Urinary incontinence—older people: where are we now?


Dr A. Wagg, University College London Hospitals, London, UK.


For many, old age is accompanied by the burden of chronic disease. Although life expectancy for women is approximately 5.4 more years than men,1 there is evidence that a greater proportion of these extra years may be spent with chronic disability.2 The rate of institutionalisation at the end of life is also increasing. For women (and men) of 65 years and older, the lifetime chance of requiring nursing home care is 43%.3 This is highly relevant, for not only do such nursing home residents experience the highest rates of incontinence in the population (Fig. 1), but there are data to suggest that the presence of incontinence of urine alone renders institutionalisation significantly more likely.4 Incontinence lies at the heart of many disease complexes (Table 1) commonly affecting older people and thus the number of people defined as sufferers is high, estimated at up to 3.9 million in the UK. Population projections for the UK demonstrate a large increase in the numbers of older women, with the greatest increase being in the 60–69 years age group. The number of women between the ages of 80–99 years is also expected to double from 2002 to 2008 with a consequently huge potential for demand on healthcare services. There is also evidence from hospital episode statistics that despite the increased prevalence in older women, there are fewer surgical procedures performed for the condition than in younger age groups.5

Figure 1.

Prevalence of urinary incontinence in institutional care and community dwelling individuals in Leicstershire (1997 data).

Table 1.  Drug therapy which may potentially aggravate or predispose to urinary incontinence.
DiureticsIncrease urinary frequency and may precipitate urge incontinence in predisposed individuals
Calcium channel antagonistsAssociated with polyuria, especially at night when fluid redistribution occurs
Anticholinergics (including antihistamines, antipsychotics, antispasmodics, antiparkinsonian agents)May precipitate confusion, especially in those with pre-existing cognitive impairment
Alpha adrenoreceptor antagonistsMay predispose to stress incontinence, due to relaxant effect on the external urethral sphincter
Non Steroidal Anti-Inflammatory DrugsSalt and water retention
H2 antagonistsConfusion
ACE inhibitorsChronic cough
Selective Serotonin Reuptake InhibitorsAssociated with new onset urinary incontinence
Benzodiazepines and NeurolepticsAny sedative medication with an appreciable hangover effect will exacerbate continence problems in those elderly people predisposed to problems


The publication of Good Practice in Continence Services6 was originally intended to mandate change. Unfortunately, the publication was downgraded to guideline status shortly before publication with the result that change in response to its recommendations has been slow and patchy.7 The National Service Framework (NSF) for Older People, however, has mandated that integrated continence services for older people (and now, by inference, for all) be in place by April 2004.8 A national survey of British Geriatric Society members (Bladder & Bowel Special interest group, 2003, unpublished data) revealed increased interest and involvement in planning for this NSF milestone. In addition, the requirement for stroke and falls services also has an impact on continence service provision, although data from the National Sentinel Audit for Stroke suggests that continence provision for these patients is still patchy, some 4 years after the initiation of the audit.9 The Royal College of Physicians of London, via its clinical Effectiveness and Evaluation Unit, has also embarked upon the pilot stage of what will hopefully be a National Audit of Continence Care of Older People, to highlight variations in standards of provision and to act as a catalyst for change.


In 1995, data from intensive testing of urinary tract function on normal, asymptomatic elderly people, half of them female, but without comparative controls were published, suggesting that normality was a rarity, with only 18% of individuals falling into this category.10 The problem with such research is that many ‘age-related’ studies have not used comparative samples of younger individuals and therefore it is difficult to ascribe their findings to age alone. A further limitation is that much of the data relating to changes in lower urinary tract function are derived from studies of individuals with lower urinary tract symptoms who have undergone urodynamic studies. Data from community dwelling, continent individuals are sparse, but where they do exist tend to confirm the associations identified by other means.

Detrusor contractile function, bladder capacity and urinary flow rates all appear to decline in association with greater age.11 There is also an increase in the prevalence of incomplete emptying as demonstrated by the existence of a significant post-micturition residual volume of urine.12 The changes in the bladder associated with prostatic hypertrophy in the male; significant detrusor muscle hypertrophy; a reduction in the number of acetylcholinesterase-containing nerves and consequent release of neurotransmitter in response to stimulation and an increase in the collagen to smooth muscle ratio which causes the bladder to be less compliant, have all been demonstrated in association with age and have been observed in age-matched females.3–18 There is also a reduction in the sensory function of the bladder, first sensation to void being that much closer to functional bladder capacity in older people, giving a perhaps functionally impaired older individual less time to reach the lavatory and void appropriately.19 Bladder capacity falls, whether measured by cystometrogram or frequency–volume chart, and urinary frequency rises.20

Studies using the urethral pressure profile21,22 or detrusor closure pressures23 have both described a reduction in urethral function in later life. There is evidence of an increased collagen content in the urethra of elderly females.24 There is also a loss of striated muscle cells in the aged female urethral sphincter.25

Likewise, the evolution of detrusor overactivity is conventionally thought to be associated with the development of significant outflow tract obstruction. This viewpoint has been reinforced by the fact that relief of the obstruction leads to bladder stability in a significant proportion.26 However, once again, the incidence of detrusor overactivity increases in association with age per se and is similar in women.


