Burgio 1998: Behavioral vs Drug Treatment for Urge Urinary Incontinence in Older Women
Burgio 1998 was a randomized, placebo-controlled, three-arm trial of 8 weeks of biofeedback-assisted behavioral training versus individually titrated oxybutynin versus placebo capsules in 197 community-dwelling women aged 55 to 92 years with urge or urge-predominant mixed incontinence, with the percent reduction in diary incontinent episodes as the primary outcome. Behavioral training reduced episodes by a mean 80.7%, oxybutynin by 68.5% (P = .04 versus behavioral training) and placebo by 39.4%.[1]
- Design and population
- Randomized, placebo-controlled, three-arm trial at a university geriatrics clinic. Community-dwelling volunteer women aged 55 to 92 years with urge or urge-predominant mixed incontinence, urodynamic bladder dysfunction and no dementia, screened at a university geriatric clinic.
- Arms
- Biofeedback-assisted behavioral training (4 visits over 8 weeks) versus individually titrated oxybutynin chloride (2.5 to 15 mg daily) versus identical-appearing placebo capsules with the same clinic visits
- Report and follow-up
- 1998 · 8-week treatment period; outcome from the 2-week posttreatment bladder diary
| Endpoint | Results | Effect | Finding |
|---|---|---|---|
| Mean percent reduction in weekly incontinent episodes on bladder diary after 8 weeks (behavioral minus oxybutynin shown as the effect; analyzed n = 63, 65, 62) | 80.7% vs 68.5% vs 39.4% | difference 12.2; P 0.04 | Met |
| More incontinent episodes after treatment than before | 1/63 (1.6%) vs 3/65 (4.6%) vs 10/62 (16.1%) | — | Descriptive |
Harms: Dry mouth: 34.9% vs 96.9% vs 54.8%; Inability to void: 6.3% vs 21.5% vs 3.2%; Constipation: 22.2% vs 38.5% vs 37.1%
Registry not recorded · Report DOI
Design
Women were recruited through advertisements and professional referrals and screened by telephone, then evaluated with history, examination, postvoid residual catheterization, two-channel supine water cystometry, laboratory tests and the Mini-Mental State Examination. Entry required urge incontinence at least twice a week for 3 months or more, urge accidents exceeding stress accidents on a 2-week diary, and urodynamic bladder dysfunction (detrusor instability, or a maximal cystometric capacity of 350 mL or less). Exclusions included a postvoid residual above 200 mL, uterine prolapse past the introitus, narrow-angle glaucoma, unstable angina, decompensated heart failure, malignant arrhythmia and a Mini-Mental State score below 20. Of 468 women evaluated, 271 were ineligible or did not participate, and 197 were randomized between July 1989 and August 1995.[1]
Randomization used computer-generated numbers in blocks of 6 within strata of incontinence type (urge only or mixed) and weekly severity (fewer than 5, 5 to 10, or more than 10 episodes). Mean age was 67.7 years, 51.3% had mixed incontinence and 52.8% had more than 10 accidents a week. The groups were comparable at baseline except that the behavioral group had more children, fewer high school graduates and more rectoceles.[1]
The primary outcome was the percent reduction in incontinent episodes between 2-week pretreatment and posttreatment bladder diaries, scored by a research assistant blinded to group. Secondary outcomes were patient-perceived progress, satisfaction and willingness to continue treatment. Drug and placebo assignment was double-blind; patients and nurse practitioners could not be blinded to behavioral versus drug treatment. Analysis was by intention to treat on 190 women (seven had no follow-up diary), using the most recent diary for women who stopped early. Mean reductions were compared with a rank-based ANOVA followed by Duncan multiple range tests at an overall alpha of .05; the report states no prespecified hypothesis or margin.[1]
Results
