Objective: To evaluate the effect of a low-intensity behavioral therapy program on urinary incontinence in older women.

Methods: A randomized clinical trial for community-dwelling women at least 55 years reporting at least one urinary incontinent episode per week was conducted. Women were randomly assigned to a behavioral therapy group (n = 77) or a control group (n = 75). The treatment group had six weekly instructional sessions on bladder training and followed individualized voiding schedules. The control group received no instruction but kept urinary diaries for 6 weeks. After this period, the control group underwent the behavioral therapy protocol. Using per-protocol analyses, t and ² tests were used to compare the treatment and control groups, and paired t tests were used to evaluate the efficacy of behavioral therapy for all women (treatment and control groups before and after behavioral therapy).

Results: Women in the treatment group experienced a 50% reduction in mean number of incontinent episodes recorded on a 7-day urinary diary compared with a 15% reduction for controls (P = .001). After behavioral therapy, all women had a 40% decrease in mean weekly incontinent episodes (P = .001), which was maintained over 6 months (P < .004). Thirty (31%) women were 100% improved (dry), 40 (41%) were at least 75% improved, and 50 (52%) at least 50% improved. There were no differences in treatment efficacy by type of incontinence (stress, urge, mixed) or group assignment (treatment, control).

Conclusion: A low-intensity behavioral therapy intervention for urinary incontinence was effective and should be considered as a first-line treatment for urinary incontinence in older women.

^aDepartment of Obstetrics, Gynecology, and Reproductive Sciences, University of California, San Francisco, San Francisco, California, USA
^bDepartment of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, California, USA
^cKaiser Foundation Research Institute, Division of Research, Oakland, California, USA
^dKaiser Permanente Medical Center, Department of Urology, Oakland, California, USA

This work was supported by Direct Community Benefit Investment, Kaiser Foundation Research Institute.

(Obstet Gynecol 2002:100:72-78. © 2002 by The American College of Obstetricians and Gynecologists.)

Urinary incontinence is one of the most prevalent health conditions in women, affecting an estimated 30-50% of community-dwelling older women, with 15% experiencing daily incontinent episodes.^1-4 In addition to substantial medical, quality of life, social, and psychologic consequences, urinary incontinence is also associated with a large economic burden estimated at over $31 billion (1995 dollars) annually in the United States.^5-7

Many treatments are commonly used for urinary incontinence, including behavioral, pharmacologic, and surgical therapies. Behavioral interventions are often recommended as the initial therapy for stress and urge incontinence.⁷ Behavioral therapy may include any combination of assisted toileting, bladder training, and pelvic muscle rehabilitation, including pelvic muscle exercises, biofeedback therapy, vaginal weight training, and pelvic floor electrical stimulation.⁸ Although there are no definitive data on the treatment mechanisms of behavioral therapy, these modalities are hypothesized to improve bladder control by teaching patients how to control the physiologic responses of the bladder and pelvic muscles that mediate continence.^9-11 In randomized controlled trials, behavioral therapy programs have been reported to reduce the number of stress and/or urge incontinent episodes by 50-80%.^8,9,12,13 One-fourth to one-half of women achieve near continence with behavioral therapy, which is similar to the efficacy of pharmacologic therapy.^8,9,14,15

Most of the prior randomized controlled trials evaluating behavioral therapy have used aggressive, time-intensive regimens, including biofeedback, electrical stimulation, or vaginal cones.^9,16-18 One trial that evaluated a lower-intensity behavioral program including bladder training, urge suppression, educational sessions, and voiding diaries observed a significant decrease in the number of weekly incontinent episodes.⁸ Most prior studies have used urodynamic testing to prove the diagnosis of urinary incontinence for study entry.^8,9,16,18 Because incontinence is often diagnosed by primary care physicians who could immediately begin a low-intensity behavioral therapy program before diagnostic testing, the effectiveness of low-intensity behavioral therapy programs for women with reported incontinence symptoms is important. We conducted this randomized controlled clinical trial to evaluate the efficacy of bladder training compared with controls to treat older women with urge, stress, or mixed urinary incontinence defined by women's description of bladder symptoms.

