Introduction
Urinary incontinence (UI) is a common symptom in women, more so in those who are older. From a societal and National Health Service (NHS) perspective, UI is associated with an increased risk of falls, hip fractures, urinary tract and skin infections, and psychiatric co-morbidity.Reference Brown, Vittinghoff, Wyman, Stone, Nevitt, Ensrud and Grady1–Reference Wagner, Hu, Bentkover, LeBlanc, Stewart, Corey, Zhou and Hunt3 From the individual perspective, personal health and hygiene problems, restriction of employment and leisure activities, avoidance behaviour (going out, sexual activity), and increased risk of institutionalization are also important. In addition, there is a significant burden on the carers of those suffering from incontinence, especially if there is co-existent cognitive or physical disability.
Estimates from a decade ago suggested that there were about five million people in the UK with continence problems requiring healthcare,Reference Brown, Vittinghoff, Wyman, Stone, Nevitt, Ensrud and Grady1 including one in three of the over-80 age group. The annual NHS cost of treating urinary incontinence was estimated at £526 million and overall societal cost at £743 million in the Leicestershire MRC Study.Reference Turner, Shaw, McGrother, Dallosso and Cooper4 However, these figures are not current, and are likely to be a gross underestimate in view of the ageing population.
In this article, the evidence for various non-pharmacological (NP) interventions is reviewed, including lifestyle interventions, physical therapies, behavioural therapies and containment. Anticholinergics have been reviewed previously,Reference Mappilakkandy and Mistri5 and surgical interventions will be considered in a future article.
An experienced clinical librarian advised on the search strategy (available on request). The relevant Cochrane Reviews were reviewed and referenced. This was a pragmatic review, rather than an exhaustive systematic review of the literature, aimed at enabling the reader to make informed management decisions.
Assessment
The initial assessment should be used to guide therapy by incorporating the following elements:
(1) Categorization of the type of incontinence – stress incontinence, urge incontinence, mixed incontinence, others (situational, overflow, polyuria-related, associated with cognitive deficit).
(2) Assessment of severity from the patient point of view, including symptom burden and impact on quality of life.
(3) Identification of potentially reversible associated conditions.
(4) Elucidation of past medical history including a detailed obstetric history.
(5) Screening for indications of referral (urgent versus non-urgent).
Three-day bladder diaries, frequency volume charts, and urodynamic testing can be considered where the type of incontinence is unclear, when NP and pharmacological interventions have failed, and where surgery is being considered, or has been unsuccessful.
Initial evaluation should also incorporate a functional and cognitive assessment aimed at differentiating the following two categories of people: the cognitively intact and physically able person, who can live independently; and the frail elderly with cognitive dysfunction and/or physical disability requiring carers. This will allow tailoring of advice based on individual circumstances.
Treatment goals
The initial aim of successful management of incontinence should always be ‘independent continence’. The approach depends on the cognitive and physical status of the individual, though some form of NP intervention is preferred initially. Carer involvement should be encouraged right from the outset. A further assessment should be planned after a suitable interval in liaison with the patient, to identify failure of a NP approach, and to consider appropriate pharmacological therapy. Successful achievement of continence with drugs is termed ‘dependent continence’. If unsuccessful, consideration of surgical options is warranted. A secondary goal of ‘contained incontinence’ using containment methods is the last resort where incontinence persists despite all interventions, or to be used while other interventions are in process.
Serial use of validated quality of life (QOL) assessments enables a greater understanding of the impact of incontinence and the benefits of treatment to patients, which cannot always be quantified by an estimation of symptom burden.Reference Kelleher and Cortes6 Thus QOL assessment should be incorporated into routine assessment of patients presenting with UI. A patient's self-assessment of the appropriate balance of therapeutic effect and adverse effects is helpful in dose titration, and a review of QOL charts can aid this assessment.
Non-pharmacological management
The non-pharmacological management of UI will be considered under the following categories:
(A) Lifestyle interventions
(1) weight reduction
(2) reduction of caffeine intake
(3) modifying fluid intake
(4) management of constipation
(5) smoking cessation
(6) urgency control techniques
(B) Physical therapies
(1) pelvic floor muscle training
(2) biofeedback
(3) electrical stimulation
(4) magnetic stimulation
(C) Behavioural therapies
(1) bladder training
(2) toileting schedules
(D) Containment
Lifestyle interventions
Patient education and an empathetic approach are essential elements of management. The importance of allowing adequate time for sensitive exploration of symptoms and their impact should never be underestimated. A dedicated continence clinic with experienced personnel facilitates the whole process.
