Protocol- and Therapist-Related Variables Affecting Outcomes of Behavioral Interventions for Urinary and Fecal Incontinence

Transcription

1 GASTROENTEROLOGY 2004;126:S152 S158 Protocol- and Therapist-Related Variables Affecting Outcomes of Behavioral Interventions for Urinary and Fecal Incontinence JEANNETTE TRIES Center for Disorders of Incontinence and Elimination, Aurora Sinai Medical Center, Milwaukee, Wisconsin Biofeedbacktechniques used to treat urinary and fecal incontinence lackstandardization. Most early protocols used a pressure device placed within the vagina or anal canal, or electromyographic (EMG) sensors in the same locations, to measure the external anal sphincter (EAS) or pelvic floor muscle (PFM) contractile function, and most early studies provided feedbackfrom a single physiological transducer. The goal was to improve bowel and bladder control by improving EAS or PFM contractile function. Protocols that have resulted in the most consistent reductions in urinary incontinent episodes used 2 or more channels of physiological information to reinforce stable abdominal and bladder pressures concurrently with PFM contraction. For fecal incontinence, more significant treatment results were derived when protocols measured (1) patient perception of sensory cues associated with rectal distention and potential loss of stool, (2) short-latency EAS contraction when perceiving rectal distention, (3) inhibition of (extraneous muscle) activity that would increase intra-abdominal pressure during EAS contraction, and (4) reinforcement of sustained (up to 30 seconds) contractions rather than only brief 1- to 2-second contractions. Limited data support the use of surface abdominal EMG measures as indices of extraneous muscle activity associated with increased intra-abdominal pressure and anal or vaginal EMG probes to obtain measures of PFM function. Better results may also be obtained when there are at least 4 training sessions, when daily home exercises are prescribed, and when the therapist is well trained and experienced. These inferences are based for the most part on indirect evidence, and more studies are needed that compare different treatment protocols. Biofeedback, pelvic floor muscle (PFM) exercises, urge-control strategies, timed voiding, and habit training are all behavioral treatments that have been reported to be effective in reducing fecal and urinary incontinence. 1 3 Because they are noninvasive, behavioral treatments for incontinence have obvious advantages over surgery and drugs. However, behavioral treatments have not been adequately assessed in controlled studies. Biofeedback can be defined as the use of an instrument that delivers a concurrent measurement of selected biological responses to enable the individual to alter his/her physiology in directions associated with improved function. 4 Because biofeedback is assumed to work by altering physiology, all biofeedback protocols should be assessed on the basis of 2 primary criteria: (1) the adequacy of the physiological measurement methods in terms of reliability and validity and (2) the reinforcing (motivational) characteristics of the procedure. Reinforcing characteristics of a biofeedback protocol include the quality of the feedback display, the effectiveness of the methods used to shape a change in response, time spent in treatment, and strategies that promote generalizing physiological changes from treatment to daily life. Unfortunately, many reports on biofeedback for incontinence lack precise descriptions of the reliability and sensitivity of the physiological measures and the methods used to motivate learning in the treatment. Physiologic changes that are presumed to have been altered by the manipulation are often not reported. When physiologic changes are reported, they are not consistently associated with functional improvements. Although the competence of the biofeedback professional is a variable in biofeedback outcomes, therapist-related variables are seldom described sufficiently to allow for evaluation or replication. Given these problems, it has been difficult to determine which biofeedback protocols lead to optimum symptom reduction. Variability in the Protocols Used for Biofeedback Treatment of Incontinence Reinforcement of a Single Physiologic Variable The earliest biofeedback studies for urinary and fecal incontinence used a pressure device placed within the vagina or anal canal to measure external anal sphinc- Abbreviations used in this paper: EAS, external anal sphincter; EMG, electromyographic; PFM, pelvic floor muscle by the American Gastroenterological Association /04/$30.00 doi: /j.gastro