Materials and Methods

Women 55 years and older reporting urinary incontinence were recruited at a northern California health maintenance organization. This study protocol was approved by the institutional review board. Physicians and nurse practitioners were asked to refer ambulatory female patients who reported at least one urinary incontinent episode per week over the past 6 months. A consecutive cohort of participants were recruited in 1995 until enrollment of the estimated sample size and 25% for dropout was completed. Participants lived independently in the community and were functionally capable of independent toileting. Women with uncontrolled diabetes mellitus, urinary tract infection (dipstick urinalysis positive for leukocytes, nitrites, or blood followed by a urine culture with more than 1000 colony-forming units per high power field), or history suggestive of urinary obstruction, overflow, or functional incontinence, or urinary tract anomalies were excluded from this study. Current pharmacologic therapy for incontinence was not an exclusion criterion, but women were asked to remain on the same drug and dose throughout the study.

All referred women meeting initial criteria completed a 1-week, standardized urinary diary recording diurnal and nocturnal voluntary micturition and incontinent episodes. Women were evaluated by the senior investigator (KS) to verify study eligibility and collect demographic and medical history data. No pelvic examination or measure of postvoid residual urine volume was performed. Type of incontinence was classified as stress, urge, or mixed based on a participant's description of her incontinence symptoms and standard diagnostic criteria.⁷

Women were then randomly assigned to the behavioral therapy group (n = 77) or control group (n = 75) using a random number table allocation enclosed in sealed envelopes (opened by KS). Women in the behavioral therapy group had six weekly 20-minute group instructional sessions with three to five participants on bladder training by one of three nurse educators. The initial session (45 minutes) included an educational program on the structure and function of the urinary tract, normal voiding, and incontinence symptoms and causes.¹⁹ Bladder training included participant education and development of individualized voluntary voiding schedules based on each participant's baseline daytime voiding frequency recorded on her voiding diary.⁸ Participants received verbal and written instructions on pelvic muscle exercises (Kegel exercises, Kaiser Patient Handout, "Kegel Exercise," Oakland, CA, Kaiser Permanente Health Education Department, January 1995). No fluid modifications were used. In sessions 2 through 6 (20-30 minutes), participants reviewed their prior week's voiding diary, discussed diary questions, and set new voiding schedule goals for the coming week. At each visit, voiding intervals were increased by 30 minutes as tolerated with the goal of a 2.5-3 hour interval. Participants maintained a daily urinary diary during the 6 weeks of behavioral therapy. A 7-day urinary diary was repeated 6 months after completion of behavioral therapy. Participants were encouraged to maintain a voiding schedule that best suited their lifestyles after completion of the study.

Women in the control group received no instruction but kept urinary diaries for 6 weeks. After this period, they underwent behavioral therapy for 6 weeks with the same protocol as the treatment group.

The primary outcome measure was the number of incontinent episodes per week recorded on a 7-day urinary diary. Secondary outcome measures included number of diurnal and nocturnal voids per week recorded on the diary. Women were asked at 6 months to rank how the behavioral therapy program had "helped them in dealing with (their) urine leakage problem," rated as not at all, slightly, moderately, or a great deal.

A sample size of 60 women in each group provided 90% power (_2-tailed = 0.05) to detect an effect size of 50% (reduction in incontinent episodes of approximately 70% in the treatment group compared with 20% in the control group), the effect size observed by previous randomized controlled trials of behavioral therapy.^8,9 t and ² tests were used to compare the treatment and control groups at baseline and after 6 weeks of either therapy (treatment group) or observation (control group). Per-protocol analyses including only participants completing the first 6 weeks of the trial (therapy [treatment group] or observation [control group]) were performed. Paired t tests were used to compare measurements at baseline, completion of therapy (6 weeks), and 6 months after completing behavioral therapy for all participants. Repeated measures analyses of variance were done to test the effects of group allocation, time, and type (stress, urge, mixed) and severity (1-7, 8-14, 15-21, more than 21 incontinent episodes per week) of incontinence on improvement in incontinence frequency. Although participants and clinicians were not blinded to group allocation, statistical analysts were blinded.

