Purpose  

Biofeedback is well established as a treatment for fecal incontinence but little is known about factors that may be associated with its effectiveness. This study assessed short-term outcomes, predictors of patients who completed treatment, and predictors of treatment success.

Methods  

This study was a retrospective review of consecutive patients treated with biofeedback at a tertiary referral colorectal clinic during ten years. Clinical, physiologic, and quality of life measures were collected prospectively at the time of treatment. Regression analysis was performed.

Results  

Of 513 patients, 385 (75 percent) completed the treatment program. Each outcome was improved for more than 70 percent of patients. Incontinence scores decreased by 32 percent (from 7.5 to 5.2 of 13), patient assessment of continence increased by 40 percent (from 5.3 to 3.2 of 10), quality of life improved by 89 percent (from 0.34 to 0.67 of 1.0), and maximum anal sphincter pressure increased by a mean 12 mmHg (14 percent; from 90 to 102 mmHg). Patients who did not complete treatment were younger, were more likely to be male, and had less severe incontinence. Treatment success was predicted by completion of all treatment sessions (odds ratio, 10.34; 95 percent confidence interval, 4.46–24.19), female gender (odds ratio, 4.11; 95 percent confidence interval, 1.04–7.5), older age (odds ratio, 1.02 per year; 95 percent confidence interval, 1–1.04), and more severe incontinence before treatment (odds ratio, 1.19 per unit increase in St. Mark’s score; 95 percent confidence interval, 1.05–1.34).

Conclusions  

More than 70 percent of patients in this large series demonstrated improved short-term outcomes. Treatment success was more likely in those who completed six training sessions, were female, older, or had more severe incontinence. Patients were less likely to complete treatment if they were male, younger, or had milder incontinence.