2 January Supplement 2004 PROTOCOL- AND THERAPIST-RELATED VARIABLES S153 ter (EAS) or pelvic floor muscle (PFM) activity. 5 7 The goal of these early protocols was to improve bowel and bladder control by improving EAS or PFM contractile function. Instead of measuring PFM activity, Cardozo et al. 8,9 used visual feedback of bladder pressure during a cystometrogram to reinforce inhibition of detrusor instability in subjects with urge urinary incontinence. Although this protocol reinforced changes in bladder smooth muscle, it called for central nervous system mediation because subjects were instructed to maintain overall body relaxation and to breathe slowly and diaphragmatically to facilitate bladder inhibition. In contrast, most protocols for incontinence that have reinforced a single physiologic variable used either pressure or electromyographic (EMG) measures of PFM contractile activity from the EAS, vagina, or perianal skin surface Multichannel (Multiple Physiologic Variables) Biofeedback Using a manometric 3-balloon assembly probe and continuous polygraph feedback, Engel et al. 21 treated fecal incontinence by reinforcing changes in 3 distinct physiologic variables rather than just in EAS contraction. The responses that were reinforced included (1) perception of sensory cues associated with rectal distention and potential loss of stool, (2) a short-latency EAS contraction when perceiving rectal distention, and (3) inhibition of activity that would increase intra-abdominal pressure during EAS contraction. Significant clinical improvement was obtained in this and other uncontrolled investigations that used similar procedures This manometric protocol for fecal incontinence was modified to treat urinary incontinence in a series of studies Using rectal and EAS pressure measures obtained from the 3-balloon assembly probe, subjects were reinforced for selectively contracting the EAS during bladder urgency while maintaining stable intraabdominal and bladder pressure. EAS contraction was treated as a surrogate for external urethral contraction, which is technically more difficult to measure. This assumption is reasonable (in the absence of specific injury to the EAS or external urethral sphincter) because the innervation to the 2 sphincters is the same, and the 2 sphincters can be observed to contract together when both are measured simultaneously. In this biofeedback protocol, urgency to urinate was triggered (reproduced) by infusion of sterile water into the bladder as is done during a cystometrogram. Therefore, this protocol simultaneously measured and reinforced several variables: (1) awareness of bladder fullness and perception of cues associated with the onset of uninhibited bladder activity, (2) selective EAS contraction during bladder urgency, (3) inhibition of abdominal wall contractions (detectable as an increase in rectal pressure) that might increase bladder pressure during EAS contraction, and (4) inhibition of bladder contractions (detectable as an increase in bladder pressure in the absence of an increase in rectal pressure). Counterproductive responses that would increase intraabdominal and bladder pressures were suppressed because the reinforcement of EAS contraction was contingent on maintaining stability during the recording of rectal pressure. Some researchers have modified the manometric protocol by substituting surface abdominal EMG electrodes for rectal pressure to measure extraneous abdominal contraction Vaginal or anal canal EMG electrode probes, or perianal electrodes, have also replaced manometric EAS measures in some studies (Figure 1). Although reliability and validity data for surface EMG are limited, they do support the use of surface abdominal EMG measures as indices of extraneous muscle activity associated with increased intra-abdominal pressure 39 and the use of anal or vaginal EMG probes to obtain measures of PFM function. 40 Possible Advantages of a Multivariable Protocol Protocols that have resulted in the most consistent reductions in urinary incontinent episodes, between 76% and 87% when postprostatectomy patients with continual leakage were excluded, included reinforcement of stable abdominal and bladder pressures concurrently with PFM contraction Although seemingly more complex than single-variable biofeedback, multivariable protocols appear to produce greater reductions in urinary incontinence and to achieve this in fewer sessions (average approximately 5 sessions) when compared with protocols that use a single measure of PFM function (average approximately 11 sessions). 41 For example, when Burgio et al. 30 reinforced selective PFM contraction while maintaining stable intra-abdominal pressure, they obtained an 81% reduction in urge urinary incontinence episodes in 65 female patients, with 74% of the women requiring only a single session of biofeedback. Only 17% of the patients needed a second biofeedback session that combined bladder pressure feedback with selective PFM training. Although no subjects received more than 2 biofeedback sessions, all had 4 clinic visits to review home exercises, diary use, and urge-control strategies. There are several apparent advantages to using multivariable protocols. First, maintaining a stable intra-abdominal pressure during EAS or PFM contraction is