Baseline weekly incontinent episodes were 15.8, 15.9 and 15.4. After treatment the means were 2.8 with behavioral training, 5.7 with oxybutynin and 8.2 with placebo (P = .005 across groups). The mean percent reduction was 80.7%, 68.5% and 39.4% (P < .001 across groups). Behavioral training beat oxybutynin (P = .04) and placebo (P < .001), and oxybutynin beat placebo (P = .009).[1]
A reduction of at least 50% occurred in 57/63 women (90.5%) with behavioral training, 50/65 (76.9%) with oxybutynin and 40/62 (64.5%) with placebo (P = .002). A reduction of at least 75% occurred in 46/63 (73.0%), 37/65 (56.9%) and 20/62 (32.3%) (P < .001). Complete dryness occurred in 19/63 (30.2%), 15/65 (23.1%) and 8/62 (12.9%), a difference that did not reach significance (P = .07). Incontinent episodes were more frequent after treatment than before in 1/63, 3/65 and 10/62 women.[1]
Women rated themselves "much better" in 43/58 (74.1%) of the behavioral group, 28/55 (50.9%) of the oxybutynin group and 14/52 (26.9%) of the placebo group (P < .001). Comfort with continuing the treatment indefinitely was reported by 55/57 (96.5%), 29/53 (54.7%) and 22/51 (43.1%), and a wish to receive another form of treatment by 8/57 (14.0%), 40/53 (75.5%) and 37/49 (75.5%). Among women who were not completely dry, 14.5% of the behavioral group asked to add drug treatment and 53.3% of the oxybutynin group asked to add behavioral treatment.[1]
In the 105 women with posttreatment cystometry, cystometric capacity changed by +17.3 mL with behavioral training (P = .30), +70.9 mL with oxybutynin (P < .001) and -5.9 mL with placebo (P = .61).[1] A later urodynamic analysis of the same 105 women reported +68.9 mL with oxybutynin and -6.0 mL with placebo, and found that structural equation modeling did not show the clinical improvement to be mediated by the urodynamic or voiding-frequency changes.[2]
Women who were not dry or not satisfied after 8 weeks could add the other treatment in an extension phase. Eight women in the behavioral group (12.7%) added oxybutynin, and their mean reduction rose from 57.5% to 88.5% (P = .034); 27 in the oxybutynin group (41.5%) added behavioral training, and their reduction rose from 72.7% to 84.3% (P = .001).[3] The extension had no concurrent control.[3] In a secondary analysis of the 131 women with nocturia at baseline, the median reduction was 0.50 episodes per night with behavioral training, 0.30 with oxybutynin (P = .02 versus behavioral training) and 0.00 with placebo (P < .001 versus behavioral training; P = .007 versus oxybutynin).[4]
Surgery and treatment delivery
All groups had 4 clinic visits at 2-week intervals over 8 weeks with a daily bladder diary, and nurse practitioners delivered every intervention.[1]
Behavioral training began with anorectal biofeedback to teach selective contraction and relaxation of the pelvic muscles with the abdominal muscles relaxed. The second visit taught urge strategies: pausing, sitting down if possible, relaxing and repeatedly contracting the pelvic muscles to diminish urgency before walking to the toilet. At the third visit, women who had not achieved a 50% reduction in accidents repeated pelvic muscle biofeedback; the fourth visit reviewed progress. Home practice was 45 pelvic muscle exercises daily (15 exercises three times a day), with holds built up to 10 seconds, practice in lying, sitting and standing positions, contractions during provoking activities and practice interrupting or slowing the urinary stream once a day. In the behavioral group, 48 of 65 women (73.8%) received one anorectal biofeedback session, 6 (9.2%) a second anorectal session and 11 (16.9%) a second session with combined bladder-sphincter biofeedback; none had more than 2.[1]
