| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
a Center for Aging and the Division of Geriatric Medicine and Gerontology, University of Alabama at Birmingham
b Department of Psychology, University of Alabama at Birmingham
Decision Editor: John E. Morley, MB, BCh
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Participants were 214 community-dwelling women, aged 40 to 90 years, with a history of stress, urge, or mixed urinary incontinence, persisting at least 3 months with a frequency of two or more episodes of urine leakage per week. Each participant kept a 14-day bladder diary documenting the time and circumstances of each incontinence episode.
Results. The mean age of participants was 63.5 years; 16.9% were African American. Women with predominantly urge incontinence (n = 138) reported a daily frequency of 2.1 incontinent episodes. Although there was a statistically significant difference between Week 1 (2.4 episodes per day) and Week 2 (2.0 episodes per day; p < .0001), five days were necessary to obtain an internal consistency of .90 for Cronbach's alpha. Women with predominantly stress incontinence (n = 76) had no statistical difference between Week 1 and Week 2 in frequency of incontinence, reporting an average 2.2 accidents per week in Week 1 and 2.1 in Week 2. However, 7 days of bladder diary were required before adequate internal consistency was reached.
Conclusion. Seven consecutive days of bladder diary provides a stable and reliable measurement of the frequency of incontinence episodes in community-dwelling women.
THE assessment of health-related symptoms and behaviors relies heavily upon self-reported data. Self-reported information is usually obtained through a clinical interview and may also include documentation in health diaries. The reliability and validity of self-reported data is important and depends on several factors, including the method of self-report, the type of information being sought, and a variety of context effects, which are the social and cognitive factors related to the processing of requested information (1).
The bladder diary (also referred to as a voiding or urinary diary, frequency-volume or incontinence chart, or bladder record) is a common method used to evaluate the frequency and characteristics of incontinence episodes in both research and clinical practice. The Agency for Health Care Policy and Research Clinical Practice Guideline (1996), the First International Consultation on Incontinence (1999), and the International Continence Society (1990) have all recommended the bladder diary to assess patients with urinary incontinence.
The bladder diary has also been used routinely as a primary outcome measure in incontinence research including trials funded by the National Institutes of Health (2)(3)(4). The bladder diary has been described as the best method to assess incontinence frequency in ambulatory patients. Although electronic monitoring devices have been developed to monitor incontinence in community-dwelling patients, their use is prohibitive because of their cost and impracticality (5). The diary has advantages over other self-report methods of data collection: It is a prospective method that reduces recall error and results in higher levels of reporting for most conditions (6). As Wyman (5) observed, the bladder diary has the advantage of assessing incontinence "in the individual's own environment and under actual daily life conditions." A limitation of diaries, however, is that they have the potential effect of changing the monitored behavior (2)(6)(7).
Although the bladder diary is a widely accepted method of assessing incontinence frequency in ambulatory patients, very little research has been conducted on the reliability and validity of the diary. To assess reliability, previous investigators have used the retest method using correlation analysis as well as t tests. One study compared a one-week diary with a two-week diary and concluded that one week was acceptable (5). A second study compared the first three days with the last four days of a 7-day diary and found no differences (8).
The purpose of the present study was to investigate the reliability of a 14-day diary for assessing the frequency of incontinence episodes. Specifically, we sought to determine the number of consecutive days needed to obtain adequate internal consistency using Cronbach's alpha. While previous reports in the literature have recommended 3, 7, or 14 days of keeping a bladder diary, no research to date has specifically addressed the question of how many days are necessary to reach a satisfactory level of internal consistency (9)(11)(12).
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Procedures
Participants in both trials underwent a clinical evaluation consisting of a medical history and physical examination, post-void catheterization for residual urine, urodynamic evaluation, and screening for dementia using the Mini-Mental State Examination. Participants were excluded from participation in either study if they scored below 24 on the Mini-Mental State Examination.
Following the clinical evaluation, participants were provided with 14 days of bladder diary booklets in which the participants documented the time of every void, the time of every incontinence episode, including the volume of urine loss (large or small), and the circumstances surrounding each episode. Participants were instructed verbally by a research nurse on how to complete the bladder diaries. The main purpose of the diary was to document pretreatment frequency of incontinence. All participants completed 14 days of the baseline diary.