3 S154 JEANNETTE TRIES GASTROENTEROLOGY Vol. 126, No. 1 Figure 1. Diagram of a modified manometric protocol that substitutes surface abdominal EMG electrodes for rectal pressure to measure extraneous abdominal contraction. The figure also shows an intravaginal EMG probe positioned above the urogenital diaphragm (arrows). A catheter shown in the bladder records changes in bladder pressure and transmission of intra-abdominal pressure to the bladder from contracting abdominal muscles. (Adapted with permission. 38 ) incompatible with extraneous abdominal wall contraction, which could increase bladder or rectal pressure in situations of urgency. If maladaptive increases in intraabdominal pressure can be reduced during times of urgency, individuals may be more successful in maintaining a positive urethral or anal canal closure pressure relative to rectal or bladder pressure, even when PFM strength has not been completely restored. Second, extraneous muscle activity or intra-abdominal pressure could be transmitted to the PFM recording device and potentially add artifacts to the recordings. 36,42 On the other hand, some clinicians feel that extraneous muscle contraction might actually augment PFM contraction measures through reciprocal muscle facilitation. The effect of such facilitation may or may not be useful in the initial stages of reinforcing PFM contraction to improve strength. This is an empiric question that remains unstudied. Alternatively, PFM contraction that is facilitated by extraneous muscle activity may be maladaptive. Indeed, it has been observed that when attempts at PFM contraction are associated with increased intra-abdominal pressure, urethral resistance will decrease in a significant number of cases. 43 Therefore, treatment protocols that measure only PFM activity may be less effective than protocols that control for such extraneous contraction and/or increases in intra-abdominal pressure during PFM or EAS contraction. No studies have directly compared single-channel with multichannel biofeedback training protocols. The degree to which a PFM contraction can be produced selectively without cocontraction of extraneous muscle may more closely reflect functional improvement than measures of maximum EAS or PFM contraction alone. This may explain why some studies have reported functional improvement in bowel and bladder control without associated improvements in EAS or PFM strength. Moreover, other parameters such as PFM resting muscle tone, latency to maximum contraction, and stability of the contraction over an extended time period often change with treatment. Currently, it is not known whether changes in muscle characteristics other than maximum contraction are associated with treatment outcome. 44 Coordination Training: Reinforcing a Short-Latency EAS Contraction to the Perception of a Rectal Stimulus Several reports have concluded that when the classic manometric protocol is used, continence improves primarily from the acquisition of a short-latency EAS response to rectal stimuli. 3,21 This procedure is aptly

4 January Supplement 2004 PROTOCOL- AND THERAPIST-RELATED VARIABLES S155 termed coordination training 2 because it reinforces coordination of the EAS contraction with increases in intrarectal pressure and with the associated perception of a rectal stimulus. In support of this procedure, 2 studies that attempted to identify the effective components of manometric biofeedback training for fecal incontinence found that only improvement in rectal sensation was associated with improved stool continence. 45,46 Both reports concluded that EAS strength was not modified with biofeedback. However, not all subjects with improved sensation developed continence; this finding indicates that rectal sensation is a necessary but not sufficient variable for continence. There are methodological weaknesses in these 2 components analysis studies that limit the generalizability of the findings. 45,46 The observation that biofeedback improves rectal sensation but not EAS strength could be because of an order effect. 2 Because a portion of rectal sensitivity is accounted for by the integrity of receptors within the PFM, it is not known if sensation would also improve after EAS strength training but without sensory training. The failure to show changes in EAS strength may reflect inadequacies in the protocol rather than intrinsic sphincter weakness that is unresponsive to operant procedures. Moreover, the 2 component analysis studies reinforced EAS contraction verbally, with the examiner using instrumental feedback to guide subjects. 