Results

We enrolled 152 women in the study, with 77 randomized to the treatment group and 75 to the control group. Eleven women in the treatment group and 18 in the control group were lost to follow-up or dropped out of the study before completing 6 weeks of data collection and were not included in the final analyses. There were no differences in demographic characteristics, medical, gynecologic, or urologic history, current urologic symptoms, or incontinence severity at baseline between women completing and not completing the behavioral therapy versus the control period of the study (6 weeks). The treatment (n = 66) and control (n = 57) groups were similar in demographic characteristics, medical, gynecologic, and urologic history, and current urologic symptoms at study entry (Table 1).

At 6 weeks, the treatment group had a 50% reduction in mean number of incontinent episodes compared with a 15% reduction for the control group (P = .001) because of improved diurnal incontinence frequency (P = .02, Table 2). There were no differences in diurnal, nocturnal, or total micturition frequencies between the treatment and control groups. After behavioral therapy, women in the treatment group experienced improved diurnal and total incontinence frequencies (P = .001, paired t test versus baseline) but no change in nocturnal incontinence or micturition frequencies (Table 2). Improvement in incontinence was not observed in the control group (P = .37). There were no adverse events in either group.

After completing behavioral therapy, women in the control group had a decrease in weekly incontinent episodes (P = .03, paired t test versus week 6). They experienced no change in diurnal, nocturnal, or total micturition frequencies.

To evaluate the efficacy of behavioral therapy, pre- and post-therapy results for the treatment and control groups were combined (n = 122, Table 3). Women experienced a 40% reduction in mean weekly incontinent episodes after behavioral therapy (P = .001), which was maintained over 6 months (P = .004). This improvement was associated with a significant decrease in diurnal and nocturnal incontinence frequencies after bladder training (6 weeks) and only diurnal frequency at 6 months. After behavioral therapy, 30 (31%) of women were 100% improved (dry), 40 (41%) were at least 75% improved, 50 (52%) at least 50% improved, 63 (65%) at least 25% improved, and 27 (28%) had no improvement or a worsening of incontinence frequency. Improvement was also observed in diurnal micturition frequency at 6 weeks and 6 months and total micturition frequency at 6 months after therapy (Table 3).

We observed an effect of time on participation in behavioral therapy and improvement in incontinence frequency (Figure 1, P < .002). Women had continued improvement in incontinence each week, reaching a statistically significant decrease in weekly incontinent episodes after 4 weeks of therapy (P = .001 versus baseline). No differences were observed, however, between women undergoing therapy immediately (treatment group) compared with after an observation period (control group) or by type or severity of incontinence. There was no effect modification observed for type or severity of incontinence.

Figure 1. Temporal effect of behavioral therapy during 6 weeks of treatment. Average number of incontinent episodes per week and 95% confidence intervals are presented by week of therapy. Participants experienced continued improvement in incontinence each week with a statistically significant decrease in weekly incontinent episodes after 4 weeks of therapy (P = .001 versus baseline). Subak. Behavioral Therapy for Incontinence. Obstet Gynecol 2002.

Six months after behavioral therapy, 33% of participants reported that the program had helped them a great deal with their urine leakage problem, 26% reported the program to be moderately helpful, 29% slightly helpful, and 12% not at all helpful.

Discussion

Women experienced a significant improvement in incontinence frequency after a low-intensity behavioral therapy program using bladder training. This effect was observed in the therapy group (treatment) in the randomized controlled trial (P = .001) and in all women after therapy (combined treatment and control groups for pre- and post-therapy results, P = .001). Improvement was maintained over 6 months (P = .004). This low-intensity behavioral therapy program resulted in statistically significant as well as clinically significant improvement in incontinence: half of participants experienced at least a 50% decrease in incontinence frequency, and 31% were dry at the end of therapy. A majority of participants (59%) reported that the behavioral therapy program moderately or greatly helped them with their incontinence problem. In addition, the significant treatment effect was achieved after only 4 weeks of behavioral therapy.

Behavioral therapy is a low-risk, low-cost intervention, proven to provide significant improvement in incontinence symptoms. Our trial results support the Agency for Health Care Policy Research guidelines for managing incontinence in adults recommending that "the first choice [of treatment] should be the least invasive treatment with the fewest potential adverse complications that is appropriate for the patient."⁷ Bladder training is an excellent initial step in the management of urinary incontinence in older women, which can be done before more invasive and expensive diagnostic testing and therapeutic modalities.⁷ It is also well suited for older women in whom complications and side effects of other therapies are substantial. Ideally, behavioral therapy can be initiated by any provider in a "see-and-treat" fashion. This is particularly important in a primary care setting. Women who discuss incontinence with their providers can immediately begin a behavioral program that requires minimal additional physical examination, diagnostic testing, staff time, or training.