There are several physiological changes occurring in late life, which lead to an increasing likelihood of developing nocturnal frequency. Evidence suggests that night-time voiding is one of the most bothersome of lower urinary tract symptoms.27 Normally, adults produce two-thirds of their 24-hour urine output by day and the other third by night. In older individuals this changes, renal concentrating ability falls and glomerular filtration rate increases in the supine position. There is a redistribution of fluid at night, particularly if the individual has venous insufficiency or is on medication, which predisposes to the development of peripheral oedema. In addition, some older adults have a delayed diuresis in response to a fluid load and lose their diurnal rhythm of ADH secretion.28 When taken together, this means that the kidneys are working harder overnight to produce greater quantities of more dilute urine, the amount of which may be in excess of functional bladder capacity (reduced in older women). All this is in the absence of any pathology, such as heart failure, diabetes or hypercalcaemia which may exacerbate the situation. Ten to 15 per cent of older women suffer from nocturia; some of this is undoubtedly due to excessive fluid intake in the form of fixed drinking habits. Causes of rising at night for reasons other than passing urine, such as insomnia, should not be neglected in the assessment. An association with sleep apnoea and sleep disturbance in the elderly has been described.29

There is evidence for the efficacy of DDAVP,30 early evening diuretic31 and limited evidence for daytime recumbence32 but these are not well tolerated by all. A recent observation on the use of aspirin, presumably acting via prostaglandins to reduce the glomerular filtration rate has been made.33 The results of a systematic attempt to examine this are awaited.

In particular, the usefulness of DDAVP appears to be limited to those individuals with true nocturnal polyuria, defined as producing >50% of total daily urine output at night,34 and its use may be hampered by drug–drug interactions predisposing to hyponatraemia or excessive secondary drinking habits. Primary nocturnal enuresis may persist into adulthood but the reported incidence of nocturnal enuresis in community-dwelling women is 3%, only a tiny proportion of these women will have primary nocturnal enuresis.35

Treatment with imipramine has shown a marked antidiuretic effect on patients with nocturnal polyuria, this appears to be a vasopressin-independent effect, mediated by α-receptors in the proximal convoluted tubule. Data on its effectiveness originates from use in younger patients. DDAVP is used although the risk of adverse events on those over 65 has resulted in a requirement for caution and early monitoring of serum sodium.36 For institutionalised, highly dependent patients, the problem of nocturnal incontinence continues to require a major effort to manage effectively. The continuing developments in pad technology and the use of barrier creams can often minimise the disruption of sleep patterns.


Many of the diseases leading to incontinence are associated with chronic impairment of function. Elderly women experience the same bladder problems as other adults but their ability to respond and to compensate for problems, which a younger adult may find trivial is often compromised. Concomitant disease, disability and drug therapy, in particular, may serve to render an elderly person incontinent (Table 1). Whilst faecal loading causes faecal incontinence, and may cause urinary retention, the relationship between loading and urinary incontinence is unclear, though often claimed. The influence of heart failure upon nocturnal frequency is well recognised. Physical problems around access to lavatory facilities and privacy given that access also need to be considered, particularly in institutional settings. The provision of commodes or other urine collection devices, appropriate grab rails and raised toilet seats may be essential in maintaining continence for some people. Strategies aimed at maintaining mobility of elderly individuals also have a positive effect upon urinary incontinence. The dexterity of elderly people should also be taken into account when devising a management strategy. Often the addition of velcro fastenings to clothes, rather than buttons or zip fasteners can make the difference between continence or incontinence.


Falls are common in the elderly and are the main cause of death due to trauma in the population over 65 years of age. In any preceding year up to 34% of over-65-year-olds will have fallen. Recurrence in the following year is 60%–70%. Data from a long-term study of osteoporosis risk suggest that patients with urge incontinence greater than weekly are at increased risk of falls (OR 1.26, 95%CI 1.1–1.4) and sustaining a fracture (OR 1.34, 95%CI 1.1–1.7) than those without.37 There was no association with either stress incontinence or nocturnal frequency; however, the presence of urinary incontinence was one of the four strongest predictors of falls in a large cohort of community dwelling elderly.38 Unfortunately, the relationship between falls and urinary urgency has not been recognised in this paper and there is no evidence from any intervention study that treating the incontinence reduces the incidence of falls.


Urinary incontinence is an indicator of poor prognosis in stroke sufferers and may simply be a marker of poor functional and cognitive status. However, its presence is not limited to highly impaired people and is associated with substantial morbidity. 30-day survivors of stroke are 3.9 times likely to die within 1 year and twice as likely to die within 2 years than continent stroke survivors.39,40 There is some evidence to suggest that patients who were incontinent prior to suffering a stroke might also be at risk of a worse outcome in terms of functional status at 6 months and be more likely to be discharged to institutional care.41–43 The presence or absence of urinary incontinence has also been used as a marker of good functional recovery and still is a stronger predictor than any of the more complex predictive scores used in stroke disease.44 The incidence of new onset incontinence following stroke has been estimated at between 2.5% and 17%, but there are little naturalistic data on the evolution of symptoms following an index stroke event.45,46


Many commonly used drugs may, through a variety of mechanisms, have an adverse effect upon the function of the lower urinary tract or the physical ability of an elderly person to cope with pre-existing urinary symptoms. Some of these are listed in Table 2. The elderly are more likely to be taking a number of different drugs; interactions between these and between any drug treatment for incontinence and pre-existing medication should be taken into account when prescribing for an elderly woman.

Table 2.  Concomitant diseases which may have an impact upon urinary incontinence.
Diseases affecting mobility
Arthritis hip fracture
Peripheral vascular disease
Parkinson's disease
Nervous system disorders affecting cognition and neural control mechanisms
Parkinson's disease
Other medical conditions
Diabetes mellitus; causing polyuria and autonomic neuropathy
Congestive heart failure; leading to excess nocturnal urinary production
Venous insufficiency; a similar mechanism
Chronic lung disease
Exacerbation of stress incontinence


Bladder problems in older women are common and older people are different, both in terms of underlying physiology and in the capacity of an older person to compensate for the increased burden. A different, wider looking approach to assessment and management is required.