45,46 It is generally accepted that the degree to which a physiologic variable can be shaped with operant procedures depends on the accuracy and immediacy of the feedback. It is unlikely that therapist-produced verbal reinforcement can provide the kind of precise sensory information needed to direct and update motor plans and thereby improve EAS contraction, especially when sphincter afferent and efferent activity is compromised by trauma or disease. Moreover, it is unreasonable to expect that reinforcement for only a short-latency EAS contraction would change muscle endurance or maximum contraction. Because the protocol used in these studies did not directly reinforce activity that could alter weak muscle, a comparison between strengthening and sensory discrimination effects cannot be made. Sustained PFM/EAS Contractions Several studies that used protocols designed primarily to improve EAS strength 19,20,47,48 reported significant reductions in fecal incontinence. Two of those studies also obtained measurable changes in EAS strength using protocols that reinforced sustained EAS contraction of 10 to 30 seconds while controlling for increases in intra-abdominal pressure. 47,48 Chiarioni et al., 47 using a protocol that reinforced sustained 30-second EAS contractions, reported an 85% reduction in fecal incontinence for 14 subjects with chronic diarrhea who had failed prior medical treatment. Subjects improved their ability to sustain an EAS contraction from 19 to 38 seconds. All subjects who developed continence learned to sustain a 30-second EAS contraction. These findings differ from those obtained by Loening- Baucke, 49 who drew subjects from a similar population. Fecal incontinence was reduced by only 50% after biofeedback combined with conventional medical treatment, and outcomes did not differ from those for subjects who received conventional therapy alone. 49 Moreover, biofeedback did not improve EAS strength. The disparity between the outcomes in these 2 studies 47,49 appears to be more an effect of the differences in the protocols than any difference in the study samples. In the Loening-Baucke study, 49 subjects received reinforcement for only a short-latency EAS contraction to a rectal stimulus. No attention was given to training a sustained EAS contraction that would exceed the time required for recovery of internal anal sphincter inhibition after rectal distention. Moreover, the home exercise program prescribed was not specified, so it cannot be assessed relative to its potential to strengthen the EAS. 49 In contrast, Chiarioni et al. 47 were systematic in reinforcing sphincter contractions of greater intensity and duration (up to 30 seconds) in response to rectal distention and in the absence of extraneous abdominal contraction. Patients were also assigned a well-defined home exercise program that included 20- to 30-second EAS contractions repeated 3 times daily. It appears that changes in EAS and PFM strength can be altered with biofeedback when the protocol reinforces sustained EAS contraction while controlling for extraneous activity. Learning to selectively contract the EAS for 10 to 30 seconds requires a distinctly different skill than learning a short-latency contraction because individuals must learn to coordinate their breathing during the longer contraction. This skill has utility in daily life during urgency: if a rhythmic breathing pattern can be maintained during EAS contraction, there is less tendency for intra-abdominal pressure to increase. Learning to perceive a rectal stimulus and then to immediately contract the EAS for 1 to 2 seconds is an important but less complex task. Although the reinforcement of a shortlatency EAS contraction improves the rectosphincteric response and is associated with improved bowel control, there are indications that such training does not alter EAS or PFM strength. 45,46 Currently, it is not known if changing EAS strength can produce additional effects on outcome. However, it would seem possible that in some

5 S156 JEANNETTE TRIES GASTROENTEROLOGY Vol. 126, No. 1 individuals, improving EAS strength could mean the difference between bowel control that is slightly improved and that which is greatly improved. This difference could have considerable impact on quality of life. Quality of the Reinforcing Stimulus The quality of the feedback display is a critical biofeedback variable because it informs the patient and therapist whether moment-to-moment alterations in motor responses have been successful. Initially, manometric polygraph recordings displayed on a strip chart recorder were used for visual reinforcement. Currently, polygraph presentations are computerized, but the graphics differ by manufacturer. For example, some visual displays juxtapose EAS and abdominal measures on the same graph by using the same ordinate to emphasize the desired