Weight reduction. UI is associated with high body mass index (BMI),Reference Asplund and Aberg7–Reference Hannestad, Rortveit, Daltveit and Hunskaar11 and obesity is an independent risk factor for developing overactive bladder and stress incontinence.Reference Dallosso, McGrother, Matthews and Donaldson9 A 6-month weight loss programme reduced incontinence episodes in women with a BMI of 25 to 50, with a significantly greater reduction in the frequency of stress incontinence episodes, and no significant change in urge incontinence episodes.Reference Subak, Wing, West, Franklin, Vittinghoff, Creasman, Richter, Myers, Burgio, Gorin, Macer, Kusek and Grady12 Surgically induced weight loss in morbidly obese women has been shown to improve incontinence symptoms.Reference Bump, Sugerman, Fantl and McClish13–Reference Burgio, Richter, Clements, Redden and Goode15 Weight loss advice is appropriate in women with a BMI > 30.
Reduction of caffeine intake. While average caffeine intake has no significant diuretic effect, it can promote detrusor excitability. Administration of caffeine was associated with a statistically significant increase in detrusor pressure on bladder filling in those with detrusor instability, compared with normal subjects.Reference Creighton and Stanton16 Mean caffeine intake of women with detrusor instability has been reported to be higher than that of controls,Reference Arya, Myers and Jackson17 and a reduction in caffeine intake has been associated with a significant decrease in incidents of urgency and frequencyReference Bryant, Dowell and Fairbrother18 and a non-significant reduction in incontinence episodes.Reference Tomlinson, Dougherty, Pendergast, Boyington, Coffman and Pickens19 There is some evidence that carbonated drinks and artificial sweeteners may increase bladder symptoms, and it is worth considering a trial period of complete cessation of any identified bladder irritants to assess symptom response.
Modification of fluid intake. There is a weak correlation between fluid intake and severity of UI in women (over 55) with stress UI, and no correlation in women with urge UI.Reference Wyman, Elswick, Wilson and Fantl20 Nonetheless, after a week of caffeine restriction, reduction of fluid intake led to a significant reduction in urgency and incontinence.Reference Swithinbank, Hashim and Abrams21 Another study suggested that nocturnal toileting and evening fluid restriction may reduce nocturnal urine loss by a small but useful amount in older patients with severe urge incontinence.Reference Griffiths, McCracken, Harrison and Gormley22 A trial of regulated fluid intake in terms of volume and timing is relatively easy and free of side-effects, can be applied by carers in those with physical or cognitive limitations, and should be advised in conjunction with reduced caffeine intake, prior to considering anticholinergic medications.
Management of constipation. A study of Chinese women suggested that constipation was associated with the risk of stress and urge incontinence,Reference Song, Zhang, Song and Xu23 but no studies were found which looked at the effects of treating constipation on incontinence. Medical treatment of constipation did improve lower urinary tract symptoms (urgency, frequency, dysuria).Reference Charach, Greenstein, Rabinovich, Groskopf and Weintraub24 Anecdotally, constipation is associated with urinary retention, and maintenance of a regular bowel habit is important in the management of continence.
Smoking cessation. Smoking may exacerbate incontinence due to raised intra-abdominal pressure associated with chronic cough, and nicotine-related detrusor overactivity. The prevailing evidence includes conflicting views on the association between smoking and UI, with a positive association between UI in current smokers compared with those who have never smoked in some studies,Reference Dallosso, McGrother, Matthews and Donaldson9, Reference Hannestad, Rortveit, Daltveit and Hunskaar11 and no association in others.Reference Roe and Doll10, Reference Burgio, Matthews and Engel25 In one study, smoking was associated with an increased prevalence of urinary tract symptoms (current smokers were at highest risk, and ex-smokers had moderate risk, compared with non-smokers).Reference Koskimaki, Hakama, Huhtala and Tammela26 There are no studies that show the effects of smoking cessation on UI, i.e. no conclusive evidence that smoking cessation will improve UI. However, in view of the higher associated symptom burden and association with vascular and neoplastic diseases, smoking cessation should be encouraged while definitive evidence is awaited.