Diagnosis of Urge and Stress Incontinence
The diagnostic criteria for urge and stress incontinence were made on the basis of a clinical interview, the baseline bladder diary, and a urodynamic evaluation. The criteria were developed on the basis of the recommendations of the International Continence Society (10).
Urge incontinence..-- To be diagnosed as urge incontinent, the participant must have reported in the clinical interview that she experienced an involuntary loss of urine associated with a strong desire to void and that the condition persisted for at least three months. In the two-week baseline bladder diary, the participant must have documented at least two urge accidents per week (on average). If the participant had concomitant stress incontinence, the predominant pattern must have been urge incontinence (i.e., the number of urge accidents must have exceeded the number of stress accidents in the bladder diary). On urodynamic evaluation, there must have been cystometric evidence of bladder dysfunction. One or both of the following must have been observed: (i) detrusor instability defined as any detrusor contraction that occurs during passive filling or with provocative maneuvers (indicating motor urgency), or (ii) maximal cystometric capacity, the volume at which the participant reported a strong desire to void, was less than 350 ml (indicating sensory urgency).
Stress incontinence..-- To be diagnosed as stress incontinent, the participant must have reported in the clinical interview that she experienced an involuntary loss of urine associated with physical exertion and that the condition persisted for at least three months. On the two-week baseline bladder diary, the participant must have documented at least two stress accidents per week (on average). If the participant had concomitant urge incontinence, the predominant pattern must have been stress incontinence (i.e., the number of stress accidents must have exceeded the number of urge accidents in the bladder diary). On urodynamic evaluation, there must have been evidence of stress incontinence with provocation (sitting up from lying; standing up from sitting; and coughing while lying, sitting, and standing).
Data Management and Analyses
Analyses were performed separately for the urge and stress groups. First, an average daily frequency was calculated for each of the 14 days. Paired t tests were used to compare the mean daily frequencies of incontinent episodes for Week 1 and Week 2. Next, correlation analysis was performed to examine test–retest reliability between Week 1 and Week 2. Last, Cronbach's alpha was used to evaluate the internal consistency of various lengths of data collection from 2 days to 14 days. An internal consistency of .90 was considered sufficient for achieving reliable reports of the frequency of incontinence episodes.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Women with predominantly stress incontinence (n = 76) had a mean age of 60.0 years with a range of 40 to 73 years. Sixteen percent of participants were African-American, and 84.0% were white. Over half of the women (52.6%) in this group experienced pure stress incontinence, while 47.4% experienced mixed stress and urge incontinence. In this group, 26.3% documented mild incontinence, 26.3% documented moderate incontinence, and 47.4% documented severe incontinence.
Frequency and Pattern of Incontinent Episodes
Women with predominantly urge incontinence recorded an average daily frequency of 2.2 incontinence episodes over the 14-day bladder diary period, while women with predominantly stress incontinence recorded an average daily frequency of 2.1 incontinence episodes. Fig. 1 presents the mean daily frequency of incontinence episodes across the 14-day period.
|
Women with stress incontinence also had more incontinence episodes in the first few days of recording. However, the number of accidents appeared to stabilize at 3 days, and only Day 1 was significantly greater than the 14-day average of 2.1 (p < .05). Days 5 and 10 were significantly lower than the 14-day average (p < .05). Women with stress incontinence reported an average 2.2 accidents per day in Week 1 and 2.1 accidents per day in Week 2 (t = .94 [df = 75], p = .35).
Daily frequency of accidents in Week 1 was highly correlated with frequency of accidents in Week 2 for women with predominantly urge incontinence (r = .93, p < .0001) as well as for women with predominantly stress incontinence (r = .86, p < .0001).
Internal Consistency
Values for the Cronbach's alpha measure of internal consistency as a function of days of recording are presented in Fig. 2. Women with urge incontinence required five days of keeping a bladder diary to reach adequate internal consistency, defined as reliability estimate of .90. Women with stress incontinence required seven days of keeping a bladder diary to reach adequate internal consistency.