response of keeping abdominal activity low during EAS contraction. Protocols that allow for the most flexible presentation of physiologic information appear to have an advantage over those with less flexibility. Flexibility in the presentation of data also applies to thresholds, auditory signals, and gain adjustments. A skilled therapist can alter these functions to meet the learning needs of the individual. The process by which feedback information is methodically manipulated into a meaningful format could conceivably determine whether a biofeedback procedure can successfully transform responses that are initially weak and without conscious awareness into information that can be generalized to the patient s internal sensory cues. To date, no consensus has been achieved with respect to the optimal manner for presenting physiologic data during biofeedback training for incontinence. As a result, the development and purchase of biofeedback equipment occurs on the basis of intuition, manufacturers claims, costs, and other expediencies. Nonspecific Variables Biofeedback is seldom used without adjunctive behavioral strategies such as diaries, urge-control strategies, voiding schedules, dietary or fluid manipulations, and PFM exercises done at home. Only a limited number of studies have examined the effects of these nonspecific components of treatment Whitehead et al. 50 showed that among fecally incontinent children with myelomeningocele, behavioral toilet training (which is often combined with biofeedback training in this patient group) accounted for most of the clinical improvements seen with biofeedback. However, these investigators identified a subgroup of patients with lower and less complete spinal cord lesions and with more frequent bowel movements who obtained significant additional benefit from biofeedback. In a recently published study, Norton et al. 51 compared education about how to cope with incontinence with biofeedback and found no benefit for biofeedback. In contrast to this finding, however, Heymen et al. 52 have described a staged-intervention study in which 1 month of education and medical management of incontinence preceded biofeedback or PFM exercise training. This study showed that although many patients obtained adequate relief with education and advice alone, two thirds of those who were still incontinent after a month of this nonspecific treatment went on to achieve adequate relief of their incontinence when behavioral training was provided. Therapist-Related Variables Affecting Outcomes In part, several researchers have attributed favorable treatment outcomes to the skill and personal characteristics of the therapists who provided the behavioral treatment. 16,34 Qualities acknowledged to influence outcomes include training, knowledge of the physiologic and learning principles that underlie the treatment, communication skills, warmth, and patience. Biofeedback therapists for incontinence come from many different professions, including psychology, physical and occupational therapy, medicine, and nursing. Currently, there are no minimal standards for therapists who use biofeedback for incontinence. Recommendations for Research To help researchers and clinicians design treatment protocols that optimize outcomes, a preliminary set of guidelines is offered. Where possible, these guidelines were based on the preceding literature review. However, this sparse literature has been supplemented in places by the author s experience in order to provide a more complete set of guidelines. For most applications of biofeedback for incontinence, protocols should include a minimum of 4 treatment sessions of at least 1-hour duration, with intervals between treatments ranging from 1 to 4 weeks. Given the apparent benefits of reinforcing multiple physiological modalities, protocols should include measures of either intra-abdominal pressure or abdominal EMG in addition to measures of PFM or EAS function. EAS or PFM contraction should be reinforced for at least 10 seconds in addition to short-latency contractions. Because improving the EAS response to rectal sensation has been associated with reductions in fecal incontinence, protocols for fecal incontinence should include reinforcement of shortlatency EAS contraction in response to the perception of

