metastudy was that none of
the trials in the systematic reviews undertook long-term
follow up of patients included in the studies to establish
benefits over time which again requires more complex study
designs, methods and incurs considerable expenses. However,
such long-term follow up of cohorts are required if we want
to not only explore the effectiveness of current and future
interventions and treatments but also to study prevention of
incontinence over the life course.
27
B. Roe et al.
What is already known about this topic
Systematic reviews on bladder training, prompted
voiding, habit retraining and timed voiding are avail-
able in the Cochrane Library.
There is evidence that bladder training is effective for
the management of urinary incontinence in adults and
suggestive evidence for the use of prompted voiding.
There is insufficient evidence to support habit retraining
or timed voiding, despite timed voiding, in particular,
being common custom and practice for the management
of urinary incontinence in people with cognitive and
physical impairments.
What this paper adds
It is unclear whether diagnostic and assessment inves-
tigations for urinary incontinence will improve health
outcomes for people with comorbidities, cognitive and
physical impairments.
Frequency of urinary incontinence is the most com-
monly cited outcome measure in randomized controlled
trials on bladder training and voiding programmes, but
limited or no data are available on other health out-
comes, change in dependency status, quality of life or
cost effectiveness.
The theory under-pinning bladder training and voiding
programmes for the management of urinary incontin-
ence, the research design and interventions, outcomes
and long-term follow up need to be considered in future
research.
It is a requirement that nurses, along with other healthcare
professionals base their clinical practice on established
evidence of effectiveness (Closs & Cheater 1999, Roe &
Moore 2000). This synopsis using metastudy techniques has
synthesized the current evidence on theory, methods, analysis
and outcomes of the four main behavioural interventions,
bladder training, prompted voiding, habit retraining and
timed voiding, used by nurses for the management of urinary
incontinence which is easily accessible to inform their clinical
practice, as well as identifying gaps where future research
endeavours should be targeted to advance clinical practice.
This synopsis also contributes to the dissemination of
evidence to its target audience of international nurses who
may not have access to or the time to read individual
Cochrane systematic reviews.
As discussed above, trial designs and methods are evolving
with more recent trials combining therapies and having more
28  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
complex designs than those from the 1970s and 1980s, with
greater emphasis on quality, power calculations, sample size,
blinding and intention-to-treat analysis (MRC 1998). Meth-
ods for the systematic review of research evidence are also
evolving with synthesis and meta-analysis of quantitative
research, notably RCTs (Egger et al. 2005), being electron-
ically disseminated via the Cochrane Library. Systematic
review, metastudy, metasynthesis or integration of qualitative
research evidence are also evolving and methods being
developed (Thorne et al. 2004, Whittemore & Knafl 2005),
although there is not a specific electronic library for their
dissemination. The challenge is now to systematically review
and integrate evidence from quantitative and qualitative
research, which The Joanna Briggs Foundation (2005) is
pioneering. Synopses of systematic reviews across related
subjects, diagnostic categories and groups of patients, such as
those with long-term conditions, for example, of quantitative
evidence, qualitative evidence or those integrating both
quantitative and qualitative evidence are warranted. This
study has provided a synopsis of systematic reviews of four
main behavioural interventions used by nurses for the
management of urinary incontinence. The individual reviews
took nearly 10 years to be completed and published and are
in the process of being added to and updated, which is a
requirement of the Cochrane Collaboration and Library.
None of these reviews received external funding, which is an
important consideration for future systematic reviews irres-
pective of the types of evidence being synthesized and
integrated. The metastudy framework and techniques while
developed for review of qualitative studies, provides a useful
methodology for the descriptive comparison and contrast of
systematic reviews of quantitative evidence.
Conclusion
The number of trials included in the systematic reviews and
evidence on the effectiveness of voiding programmes, in
particular, is very limited. There is evidence that bladder
training is effective for the management of urinary incontin-
ence in adults and is recommended for clinical practice,
although long-term follow up studies are required. Bladder
training has also been combined with pelvic floor muscle
exercises and other cognitive and lifestyle modification
techniques and so combination interventions need to be
studied in future research for the management of urinary
incontinence. There is limited evidence on the effectiveness of
prompted voiding for adults reliant on carers to avoid
incontinence and studies, which include follow up are
required. The systematic review on bladder training included
different types of urinary incontinence that were stratified
JAN: REVIEW PAPER
whereas those on voiding programmes made no attempt to
accurately diagnose the type of incontinence and may reflect
that patients had limited cognitive and physical abilities as
well as clinical custom and practice. It is unknown whether
intensive diagnostic and assessment investigations will
improve health outcomes or people with comorbidities and
impairments. It is also unknown whether systematic voiding
programmes are more effective for different types of urinary
incontinence and whether the more dependent individuals
they are targeted at might benefit equally from regular
voiding opportunities. The theories underpinning bladder
training and voiding programmes, their future classification
and interventions studies are warranted (Ostaszkiewicz et al.