Urgency control techniques. An immediate response to void following an urge should be discouraged, and advice provided about: (a) deep breathing techniques to relax the bladder; (b) simple distraction techniques (activities that require mental concentration) to delay voiding; (c) use of a pelvic floor muscle contraction or crossing of legs to suppress the micturition reflex; and (d) self-motivation to progressively increase delay to voiding.
Physical therapies
Pelvic floor muscle training. Pelvic floor muscle training (PFMT) was first defined in 1936 as ‘tensing and relaxing of the pelvic floor muscles’ for prevention and treatment of urinary and faecal incontinence.Reference Morris27 The first report of PFMT for treatment of urinary incontinence was a case series of 64 females with stress urinary incontinence where PFM exercises with a pressure biofeedback perineometer were effective.Reference Kegel28
A Cochrane review (updated 2010) identified 14 trials that addressed whether PFMT was better than no treatment, placebo or control treatments.Reference Dumoulin and Hay-Smith29 The reviewers concluded that PFMT was better than no treatment, placebo drug, or inactive control treatments for women with stress, urge, or mixed incontinence, with no significant adverse effects and supported the widespread implementation of PFMT as first-line management of stress, urge and mixed forms of UI. There is no unanimous regime for a PFMT programme, but current guidelines suggest that programmes should consist of at least eight contractions performed three times a day, and that such a trial of at least 3 months worth of supervised PFMT be the first-line treatment in women with stress or mixed UI.30 A Cochrane review of the conservative management of post-prostatectomy UI did not provide definitive conclusion as to the benefits or harms of PFMT.Reference Hunter, Moore and Glazener31
Biofeedback. Biofeedback is a technique used to promote awareness of the physiological action of pelvic floor muscles, which works by helping a patient to identify and contract their pelvic floor muscles, thus enhancing PFMT. Patients may have access to visual, auditory, or tactile feedback when electrodes placed vaginally, rectally or on the surface detect and amplify a pelvic floor contraction.Reference Abrams, Cardozo, Fall, Griffiths, Rosier, Ulmsten, van Kerrebroeck, Victor and Wein32
In one randomized controlled trial (RCT), biofeedback-assisted behavioural training was significantly more effective in reducing incontinence episodes (80% reduction), compared with oxybutynin or placebo.Reference Burgio, Locher, Goode, Hardin, McDowell, Dombrowski and Candib33 The study included women with urge or mixed incontinence up to the age of 92 years (mean 68 years) and ~75% received a single session of biofeedback, with the rest receiving further sessions. Biofeedback as part of a multi-component behavioural training programme for urge incontinence in older women did not appear to enhance efficacy over and above that achieved by behavioural training using other methods (verbal feedback/self help booklet).Reference Burgio, Goode, Locher, Umlauf, Roth, Richter, Varner and Lloyd34 It may be that biofeedback is more appropriate for a subgroup of people who are unable to identify their pelvic floor muscles.
Biofeedback and pelvic muscle exercises are also effective in the treatment of stress incontinence in older women, with a suggestion of continued improvement up to 6 months in more severe incontinence (>8 incontinence episodes/week).Reference Burns, Pranikoff, Nochajski, Hadley, Levy and Ory35 Benefits were maintained and improvement continued for at least 6 months post-intervention. A systematic review concluded that there was no evidence that PFM exercises with biofeedback were more effective than PFM exercises alone for the treatment of stress UI, and that there was no clear evidence of the efficacy of PFMT for prevention of stress UI.Reference Berghmans, Hendriks, Bo, Hay-Smith, de Bie and van Waalwijk van Doorn36
Current guidelines recommend that biofeedback should not be used as routine, and may be considered in women who cannot actively contract pelvic floor muscles in order to aid motivation and adherence to therapy.30
Electrical stimulation. Electrical stimulation as a method of treatment for UI is based on contraction of smooth and striated periurethral muscles by applying electrical stimulation through a surface, vaginal or rectal electrode. This is supposed to be of benefit in stress incontinence by increasing the urethral closing pressure, but it is unclear how it helps those with detrusor overactivity. There are various methods of treatment using different types of currents and intensities. Although various RCTs have looked at electrical stimulation as a treatment for UIReference Barroso, Ramos, Martins-Costa, Sanches and Muller37–Reference Yamanishi, Yasuda, Sakakibara, Hattori and Suda40 and compared it with PFMT (aloneReference Bo, Talseth and Holme41, Reference Wang, Wang and Chen42 and in combinationReference Lo, Naidu and Cao43), there is a lack of consistency through the trials and hence inconclusive evidence. NICE guidelines recommend that electrical stimulation should not be routinely used alone or in combination with PFMT for the treatment of UI, and may be considered in women who cannot actively contract pelvic floor muscles in order to aid motivation and adherence to treatment.30