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The nonsignificant differences between Weeks 1 and 2 for the stress group suggest that incontinence is a condition in which symptoms are fairly stable over time. The results corroborate the findings of previous studies of the reproducibility of bladder diary data (5)(8). Robb (5) compared the first three days of a one-week bladder diary to the last four days of the diary in 44 older community-dwelling men. Using both correlation analysis and paired t tests, she found no significant differences between the three- and four-day data collection periods. Robb's work, however, was limited in that 54% of the participants experienced no incontinence during the first three days of diary recording, and the mean number of episodes was only 2, compared with the last four days when 55% of participants reported no incontinence, and the mean number of episodes was also 2.
Wyman and colleagues (5) found high test–retest reliability between Week 1 and Week 2 for the recording of incontinence episodes for two groups of patients, those with detrusor instability with or without sphincteric incompetence (urge incontinence or mixed urge and stress incontinence) and those with sphincteric incompetence alone (stress incontinence). On the basis of these findings, the authors concluded that a one-week diary was sufficient to obtain reliable reports of frequency of incontinence.
Our data on urge incontinence are inconsistent with these previous studies. In the present study, the early immediate decrease in frequency observed in the urge group does not reflect low reliability of the measure but is more appropriately attributed to the effects of self-monitoring. The positive influence of self-monitoring has been exemplified by the effects of food diaries on eating behavior and the effects of self-monitoring of glucose levels on diabetic control. For incontinence, self-monitoring by keeping a bladder diary can enhance the awareness of bladder habits and leakage patterns. The resulting insights into the problem may produce behavioral alterations that can potentially decrease the frequency of incontinence episodes. Such an effect was suggested by data from a clinical trial of 197 women with predominantly urge incontinence (2). In that study, the control group demonstrated a 40% decrease in the number of incontinence episodes. In addition to taking placebo medication, the control group kept bladder diaries for the 8 weeks of the trial, and it was believed that the self-monitoring effect could explain a portion of the improvement.
Considering these data and data from the present study, which indicate that urge-incontinent women had a significant reduction of incontinence from Week 1 to Week 2, it is possible that urge incontinence is more responsive to the effects of self-monitoring than is stress incontinence, which is more anatomically based.
Self-monitoring effects appeared to occur early, however, and appeared to decrease after three to seven days for both groups in the present study. Despite the difference in Weeks 1 and 2 for the urge-incontinent group, sufficient internal consistency (Cronbach's alpha >.90) was achieved in 5 days. For the stress group, 7 days were required. The daily frequency of incontinent episodes for the urge-incontinent group declined over a greater number of days, and the decline was more steady compared with the stress-incontinent group. The daily frequency of incontinence episodes for the stress-incontinent group dropped sharply over fewer days, and there was more variability between the days. These differences in consistency over time between the two types of incontinence explain why it took longer for the stress group to reach sufficient internal consistency levels.
Sufficiently high levels of internal consistency were evident within the first week for both urge and stress incontinence. In conclusion, data from our study indicate that seven consecutive days of bladder diaries should provide stable and reliable estimates of the frequency of incontinence in clinical samples of community-dwelling women.
|
Acknowledgments |
|---|
This paper was presented at the 1999 Gerontological Society of America's 52nd annual scientific meeting held in San Francisco.
We thank Artisha Moore for data management.
Received December 2, 1999
Accepted April 28, 2000
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. G. Ouslander Management of Overactive Bladder N. Engl. J. Med., February 19, 2004; 350(8): 786 - 799. [Full Text] [PDF]
|
|
|
|
 |
|
 M.-M. G. Wilson Guest Editorial. Urinary Incontinence: Bridging the Gender Gap J. Gerontol. A Biol. Sci. Med. Sci., August 1, 2003; 58(8): M752 - 755. [Full Text] [PDF]
|
|
|
|
 |
|
 P. S. Goode, K. L. Burgio, J. L. Locher, D. L. Roth, M. G. Umlauf, H. E. Richter, R. E. Varner, and L. K. Lloyd Effect of Behavioral Training With or Without Pelvic Floor Electrical Stimulation on Stress Incontinence in Women: A Randomized Controlled Trial JAMA, July 16, 2003; 290(3): 345 - 352. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Locher, K. L. Burgio, P. S. Goode, D. L. Roth, and E. Rodriguez Effects of Age and Causal Attribution to Aging on Health-Related Behaviors Associated With Urinary Incontinence in Older Women Gerontologist, August 1, 2002; 42(4): 515 - 521. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