2005, Roe et al. 2007). The most common and constant
measure of effectiveness was the frequency of incontinence.
The body of evidence on this was greatest for bladder
training, and more limited for prompted voiding, habit
retraining and timed voiding. There were insufficient studies
undertaken and therefore included in the latter two system-
atic reviews to make definitive recommendations for clinical
practice. Limited data on other health outcomes, change in
dependency status, QOL and cost effectiveness were available
and need to feature in future research. The theory underpin-
ning these interventions, future research design, long-term
follow up and outcomes measured need careful consideration
in future research and need to adhere standards for good
practice so that evidence on effectiveness of interventions is
obtained and future meta-analyses can be performed. Meta-
study techniques provide a useful methodology for the
synopsis of systematic reviews of quantitative evidence.
Acknowledgement
We would like to thank Margarete Sandelowski and Sally
Thorne for their support and encouragement in pursuing this
approach and application of metastudy techniques.
Author contributions
BR was responsible for the study conception and design. BR,
JO, JM and SW were responsible for drafting of the
manuscript. BR, JO, JM and SW performed the data
collection and data analysis. BR made critical revisions to
the paper.
References
Note the references in bold adhere to the Cochrane Library
convention, where each trial is cited by first author/principal
investigator only and year of initial publication, with all
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Systematic reviews of bladder training and voiding programmes in adults
subsequent papers and publications arising from the original
trial included under that author and date.
Castleden C.M. & Duffin H.M. (1981) Guidelines for controlling
urinary incontinence without drugs or catheters. Age and Ageing
10, 186190.
Clay E.C. (1978) Incontinence of urine: a regimen for retraining.
Nursing Mirror 146, 2324.
Closs S.J. & Cheater F.M. (1999) Evidence for nursing practice:
A clarification of the issues. Journal of Advanced Nursing 30,
1017.
Colling J., Ouslander J., Hadley B.J., Eisch J. & Campbell E. (1992)
The effects of patterned urge-response toileting (PURT) on urinary
incontinence among nursing home residents. Journal of the
American Geriatrics Society 40(2), 135141.
Colling J., Owen T.R., McCreedy M. & Newman D. (2003) The
effects of a continence program on frail community dwelling
elderly persons. Urology Nursing 23(2), 117122.
Colombo M., Zanetta G., Scalambrino S. & Milani R. (1995)
Oxybutinin and bladder training in the management of female
urinary urge incontinence. A randomised study. International
Urogynecology Journal and Pelvic Floor Dysfunction 6, 6367.
Egger M., Davey, Smith G. & Altman D.G. (2005) Systematic Re-
views in Health care. Meta-analysis in Context. BMJ Publishing,
London.
Eustice S., Roe B. & Paterson J. (2002) Prompted Voiding for
the Management of Urinary Incontinence in Adults (Cochrane
Review). In The Cochrane Library. Issue 2, John Wiley and Sons,
Ltd Chichester.
Fantl 1991
Fantl A., Wyman J.F., McCLish D.K., Harkins S.W., Elswick R.K.,
Taylor J.R. et al. (1991) Efficacy of bladder training in older
women with urinary incontinence. Journal of the American
Medical Association 265(5), 609613.
Fantl J.A., Wyman J.F., Harkins S.W. & Taylor J.R. (1988) Bladder
training in women with urinary incontinence (Abstract). Neu-
rourology and Urodynamics 7(3), 276277.
McLish D.K., Fantl J.A., Wyman J.F., Pisani G. & Bump R.C. (1991)
Bladder training in older women with urinary incontinence:
relationship between outcome and changes in urodynamic
observations. Obstetrics and Gynecology 77(2), 281286.
Wyman J.F., Fantl J.A., McClish D.K., Harkins S.W., Uebersax J.S.
& Ory M.G. (1997) Quality of life following bladder training
in older women with urinary incontinence. International
Urogynecology Journal 8(40), 223229.
Wyman J.F., McClish D.K., Ory M.G. & Fantl J.A. (1992) Changes
in quality of life following bladder training in older women with
urinary incontinence (Abstract). Neurourology and Urodynamics
11(40), 426427.
