The epidemiology of incontinence

Summarized by Robert W. Griffith, MD
December 15, 2000 (Reviewed: February 17, 2003)

Introduction

Knowledge of the epidemiology of urinary incontinence (UI) in women is not well founded. The available data are, in general, of relatively poor quality. A recent review article, summarized here, has provided an overview of our current state of knowledge, and discusses the adequacy of the supporting evidence. (The original review includes consideration of enuresis and UI in children, and UI in men, but these topics are not considered here.)

Epidemiology is the study of the distribution and causation of disease in people. The main difficulty in examining the epidemiology of urinary incontinence (UI) is the lack of a clear unifying definition. The international Continence Society defines it as "a condition where involuntary loss of urine is a social or hygienic problem and is objectively demonstrable".

Prevalence is defined as the probability of being incontinent within a defined population at a defined point of time, whereas incidence is the probability of developing incontinence over a defined time period (1, 2, or 5 years). However, the variability introduced by different time periods or points in time used in different studies makes it necessary to adapt a more pragmatic approach, defining incontinence based on symptoms alone. In addition, severity can be defined by factors such as frequency, amount and the subjective nuisance caused by the condition.

Prevalence

The prevalence of female incontinence to at least some degree, based on data drawn from 13 studies of the general population, shows 20 to 30% prevalence during young adult life, with a broad peak around middle age (30 to 40%) and then a steady increase in the elderly (30% to 50%). Each of these studies covered an age span of at least 30 years.

Types

Surveys based on questionnaires only register symptoms, whereas exact diagnoses require urodynamic studies. In the 11 studies reviewed by the authors, 48% of all incontinent women were classified as having stress incontinence, although this percentage was lower among older women. Urge incontinence, which was reported in 17%, seemed to increase with age, while mixed incontinence (34%) was slightly less frequent in the elderly. However, the proportion of stress and mixed types among older women were difficult to estimate, and there was considerable variability among studies.

One study attempted to validate the results of questionnaires against the final diagnoses made by a gynecologist following urodynamic examination. With this validation, the percentage of stress incontinence increased from 51% to 77%, mixed incontinence was reduced from 39% to 11%, and urge incontinence increased from 10% to 12%. This suggests that mixed incontinence is over-reported in most epidemiological surveys.

Severity

While the definition of severe or significant incontinence often varies among investigators, (e.g. frequency and amount of leakage, soaking of clothes, use of pads) prevalence of severe incontinence is quite consistent across different studies. Severe UI estimates range between 3% and 17% of total UI, but most studies report between 4% and 8%. Some studies find the prevalence of severe UI increases with age, while others show the opposite. If a "nuisance" or quality-of-life factor is included, approximately 20% of incontinent women overall (i.e. about 6% of all adult women) may be potential patients in need of therapy.

Natural remission

As many as one third of incontinent women report that they become continent again over time. In older people, one study showed a remission rate of 12% over 1 year. It must be emphasized, however, that longitudinal studies to support these findings are scarce.

Racial or ethnic differences

There is some evidence that white women are more susceptible to UI than black women. However, the data are somewhat contradictory, and interpretation is hampered by small sample sizes.

Risk factors

Again, well-controlled analyses of predictors for UI are quite limited. The following summary of health-related factors is based on cross-sectional studies and can only identify correlates.

Age.The prevalence of UI increased progressively with age, from 12% to 25%, in a large study of women aged 46 to 86 years. However, incontinence is not necessarily a normal aging event. Changes in the bladder and pelvic structures that occur with age can contribute to it, as can medical problems that are more common in older persons (e.g. diabetes).

Pregnancy.UI is more common in pregnant women, with prevalence rates of 30% to 60%. For most women, UI during pregnancy is self-limiting, although older women will often associate the onset of their incontinence with a pregnancy.

Childbirth.There are several reasons why childbirth predisposes women to UI. The process may result in a lax pelvic floor due to weakening and stretching of the muscles during delivery. Damage may occur as a result of spontaneous lacerations or episiotomy. Finally, stretching of the pelvic tissues can damage the pudendal and pelvic nerves, as well as the muscles and connective tissue of the pelvic floor, preventing the urethral sphincter from contracting promptly when called upon.

Although parity has been reported by some as a risk for stress incontinence, the effect is not strong, and several studies show no, or only insignificant, increase in UI with increasing parity. Obstetric trauma, causing a vesico-vaginal fistula for example, is an important cause of UI in developing countries.

Menopause.The atrophic changes in genital tissues associated with the menopause increase the susceptibility to urinary tract infections; reversal of these changes by estrogen suggests that estrogen loss contributes to the risk of UI. However, the literature on this is somewhat inconsistent.

Hysterectomy.The relevance of hysterectomy as a risk factor for UI is controversial. While a hysterectomy with oophorectomy induces a surgical menopause, pure hysterectomy may itself cause nerve damage to the pelvic and/or pudendal nerves and may disturb muscular faschia or attachments of the bladder to the pelvic wall. Women who have had a hysterectomy have reported urinary incontinence as more likely (see the article "Incontinence after hysterectomy" - link below).

Obesity.The added weight of obese tissues can bear down on pelvic tissues causing chronic strain, stretching and weakening of the pelvic tissues. Intervention studies leading to weight and BMI reduction have shown decreases in UI severity.

Urinary tract infection.This condition shows one of the clearest associations with UI, and it is one that can be usually readily corrected.

Functional impairment.Mobility problems (e.g. after a fall, in arthritis, and when there are other physical difficulties) can produce dependent continence, in which the patient is dry, but only as a result of being regularly reminded or shown to the toilet, or being given assistance. It often remains to be determined whether UI is a direct consequence of difficulties in getting to the toilet and removing clothing, or whether mobility limitations and UI are both consequences of the general frailty of old age or an underlying condition.

Cognitive impairment.UI is common in patients with dementia, in both men and women; in fact, this is the only risk factor that affects males more than females.

Other factors.Constipation, the use of diuretics and other drugs, uterine prolapse, pelvic radiation, and cigarette smoking have all been accused of being related to an increased likelihood of UI. Definitive studies are, however, lacking.

Comment

The variability in the nature of the information reviewed by the authors shows the need for large longitudinal studies using a rigorous protocol, in order to bring our knowledge of UI to a satisfactory level. Low response rates to questionnaires can bias prevalence estimates, and method of correcting for this have to be established. Some communities or ethnic groups fail to have strong health-seeking behavior, so that many cases of UI may be missed. These influences can also be affected by differing health-care systems in different countries.

The authors of the review suggest that all future epidemiological studies of UI contain a minimum data set, incorporating the following elements:

• A screening question for any involuntary urine loss
• A frequency measure (none, less than once a month, etc.)
• A quantity of urine loss for a typical episode (drops, small amounts, etc.)
• Duration (months, years)
• A validated measure of "nuisance" or use of a "quality-of-life" scale.

If future studies are done using these evaluation elements, we shall learn considerably more about the risk factors for urinary incontinence, and be in a better position to advise women on how to avoid it.

Source

• Epidemiology and natural history of urinary incontinence. S. Hunskaar, EP. Arnold, K. Burgio, Int Urogynecol J. , 2000, vol. 11, pp. 301--319

Related Links
Preventing incontinence
Incontinence after hysterectomy

Please take a moment to give us your comments. For questions about Health matters you may check our "Questions & Answers" Portal and Service.