A longitudinal study of urinary incontinence in community-based women: prevalence, incidence, resolution, and associated factors, and impact on well-being and quality of life

Abstract

Background Available data indicate that urinary incontinence (UI) is a condition in women, which adversely impacts on quality of life and daily activities. It affects women of all ages, but is particularly common in older women. It has been associated with significant physical morbidities, lowered well-being, loss of independence as well as sexual difficulties. It also causes a considerable financial burden on both individuals and the healthcare system. A detailed understanding of UI in terms of its prevalence, incidence and risk factors in women is an essential step in reducing the impact of this condition. However, the reported prevalence of UI among women varies widely between studies, with most studies reporting a prevalence of any UI in the range of 25 to 50%. This range is a result of the different definitions used for UI, the heterogeneity of study populations and the different data collection procedures applied for this sensitive health issue. UI is commonly viewed as a permanent condition once it develops. However, few studies have examined the progression and resolution of UI in community-based women with or without treatment. To date no study of the prevalence and incidence of UI in Australian women has been undertaken using a validated instrument. Also, little is known about the natural history of UI and its association with fecal incontinence. The aetiology of UI is widely recognised to be multifactorial and various risk factors have been identified in different studies. However, the estimated magnitude of risks varies widely in these studies and there is inconsistent evidence with regard to certain factors including mode of delivery, hysterectomy and hormone therapy use. Aims The aims of this doctoral thesis are to comprehensively examine UI in community-dwelling women in Australia in terms of its age-specific prevalence, risk factors, impact on quality of life, natural history (incidence and resolution) and its relationship with fecal incontinence (FI). Methods The project involved 542 community-based women aged 24 to 80 years in 2006 who were originally recruited from a previous cross-sectional study of 1423 women who participated in the Study of Androgens in Women (SAW). The SAW women were recruited from a database established from the Victorian Electoral Roll. Of the 1423 women who participated in the SAW, 754 agreed to be re-contacted regarding further research and of those, 542 women expressed interest in participating in the study of UI. A detailed self-administered questionnaire was mailed to the participants of this research at baseline in 2006 and again at follow-up in 2008. UI was assessed using a validated questionnaire, the Questionnaire for Urinary Incontinence Diagnosis (QUID) and FI by the Pelvic Floor Distress Inventory (PFDI). The PFDI was included in the follow-up study only. Definitions of stress and urge UI that conform to the standards recommended by the International Continence Society were used. The Bristol Female Lower Urinary Tract Symptoms Questionnaire (BFLUTS) was used to assess the impact of UI on condition-specific quality of life. The impact of UI on well-being was assessed using the Psychological General Well-being Index (PGWBI). Results Five hundred and six women provided data for the baseline analysis and 442 women for the follow-up data analysis. Prevalence: The overall point prevalence of any UI was 41.7% [95% confidence interval (CI): 37.2-45.8%] at baseline and increased to 44.6% [95%CI: 40.0-49.2%] by the end of the follow-up period. Of the 210 women reporting UI at baseline, 16% [95%CI: 12.9-19.3%] reported stress UI; 7.5% [95%CI: 5.2-9.8%] reported urge UI and 18% [95%CI: 14.7-21.5%] reported a mixed pattern. Stress UI was found to be the most common type among middle-aged women (25.3% of women aged 35-44 years), while urge UI was the most common type in women over the age of 75 years (24.2%). Risk factors: In logistic regression analyses, obesity (P