6 January Supplement 2004 PROTOCOL- AND THERAPIST-RELATED VARIABLES S157 Table 1. Important Elements to Include in Descriptions of Biofeedback and Behavioral Treatment Protocols Physiologic training variables used Qualifications of the therapist Description of measurement methods (equipment, feedback display, gain settings, use of thresholds) Methods of reinforcement Number or training trials Time spent in treatment Methods used to generalize skills from the clinic to daily life situation (i.e., specifics of the home exercise program) Description of all adjunctive behavioral strategies (i.e., education, urge control and habit training, dietary changes) Physiologic and functional outcome measures obtained at treatment completion and follow-up a rectal stimulus when this function is found to be impaired. Ideally, research designs should use controls for delineating the effects of sensory and EAS coordination training and distinguishing those from effects obtained through EAS strength training. Protocols should include the assignment of home exercises. Although there is no consensus as to the ideal number of PFM contractions to be used in home exercises, subjects should perform some PFM contractions daily. In our clinic, we generally assign 28 to 36 PFM contractions daily. These exercises include quick 1- to 2-second, 10-second, and 30-second contractions. The total number of contractions assigned daily is divided into 3 sets. Each set of 8 to 12 contractions is done at different times of the day and in different positions (sitting, standing, and lying). However, the exercise program is individualized on the basis of how well each patient can perform selective PFM contractions in the clinic. Follow-up data should be obtained at 3, 6, and 12 months. To allow for adequate replication, research reports should clearly define all aspects of the biofeedback and behavioral manipulation. These elements are summarized in Table 1. The literature suggests that biofeedback is generally effective in reducing urinary and fecal incontinence. However, because of the many shortcomings in research methods, we do not know which protocolrelated variables contribute most to symptom reduction. Therefore, the most efficient and effective protocols must be established empirically to advance this important treatment and set standards of care for the future. References 1. Fantl JA, Newman DK, Colling J, DeLancey JOL, Keeys C, Loughery R, McDowell BJ, Norton P, Ouslander J, Schnelle J, Staskin D, Tries J, Urich V, Vitousek SH, Weiss BD, Whitmore K. Urinary incontinence in adults: acute and chronic management. Clinical practice guideline, No. 2, 1996 Update. Rockville, MD: Agency for Health Care Policy and Research, US Dept of Health and Human Services. Public Health Service; AHCPR Publication No Heymen S, Jones KR, Ringel Y, Scarlett Y, Whitehead WE. Biofeedback treatment of fecal incontinence: a critical review. Dis Colon Rectum 2000;44: Enck P. Biofeedback training in disordered defecation: a critical review. Dig Dis Sci 1993;38: Miller NE. Biofeedback and visceral learning. Ann Rev Psychophysiol 1978;29: Kohlenberg JR. Operant conditioning of human anal sphincter pressure. J Appl Behav Anal 1973;6: Kegel AH. Progressive resistance exercise in the functional restoration of the perineal muscles. Am J Obstet Gynecol 1948;56: Kegel AH. Physiologic therapy for urinary stress incontinence. J Am Med Soc 1951;146: Cardozo LD, Abrams PD, Stanton SL, Feneley RC. Idiopathic bladder instability treated by biofeedback. Br J Urol 1978;50: Cardozo L, Stanton SL, Hafner J, Allan V. Biofeedback in the treatment of detrusor instability. Br J Urol 1978;50: Baigis-Smith J, Smith DA, Rose M, Newman DK. Managing urinary incontinence in community-residing elderly persons. Gerontologist 1989;29: Burns PA, Marecki MA, Dittmar SS, Bullough B. Kegel s exercises with biofeedback therapy for treatment of stress incontinence. Nurse Practitioner 1985;(Feb): Burns PA, Pranikoff K, Nochajski TH, Hadley EC, Levy KL, Ory MG. A comparison of effectiveness of biofeedback and pelvic muscle exercise treatment of stress incontinence in older community dwelling women. J Gerontol 1993;48: Castleden CM, Duffin HM. Guidelines for controlling urinary incontinence without drugs or catheters. Age Ageing 1981;10: Fisher W. Physiotherapeutic aspects of urine incontinence. Acta Obstet Gynecol Scand 1983;62: Henderson JS, Taylor KH. Age as a variable in an exercise program for the treatment of simple urinary stress incontinence. J Obset Gynecol Neonatal Nurs 1987;16: Susset JG, Galea G, Read L. Biofeedback therapy for female incontinence due to low urethral resistance. J Urol 1990;143: Rose MA, Baigis-Smith J, Smith D, Newman D. Behavioral management of urinary incontinence in homebound older adults. Home Healthcare Nurse 1990;8: van Tets WF, Kuijpers JH, Bleijenberg G. Biofeedback treatment is ineffective in neurogenic fecal incontinence. Dis Colon Rectum 1996;39: Schiller LR, Santa Ana C, Davis GR, Fordtran JS. Fecal incontinence in chronic diarrhea: report of a case with improvement after training with rectally infused saline. Gastroenterology 1979; 77: MacLeod JH. Management of anal incontinence by biofeedback. Gastroenterology 1987;93: Engel BT, Nikoomanesh P, Schuster MM. Operant conditioning of rectosphincteric responses in the treatment of fecal incontinence. N Engl J Med 1974;290: Cerulli MA, Nikoomanesh P, Schuster MM. Progress in biofeedback conditioning for fecal incontinence. Gastroenterology 1979; 76: Wald A. Biofeedback therapy of fecal incontinence. Ann Intern Med 1981;95: Wald A, Tunuguntla AK. Anorectal sensorimotor dysfunction in fecal incontinence and diabetes mellitus. N Engl J Med 1984; 310: Goldenberg DA, Hodges K, Hersh T, Jinich H. Biofeedback ther-