Oxybutynin and placebo were dispensed in identical capsules containing riboflavin phosphate as a marker. Dosing started at 2.5 mg three times daily and was adjusted at visits between 2.5 mg daily and 5.0 mg three times daily to reach the most effective dose each woman could tolerate. Final doses were 2.5 mg daily in 7.5%, 5.0 mg in 19.4%, 7.5 mg in 24.4%, 10.0 mg in 17.9% and 15.0 mg in 26.9%. The 8-week course was completed by 61/65 women (behavioral), 55/67 (oxybutynin) and 53/65 (placebo).[1]
Adverse events
A checklist of five known oxybutynin adverse effects, mixed with dummy symptoms, was completed at each visit. Dry mouth was reported by 96.9% of the oxybutynin group, 54.8% of the placebo group and 34.9% of the behavioral group (P < .001 across groups), and inability to void by 21.5%, 3.2% and 6.3% (P = .002). Constipation (38.5%, 37.1% and 22.2% in the oxybutynin, placebo and behavioral groups), blurred vision (15.4%, 9.7% and 9.5%) and confusion (7.7%, 11.3% and 6.3%) did not differ significantly between groups. Seven women stopped oxybutynin for adverse effects (7/67, 10.4%) and four stopped placebo (4/65, 6.2%); none is listed in the behavioral group. Attrition from all causes was 6.2% with behavioral training, 17.9% with oxybutynin and 18.5% with placebo.[1]
Limitations
The outcome was assessed at 8 weeks, and the authors state that the durability of behavioral treatment and patients' ability to sustain adherence are unknown. The report also notes that behavioral treatment depends on an active, motivated patient, so its value may be limited with cognitive impairment.[1]
Participants were volunteers who were ambulatory, had a Mini-Mental State score of 20 or more and had urodynamic bladder dysfunction, seen at one geriatric clinic; 271 of 468 evaluated women were not randomized. Behavioral and drug assignment could not be blinded, and the primary outcome was a self-kept diary scored by a blinded assistant.[1]
The placebo group, which received the same visits, diaries and nurse contact, still reduced episodes by a mean 39.4%. The authors also note that the oxybutynin arm's 68.5% reflects individualized titration and more clinician contact than routine prescribing. Women who stopped early contributed their last available diary, and attrition differed by arm (6.2% versus 17.9% and 18.5%). Oxybutynin was the only drug tested, and the behavioral arm combined biofeedback with urge-control strategies and home exercises, so the contribution of each component is not separable.[1]
In conversation
- Parallel to OrBIT (2009). Both compared a nondrug overactive bladder treatment with an antimuscarinic; ORBIT used tibial nerve stimulation against tolterodine.
ORBIT also compared a nondrug overactive bladder treatment with an antimuscarinic, but it randomized 100 adults to percutaneous tibial nerve stimulation or tolterodine, with no placebo arm. Burgio 1998 included a placebo-and-attention control, enrolled only women aged 55 years and older and measured the percent reduction in incontinent episodes.
Guidelines and approvals
- AUA/SUFU (2024). The idiopathic overactive bladder guideline cites this trial among reports of behavioral therapy added to drug therapy and states that no significant evidence supports or refutes layering behavioral and pharmacologic treatment. Statement 14 allows combining behavioral therapy, non-invasive therapy, pharmacotherapy and minimally invasive therapy in patients whose symptoms do not respond adequately to monotherapy .[5]
Report history
| Published | Report | Cutoff | Headline results | Citation |
|---|---|---|---|---|