Fonda D., DuBeau C.E., Harari D., Ouslander J.G., Palmer M. &
Roe B. (2005) Incontinence in the Frail Elderly. In Incontinence:
Management Volume 2. 3rd International Consultation on
Incontinence. Chapter 18 (Abrams P., Cardozo L., Khoury S.,
Wein A., eds), Health Publications Ltd, Plymouth, pp. 1165
1239.
29
B. Roe et al.
Frewen W.K. (1978) An objective assessment of unstable bladder of
psychosomatic origin. British Journal of Urology 50(4), 246249.
Hadley E.C. (1986) Bladder training and related therapies for urinary
incontinence in older people. Journal of the American Medical
Association 256(3), 372379.
Hu 1989
Hu T.W., Igou J.F., Kaltreider D.C., Yu L.C., Rohner T.J.,
Dennis P.J., Craighead W.E., Hadley E.C. & Ory M.G. (1989)
A clinical trial to reduce urinary incontinence in nursing
homes. Journal of the American Medical Association 261(18),
26562662.
Hu T.W., Kaltreider D.L., Igou J.F., Yu L.C. & Rohner T.J. (1990)
Cost effectiveness of training incontinent elderly in nursing homes:
a randomised clinical trial. Health Services Research 25(3), 455
477.
Kennedy A.P. (1992) Bladder re-education for the promotion of
continence. In Clinical Nursing Practice: The Promotion and
Management of Continence (Roe B., ed.), Prentice Hall, London,
pp. 7894.
Jarvis 1981
Jarvis G.J. (1981) A controlled trial of bladder drill and drug therapy
in the management of detrusor instability. British Journal of
Urology 53(6), 565566.
Jarvis G.J. (1981) The Unstable Bladder A Psychosomatic Disease?
(Abstract). Conference Proceedings of the International Continence
Society (ICS), 11th Annual Meeting, pp. 4546.
Jirovec M.M. & Templin T. (2001) Predicting success using in-
dividualized scheduled toileting for memory impaired elders at
home. Research in Nursing & Health 24(1), 18.
Lagro-Janssen 1992
Lagro-Janssen A.L., Debruyne F.M., Smits A.J. & Van Weel C.
(1992) The effects of treatment of urinary incontinence in general
practice. Family Practice 9(30), 284289.
Lagro-Janssen T. & Van Weel C. (1998) Long term effect of treat-
ment of female incontinence in general practice. British Journal of
General Practice 48(436), 17351738.
Lagro-Jansen T.L.M., Debruyne F.M.J., Smits A.J.A. & Van Weel C.
(1991) Controlled trail of pelvic floor exercises in the treatment of
urinary stress incontinence in general practice. British Journal of
General Practice 41(352), 445449.
MRC (1998) Guidelines for Good Clinical Practice in Clinical Trials.
Medical Research Council, London, UK. http://www.mrc.ac.uk/
pdf-ctg.pdf last accessed 13/10/2004.
Ouslander J.G., Schnelle J.F., Uman G., Fingold S., Nigam J.G.,
Tuico E. & Jensen B.B. (1995) Does oxybutinin add to the effec-
tiveness of prompted voiding for urinary incontinence among
nursing home residents? A placebo-controlled trial. Journal of the
American Geriatrics Society 43, 610617.
Ostaszkiewicz J., Johnson L. & Roe B. (2004a) Timed Voiding for
Urinary Incontinence in Adults (Cochrane Review). In The
30  2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Cochrane Library. Issue 1, Update Software, John Wiley and Sons
Ltd, Chichester.
Ostaszkiewicz J., Johnson L. & Roe B. (2004b) Habit Training for
Urinary Incontinence in Adults (Cochrane Review). In The
Cochrane Library. Issue 2, Update Software. John Wiley and Sons
Ltd, Chichester.
Ostaszkiewicz J., Roe B. & Johnson L. (2005) Effects of
timed voiding for the management of urinary incontinence in
adults: systematic review. Journal of Advanced nursing 52(4), 420
431.
Paterson B.L., Thorne S.E., Canam C. & Jillings C. (2001) Meta-
study of Qualitative Health Research. A Practical Guide to Meta-
Analysis and Meta-Synthesis. Sage Publications, Thousand Oaks,
CA.
Roe B. & Moore K.N. (2000) Utilization of Incontinence Clinical
practice Guidelines. Journal of Wound, Ostomy and Continence
Nursing 28(6), 287304.