Magnetic stimulation. Functional magnetic stimulation is a method that aims to stimulate the sacral nerve roots and pelvic floor muscles by using a concentrated electromagnetic field, usually applied via the seat of a chair, without the need for internal probes. Though the exact mechanism of action is unclear, it has been suggested that the electromagnetic field causes a contracture of the external urethral sphincter and may suppress detrusor over-activity.Reference Quek44
A significant improvement in symptoms (56% reduction compared with 26% with placebo) was reported in a RCT that compared continuous functional magnetic stimulation (Pulsegen device) with sham stimulation in women with UI (stress/urge/mixed) for a period of 8 weeks.Reference But45 Magnetic stimulation was better tolerated than electrical stimulation (with dropout rates of ~50% in studies). Another RCT of subjects with urge-predominant mixed UI also reported significant improvements in symptoms with magnetic stimulation.Reference But, Faganelj and Sostaric46 While there are significant symptomatic improvements, these are temporary and there is a lack of standardization of treatment protocols and patient selection. The side-effect profile needs to be established. Thus the role of magnetic stimulation for treatment of UI needs further clarification.
Behavioural therapies
Bladder training. Bladder training or retraining is a widely used conservative method of treating urge urinary incontinence, and it is also thought to be of some benefit in people with mixed or stress incontinence.Reference Wallace, Roe, Williams and Palmer47 The aim of bladder training is to regain continence by increasing periodically the time interval between voids and suppressing urge in between voids. This usually involves hourly voiding regimens with gradual increments of the inter-voiding interval by 15–30 min, as tolerated, with the aim of reaching an interval of 3–4 h between voids without incontinence. One hypothesis about the mechanism of bladder training is that the bladder capacity increases with increasing time between voids, leading to a reduction in urinary incontinence.Reference Brown, McGhan and Chokroverty2 Patient education, scheduled voiding and positive reinforcement are all essential components of a bladder training programme, which can require up to several months to be successful. The Cochrane review (2009 update) of 12 trials reported inconclusive evidence to assess the effects of bladder training, citing inadequate evidence in the form of small studies, and further research was recommended.Reference Wallace, Roe, Williams and Palmer47 Current guidelines include a bladder training programme lasting for a minimum of 6 weeks as a first-line treatment for women with urge or mixed UI, and, if there is some benefit without independent continence, then the combination of an antimuscarinic agent with continued bladder training should be considered.30
Toileting schedules. Assisted toileting schedules for the purposes of this review will be considered in the categories of timed voiding, prompted voiding and habit retraining. These differ from bladder retraining in that they are primarily used for people with cognitive impairment and physical limitations,Reference Roe, Ostaszkiewicz, Milne and Wallace48 while bladder retraining is suitable for cognitively intact and physically able people with a defined type of incontinence.
Timed voiding. Timed voiding is a passive process of toileting assistance by another person, whereby people are taken for toileting at regulated intervals.Reference Ostaszkiewicz, Johnston and Roe49 A fixed toileting schedule is set, over the whole of the 24-h period, aimed at bladder emptying before an unstable detrusor contraction. While the non-invasive nature and lack of side-effects is attractive, the process can be demanding on the time of the carers delivering the management.Reference Ostaszkiewicz, Roe and Johnston50 A Cochrane review found only two studies largely due to inconsistency in the definition of the toileting schedule, with both trials reporting a larger reduction in incontinence, with intervention.Reference Ostaszkiewicz, Johnston and Roe49
Habit retraining. Habit retraining is a form of toileting assistance given by a caregiver to adults with urinary incontinence. An individualized toileting schedule is developed after identification of an incontinent person's natural voiding pattern in an effort to pre-empt involuntary bladder emptying.Reference Ostaszkiewicz, Chestney and Roe51 In view of potential ethical considerations with respect to regular checks of voiding for people with reduced cognition, an electronic monitoring device has been suggested for checking continence status.Reference Nikoletti, Young and King52 A Cochrane review found inadequate evidence to make definitive conclusions about benefits from habit retraining.Reference Ostaszkiewicz, Chestney and Roe51 Although there was a reduction in incontinence in the monitored groups, habit retraining was labour intensive and adherence rates by care staff may be poor as a consequence.
