Variation in estimates of urinary incontinence prevalence in the community: effects of differences in definition, population characteristics, and study type.

OBJECTIVES: Prevalence estimates for urinary incontinence among community-dwelling adults vary from 2 to 55%. A review of the literature was undertaken to investigate the degree to which differences in definitions of incontinence, age, and gender of the populations studied, response rates, measurement techniques, or location could explain differences in reported prevalences.
DESIGN: A literature search was conducted to locate all studies published in English reporting the prevalence of urinary incontinence in a population-based sample of adults. MEASUREMENT: Information was abstracted for study size, response rate, type of survey, definition of urinary incontinence, and prevalence of incontinence by age group and gender. Prevalence by type of incontinence was also abstracted where available. Stratification was used to obtain prevalence estimates specific for age, gender, and frequency of incontinence. Data were examined for associations between prevalence and survey type, response rate, year, and location of survey.
RESULTS: A total of 21 studies met inclusion criteria. Stratification of reported prevalence by frequency, gender, and age substantially reduced the variation in prevalence estimates. For older women, the estimated prevalence of urinary incontinence ranged from 17 to 55% (median = 35%, pooled mean = 34%), and for daily incontinence it ranged from 3 to 17% (median = 14%, pooled mean = 12%). For older men, incontinence prevalence was estimated to be 11 to 34% (median = 17%, pooled mean = 22%), and 2 to 11% reported daily incontinence (median = 4%, pooled mean = 5%). Within studies, the prevalence of any incontinence was 1.3 to 2.0 times greater for older women than for older men. Among middle-aged and younger adults, prevalence of incontinence ranged from 12 to 42% (median = 28%, pooled mean = 25%) for women and from 3 to 5% (median = 4%, pooled mean = 5%) for men. The ratio of prevalence of any incontinence for women to men in this age group ranged from 4.1 to 4.5. Stress incontinence predominated in younger women, whereas urge and mixed incontinence predominated in older women. There was a tendency for studies using in-person interviews to report higher prevalences.
CONCLUSIONS: An accurate estimate of the prevalence of urinary incontinence depends on specifying the definition of incontinence and the age and gender groups of interest.