Other research studies have examined the effect of different types of behavioral therapy on incontinence. Berghmans et al¹⁷ presented a systematic review of randomized controlled trials for stress incontinence, finding strong evidence to support the effectiveness of pelvic floor muscle exercises alone to treat stress incontinence. Other investigators have evaluated urge and mixed incontinence, finding that 10-30% of participants experience complete resolution of incontinence and 40-80% had a 50% or greater reduction in incontinence frequency.^8-10,12 Most have studied intensive interventions that are much more time consuming than our intervention. There is limited evidence that high-intensity behavioral therapy regimens including biofeedback, electrical stimulation, vaginal cones, in-patient pelvic floor muscle exercise training, and/or medication are more effective than a low-intensity regimen for stress incontinence.¹⁷

Other studies have corroborated our observation that the effect of behavioral therapy may be similar for urge, stress, and mixed incontinence. Fantl et al⁸ evaluated bladder training with timed voiding versus no intervention (control group) in a randomized controlled trial of 131 women with urodynamically diagnosed stress, urge, or mixed urinary incontinence. They observed a significant reduction in incontinence frequency in the bladder training group (P < .001). The efficacy of bladder training was similar for women with the diagnosis of genuine stress incontinence, detrusor instability, or both.⁹ Combining several types of behavioral therapy may be more effective than a single modality. A randomized controlled trial evaluated an intensive 12-week program of bladder training, biofeedback-assisted pelvic muscle exercises, or both.¹² The combination therapy group experienced a significantly better outcome immediately after therapy (50-70% had a 50% or greater improvement in incontinence frequency) with similar improvement between groups at 3 months (40-60% had a 50% or greater improvement).

Our behavioral intervention was based on bladder training. In the randomized controlled trial, we did not observe a change in toileting behavior, suggesting that bladder training alone was not the mechanism of the treatment effect. However, we did observe a decrease in micturition frequency in the efficacy analysis, with participants recording a significant decrease in diurnal voiding frequency after therapy, which was maintained at 6 months. The only other randomized controlled trial with a similar intervention to our study found a significant reduction in diurnal micturition frequency after 6 weeks of therapy.⁸ However, their study protocol used an initial voiding schedule of 30-60 minutes rather than our protocol, which was based on each participant's baseline voiding pattern (mean initial voiding schedule was 2.2 hours). In addition, they observed a decrease in micturition frequency only in women with higher baseline weekly diurnal (more than 56 voids per week) and nocturnal (more than 4.4 voids per week) frequency. Because voiding frequency is at least in part a behavioral pattern, it may take longer than 6 weeks to change voiding behavior. During this same brief period, women may experience improved continence through better understanding of the physiologic responses of the bladder and pelvic muscles that mediate continence.

Participation in this study was subject to selection bias. Participants reported their incontinence and were willing to undergo a behavioral therapy program in a research study. However, the effect of self-selection may be minimal because the distribution of urinary incontinence type and severity in this study is close to estimates from population-based studies of older women. A study at one location inherently raises the question of generalizability. We attempted to recruit an age- and race-diverse population with varying severity of disease. In addition, the control group underwent some intervention by recording urinary diaries. We did not observe, however, a significant effect on incontinence severity during the control period. The type of incontinence was determined by the participants' description of their incontinent episodes. Urodynamic studies, pelvic examination, and measurement of postvoid residual urine volume were not performed for this study. Although this is a study limitation, it is consistent with a primary medical provider's initial evaluation and the proposed early initiation of behavioral therapy before additional evaluation or treatment.⁷

Because behavioral therapy has large potential benefit, minimal risk, and is effective for both stress and urge incontinence, it is an ideal first-line therapy that can be initiated before more invasive and costly diagnostic tests and therapeutic modalities. It is well suited for older women in whom the risks of surgical complications and/or medication side effects may be substantial.⁸ We observed improvement in incontinence after a low-intensity behavioral therapy intervention using bladder training in older women with urinary incontinence. The "best" intervention is one that is not only low risk, inexpensive, and effective, but can be initiated effectively and easily by primary care providers.