| 1998-12-16 | Initial report · primary | — | Mean percent reduction in weekly incontinent episodes on bladder diary after 8 weeks (behavioral minus oxybutynin shown as the effect; analyzed n = 63, 65, 62): 80.7% vs 68.5% vs 39.4%; difference 12.2; P 0.04 Met Mean weekly incontinent episodes after treatment (baseline 15.8, 15.9 and 15.4; P = .98): 2.8 vs 5.7 vs 8.2; P 0.005 Reduction of at least 50% in incontinent episodes: 57/63 (90.5%) vs 50/65 (76.9%) vs 40/62 (64.5%); P 0.002 Reduction of at least 75% in incontinent episodes: 46/63 (73.0%) vs 37/65 (56.9%) vs 20/62 (32.3%); P <0.001 Reduction of 100% (completely dry): 19/63 (30.2%) vs 15/65 (23.1%) vs 8/62 (12.9%); P 0.07 More incontinent episodes after treatment than before: 1/63 (1.6%) vs 3/65 (4.6%) vs 10/62 (16.1%); — Descriptive Patients describing themselves as much better: 43/58 (74.1%) vs 28/55 (50.9%) vs 14/52 (26.9%); P <0.001 Comfortable enough with treatment to continue indefinitely: 55/57 (96.5%) vs 29/53 (54.7%) vs 22/51 (43.1%); P <0.001 Wished to receive another form of treatment: 8/57 (14.0%) vs 40/53 (75.5%) vs 37/49 (75.5%); P <0.001 Mean change in cystometric capacity, mL (posttreatment cystometry in 105 of 197 subjects): +17.3 vs +70.9 vs -5.9; — | Burgio KL, Locher JL, Goode PS, et al. "Behavioral vs Drug Treatment for Urge Urinary Incontinence in Older Women: A Randomized Controlled Trial." *JAMA.* 1998;280(23):1995-2000. doi:10.1001/jama.280.23.1995 DOI |
| 2000 | Combined therapy extension · secondary | — | Mean percent reduction after adding oxybutynin to behavioral training, in the 8 subjects (12.7%) not dry or satisfied after behavioral training alone: single therapy versus combined: 57.5% to 88.5%; P 0.034 Mean percent reduction after adding behavioral training to oxybutynin, in the 27 subjects (41.5%) not dry or satisfied after drug alone: single therapy versus combined: 72.7% to 84.3%; P 0.001 | Burgio KL, Locher JL, Goode PS. "Combined Behavioral and Drug Therapy for Urge Incontinence in Older Women." *J Am Geriatr Soc.* 2000;48(4):370-374. doi:10.1111/j.1532-5415.2000.tb04692.x DOI |
| 2002 | Urodynamic analysis · secondary | — | Mean change in cystometric capacity, mL (n = 105): +17.3 vs +68.9 vs -6.0; P <0.001 (oxybutynin) Mean increase in volume at strong desire to void, mL: +40.5 vs +69.9 vs +7.8; — Detrusor instability resolved after treatment, among women who had it at baseline: 1 of 7 vs 7 of 8 vs 7 of 12; — Mean percent reduction in incontinent episodes in the urodynamic subsample (n = 105): 82.3% vs 78.3% vs 51.5%; P 0.002 | Goode PS, Burgio KL, Locher JL, Umlauf MG, Lloyd LK, Roth DL. "Urodynamic Changes Associated with Behavioral and Drug Treatment of Urge Incontinence in Older Women." *J Am Geriatr Soc.* 2002;50(5):808-816. doi:10.1046/j.1532-5415.2002.50204.x DOI |
| 2005 | Nocturia analysis · secondary | — | Median reduction in nocturia episodes per night (secondary analysis of the parent trial): 0.50 vs 0.30 vs 0.00; P 0.02 (behavioral vs oxybutynin) | Johnson TM 2nd, Burgio KL, Redden DT, Wright KC, Goode PS. "Effects of Behavioral and Drug Therapy on Nocturia in Older Incontinent Women." *J Am Geriatr Soc.* 2005;53(5):846-850. doi:10.1111/j.1532-5415.2005.53260.x DOI |
References
1. Burgio KL, Locher JL, Goode PS, et al. "Behavioral vs Drug Treatment for Urge Urinary Incontinence in Older Women: A Randomized Controlled Trial." JAMA. 1998;280(23):1995-2000. doi:10.1001/jama.280.23.1995
2. Goode PS, Burgio KL, Locher JL, Umlauf MG, Lloyd LK, Roth DL. "Urodynamic Changes Associated with Behavioral and Drug Treatment of Urge Incontinence in Older Women." J Am Geriatr Soc. 2002;50(5):808-816. doi:10.1046/j.1532-5415.2002.50204.x
3. Burgio KL, Locher JL, Goode PS. "Combined Behavioral and Drug Therapy for Urge Incontinence in Older Women." J Am Geriatr Soc. 2000;48(4):370-374. doi:10.1111/j.1532-5415.2000.tb04692.x
4. Johnson TM 2nd, Burgio KL, Redden DT, Wright KC, Goode PS. "Effects of Behavioral and Drug Therapy on Nocturia in Older Incontinent Women." J Am Geriatr Soc. 2005;53(5):846-850. doi:10.1111/j.1532-5415.2005.53260.x
5. Cameron AP, Chung DE, Dielubanza EJ, et al. "The AUA/SUFU Guideline on the Diagnosis and Treatment of Idiopathic Overactive Bladder." J Urol. 2024;212(1):11-20. doi:10.1097/JU.0000000000003985