Prompted voiding. The aim of prompted voiding is for the person to initiate their own toileting through positive feedback and reinforcement.Reference Durrant and Snape53 At regular intervals, usually at 2-hourly intervals, the person is asked if they have had leakages of urine, a check of their continence status is made and feedback given to the person, with an offer to be taken to the toilet. Positive feedback is given if the person is not wet and if they request to use the toilet. Prompted voiding can have a degree of success whereby positive feedback leads onto more self-initiated requests to use the toilet.Reference Schnelle54 It is felt that it is not the learning of a new skill that causes the success but the increased opportunity to use the toilet plus positive reinforcement from staff. Irrespective of cognitive status and type of incontinence, longer term success was indicated by a reduction in percentage of checks where incontinence was discovered by 20% or a percentage toilet use at more than 66% of checks, after 3 days of prompted voiding, suggesting continuation of such therapy.Reference Ouslander, Schnelle, Uman, Fingold, Nigam, Tuico and Bates-Jensen55 This is seen as ensuring the effectiveness of the programme by targeting the appropriate person for this type of intervention. The Cochrane reviewReference Eustice, Roe and Paterson56 opined that those people with higher cognitive ability, higher dependency levels and normal cystometry were more likely to benefit and suggested that improved mobility associated with regular prompts may have improved the continence status. Prompted voiding appears beneficial for elderly people with incontinence with reliance on motivated carers. However, further research as to the long-term benefits and cost implications remains to be carried out.
Containment
Hand-held urinals. Hand-held urinals can be used for people who are unable to access either toilet or commode easily, or in time. This may occur in those with mobility problems in the absence of bladder dysfunction and may be compounded by an overactive bladder or overzealous diuretic use. For those who can transfer to a commode, both hand-held urinals and incontinence pads are seen as containment options. There are many different types of urinals for both men and women, for use in a variety of positions, and the male urinal design has not changed a great deal over the years. Recent studies and evaluations of male and female urinals to aid selection are limited, mainly due to the varied range available.
It is recognized that female urinals are not readily used in managing incontinence, with incontinence pads being preferred.Reference Pomfret, Vickerman and Tonge57, Reference Vickerman58 One evaluation of the use of female urinals suggested that hand-held urinals are useful in maintaining independent continence for those people with lower dependency levels; and assistance would be needed for those with higher dependency.Reference Fader, Pettersson, Dean, Brooks and Cottenden59 Successful use of a female urinal depended on the user's ability to position the urinal easily and to be confident that spillage would not occur. Issues around emptying of urinals have been reduced with the introduction of drainage bags that can be attached to the urinal to aid both collection and emptying processes.
Since independent use of the urinal requires a certain degree of dexterity, a functional assessment is an essential pre-requisite to the use of urinals,Reference Vickerman60, Reference Macaulay, Clarke-O'Neill, Cottenden, Fader, Van Den Heuvel and Jowitt61 with consideration given to urine flow, ability to get out of bed at night and patient preference.Reference Colley62 It is clear that further research is required on urinal design and evaluations to aid selection of appropriate candidate and urinal type. In the interim, availability of a wide range of urinals for a trial period is probably appropriate.
Multicomponent behavioural training
A sensible combination of the available NP interventions should be presented to the person with urinary incontinence. There is some evidence that the effects of individual interventions may be additive. In addition, simple educational information and urgency control techniques can be used to improve symptom control and potentially reduce incontinence.
Conclusions
Urinary incontinence is a common problem, which is on the rise due to an ageing population. There is significant associated morbidity and societal cost, highlighting its importance as an important public health issue.
A trial of NP interventions is the preferred first-line approach for management of incontinence for all sufferers. Lifestyle interventions can be applied universally, including in those with cognitive problems with carer partnership. While physical therapies are applicable to younger people with incontinence and the ‘otherwise well’ elderly, they have limited applicability in the presence of physical disability and cognitive deficit. Here behavioural therapy especially toileting schedules have had some success, though they are labour and time intensive. Containment with hand-held urinals can be used alongside trials of other therapies, and as a last resort where all interventions at alleviating incontinence have failed. Further research is required to clarify the feasibility, acceptability and efficacy of many NP interventions.
Conflicts of interest
The authors declared no conflicts of interest.