7 S158 JEANNETTE TRIES GASTROENTEROLOGY Vol. 126, No. 1 apy for fecal incontinence. Am J Gastroenterol 1980;74: Riboli EB, Frascio MJ, Pitto G, Reboa G, Zanolla R. Biofeedback conditioning for fecal incontinence. Arch Phys Med Rehabil 1988; 69: Rao SS, Welcher KD, Happel J. Can biofeedback therapy improve anorectal function in fecal incontinence? Am J Gastroenterol 1996;91: Glia A, Gylin M, Akerlund JE, Lindfors U, Lindberg G. Biofeedback training in patients with fecal incontinence. Dis Colon Rectum 1998;41: Burgio KL, Burgio LD. Behavior therapies for urinary incontinence in the elderly. Clin Geriatr Med 1986;2: Burgio KL, Locher JL, Goode PS, Hardin JM, McDowell BJ, Dombrowski M, Candib D. Behavioral vs drug treatment for urge urinary incontinence in older women: a randomized controlled trial. JAMA 1998;280: Burgio KL, Robinson JC, Engel BT. The role of biofeedback in Kegel exercise training for stress urinary incontinence. Am J Obstet Gynecol 1986;154: Burgio KL, Stutzman RE, Engel BT. Training for post-prostatectomy urinary incontinence. J Urol 1989;141: Burgio KL, Whitehead WE, Engel BT. Urinary incontinence in the elderly: bladder-sphincter biofeedback and toileting skills training. Ann Intern Med 1985;104: Burton JR, Pearce KL, Burgio KL, Engel BT, Whitehead WE. Behavioral training for urinary incontinence in elderly ambulatory patients. J Am Geriatr Soc 1988;36: Middaugh SJ, Whitehead WE, Burgio KL, Engel BT. Biofeedback in treatment of urinary incontinence in stroke patients. Biofeedback Self Regul 1989;14: O Donnell PD, Doyle R. Biofeedback therapy technique for the treatment of urinary incontinence. Urology 1991;37: Tries J. The use of biofeedback in the treatment of incontinence due to head injury. J Head Trauma Rehabil 1990;5: Tries J, Eisman E. Biofeedback for the treatment of urinary incontinence. In: Schwartz ME, ed. Biofeedback: a practitioner s guide. New York, NY: Guilford; Workman DE, Cassisi JE, Dougherty MC. Validation of surface EMG as a measure of intravaginal and intra-abdominal activity: implications for biofeedback-assisted Kegel exercises. Psychophysiol 1992;30: Pinho M, Hoaiw K, Bielecki K, Keighley MRB. Assessment of noninvasive intra-anal electromyography to evaluate sphincter function. Dis Colon Rectum 1991;34: Tries J, Brubaker L. Application of biofeedback in the treatment of urinary incontinence. Professional Psychol Res Prac 1996;27: Bo K, Kvarstein B, Hagen R, Larsen S. Pelvic floor muscle exercise for the treatment of female stress urinary incontinence: reliability of vaginal pressure measurements of pelvic floor muscle strength. Neurourol Urodyn 1990;9: Bump RC, Hurt WG, Fantl JA, Wyman JF. Assessment of Kegel pelvic muscle exercise performance after brief verbal instruction. Am J Obstet Gynecol 1991;165: Dougherty M, Bishop K, Mooney R, Williams B. Graded pelvic muscle exercise: effect on stress urinary incontinence. J Reprod Med 1993;39: Latimer PR, Campbell D, Kasperski J. A component analysis of biofeedback in the management of fecal incontinence. Biofeedback Self Regul 1984;9: Miner PB, Donnelly TC, Read NW. Investigation of the mode of action of biofeedback in the treatment of fecal incontinence. Dig Dis Sci 1990;35: Chiarioni G, Scattolini C, Bonfante F, Vantini I. Liquid stool incontinence with severe urgency: anorectal function and effective biofeedback treatment. Gut 1993;34: Whitehead WE, Burgio KL, Engel BT. Biofeedback treatment of fecal incontinence in geriatric patients. J Am Geriatr Soc 1985; 33: Loening-Baucke V. Efficacy of biofeedback training in improving fecal incontinence and anorectal physiologic function. Gut 1990; 31: Whitehead WE, Parker L, Bosmajian L, Morrill-Corbin ED, Middaugh S, Garwood M, Cataldo MF, Freeman J. Treatment of fecal incontinence in children with spina bifida: comparison of biofeedback and behavior modification. Arch Phys Med Rehabil 1986; 67: Norton C, Chelvanayagam S, Wilson-Barnett J, Redfern S, Kamm MA. Randomised controlled trial of biofeedback for fecal incontinence. Gastroenterology 2003;124: Heymen S, Jones KR, Ringel Y, Scarlett Y, Drossman DA, Whitehead WE. Biofeedback for fecal incontinence and constipation: the role of medical management and education. Gastroenterology 2001;120(suppl 1):A397. Address requests for reprints to: Jeannette Tries, Ph.D., O.T.R., Center for Disorders of Incontinence and Elimination, Aurora Sinai Medical Center, 945 North 12th Street, Milwaukee, Wisconsin jmtries@core.com; fax: (414)