Roe B., Milne J., Ostaszkiewicz J. & Wallace S. (2007) Systematic
review of bladder training and voiding programmes for the
management of urinary incontinence in adults: A synopsis of
findings on theory and methods using metastudy techniques.
Journal of Advanced Nursing in press.
Schnelle 1983
Schnelle J.F., Traugher B., Morgan D.B., Embry J.E., Binion A.F. &
Coleman A. (1983) Management of geriatric incontinence in nur-
sing homes. Journal of Applied Behavior Analysis 16(2), 235241.
Schnelle 1989
Schnelle J. (1990) Treatment of urinary incontinence in nursing home
patients by prompted voiding. Journal of American Geriatrics
Society 38(3), 356360.
Schnelle J.F., Traugher B., Sowell V.A., Newman D.R., Petrilli C.O.
& Ory M. (1989) Prompted voiding treatment of urinary incon-
tinence in nursing home patients. A behavior management
approach for nursing home staff. Journal of the American
Geriatrics Society 37(11), 10511057.
Smith D.A., Newman D.K., McDowell B.J. & Burgio L.D. (1992)
Reduction of incontinence among elderly in a nursing home set-
ting. In Key Aspects of Elder care: Managing Falls, Incontinence
and Cognitive Impairment (Funk S.G., Tornquist E.M., Cham-
pagne M.T. & Wiese R.A., eds), Springer Publishing, New York,
pp. 196204.
Surdy T.M. (1992) Rehabilitation of Urinary Incontinent Nursing
Home Patients. Doctoral Thesis, University of Wisconsin-Mil-
waukee, Milwaukee, WI.
Tobin G.W. & Brocklehurst J.C. (1986) The management of urinary
incontinence in local authority residential homes for the elderly.
Age and Ageing 15(50), 292298.
The Joanna Briggs Foundation (2005) http://www.joannabriggs.
edu.au/about/home.php# (Accessed 28/9/2005).
Thorne S., Jensen L., Kearney M.H., Noblit G. & Sandelowski M.
(2004) Qualitative metasynthesis: reflections on methodological
orientation and ideological agenda. Qualitative Health Research
14(10), 13421365.
JAN: REVIEW PAPER
Wallace S., Roe B., Williams K. & Palmer M. (2004) Bladder
Training for Urinary Incontinence in Adults (Cochrane Review). In
The Cochrane Library. Issue 1, Update Software, John Wiley and
Sons Ltd.
Whittemore R. & Knafl K. (2005) The integrative review. Updated
methodology. Journal of Advanced Nursing 52, 18.
Willington F.L. (1975) Training and retraining for continence. Nur-
sing Times 71, 500503.
Wyman 1998
Barber M.D., Visco A.G., Wyman J.F., Fanlt R.C., Bump R.C. for the
Continence Program for Women Research Group (2002) Sexual
function in women with urinary incontinence and pelvic organ
prolapse. Obstetrics and Gynecology 99(2), 281289.
Elser D.M., Fantl J.A., McClish D.Y. for the Continence Program for
Women Research group (1995) Comparison of subjective and
objective measures of severity of urinary incontinence in women.
Neurourology and Urodynamics 14(4), 311316.
 2006 The Authors. Journal compilation  2006 Blackwell Publishing Ltd
Systematic reviews of bladder training and voiding programmes in adults
Elser D.M., Wyman J.F., McClish D.K., Robinson D., Fantl J.A. &
Bump R.C. (1999) The effect of bladder training, pelvic floor
muscle training or combination training on urodynamic para-
meters in women with urinary incontinence. Neurourology and
Urodynamics 18(50), 427436.
Theofrastous J.P., Wyamn J.F., Bump R.C., McClish D.K., Elser
D.M., Bland D.R. et al. (2002) Effects of pelvic floor muscle
training on strength and predictors of response in the treatment of
urinary incontinence. Neurourology and Urodynamics 21(5), 486
490.
Wyman J.F., Fantl J.A., McClish D.K., Bump R.C. & the Continence
Program for Women Research Group (1998) Comparative efficacy
of the behavioural interventions in the management of female
urinary incontinence. American Journal of Obstetrics and
Gynecology 179(4), 9991007.
Wyman J.F., McClish D.K., Sale B. & Camp J. (1999) Long term
follow-up of behavioural interventions in incontinent women
(Abstract). International Urogynecology Journal & Pelvic Floor
Dysfunction 10(Suppl. 1), S33.
31