References

1. Diokno A, Brock B, Brown M, Herzog A. Prevalence of urinary incontinence and other urological symptoms in the noninstitutionalized elderly. J Urol 1986;136:1022-5.

2. Hunskaar S, Vinsnes A. The quality of life in women with urinary incontinence as measured by the sickness impact profile. J Am Geriatr Soc 1991;39:378-82.

3. Thom D. Variation in estimates of urinary incontinence prevalence in the community: Effects of differences in definition, population characteristics, and study type. J Am Geriatr Soc 1998;46:473-80.

4. Brown J, Seeley D, Fong J, Black D, Ensrud K, Grady D. Urinary incontinence in older women: Who is at risk? Obstet Gynecol 1996;87:715-21.

5. Wagner TH, Hu TW. Economic costs of urinary incontinence in 1995. Urology 1998;51:355-61.

6. Wilson L, Park GE, Luc KO, Brown JS, Subak LL. Annual costs of urinary incontinence. Obstet Gynecol 2001;98:398-406.

7. Fantl J, Newman D, Colling J, DeLancey J, Keeys C, McDowell B. Urinary incontinence in adults: Acute and chronic management. Rockville, MD: AHCPR, 1996.

8. Fantl JA, Wyman JF, McClish DK, Harkins SW, Elswick RK, Taylor JR, et al. Efficacy of bladder training in older women with urinary incontinence. JAMA 1991;265:609-13.

9.Burgio K, Locher J, Goode P, Hardin JM, McDowell BJ, Dombrowski M, et al. Behavioral vs drug treatment for urge urinary incontinence in older women: A randomized controlled trial (see comments). JAMA 1998;280:1995-2000.

10. Fantl JA. Behavioral intervention for community-dwelling individuals with urinary incontinence. Urology 1998;51(Suppl. 2A):30-4.

11. McClish DK, Fantl JA, Wyman JF, Pisani G, Bump RC. Bladder training in older women with urinary incontinence: Relationship between outcome and changes in urodynamic observations. Obstet Gynecol 1991;77:281-6.

12. Wyman JF. Quality of life of older adults with urinary incontinence (editorial, comment). J Am Geriatr Soc 1998;46:778-9.

13. Lagro-Janssen TL, Debruyne FM, Smits AJ, van Weel C. Controlled trial of pelvic floor exercises in the treatment of urinary stress incontinence in general practice. Br J Gen Pract 1991;41:445-9.

14. Appell RA. Clinical efficacy and safety of tolterodine in the treatment of overactive bladder: A pooled analysis. Urology 1997;50(Suppl. 6A):90-6, discussion 97-9.

15. Anderson R, Mobley D, Blank B, Saltzstein D, Susset J, Brown J. Once daily controlled versus immediate release oxybutynin chloride for urge urinary incontinence. OROS Oxybutynin Study Group. J Urol 1999;161:1809-12.

16. Wyman JF, Fantl JA, McClish DK, Bump RC. Comparative efficacy of behavioral interventions in the management of female urinary incontinence. Continence Program for Women Research Group. Am J Obstet Gynecol 1998;179:999-1007.

17. Berghmans LC, Hendriks HJ, Bo K, Hay-Smith EJ, de Bie RA, van Waalwijk van Doorn ES. Conservative treatment of stress urinary incontinence in women: A systematic review of randomized clinical trials. Br J Urol 1998;82:181-91.

18. Bo K, Talseth T, Holme I. Single blind, randomised controlled trial of pelvic floor exercises, electrical stimulation, vaginal cones, and no treatment in management of genuine stress incontinence in women. BMJ (Clin Res Ed) 1999;318:487-93.

19. Sandler M, Daley MC, Ebner GD, Gartley CB, Jeter KF, Watters HC. Incontinence urinary leakage, a common and treatable condition. San Bruno, CA: Krames Communications, 1995.

Address reprint requests to: Leslee L. Subak, MD, University of California, San Francisco, Department of Obstetrics, Gynecology, and Reproductive Sciences, 1635 Divisadero Street, Suite 600, San Francisco, CA 94115; E-mail: subakl@obgyn.ucsf.edu

Received August 31, 2001.
Received in revised form January 18, 2002.
Accepted February 7, 2002.