Screening for Urinary Incontinence in Women: A systematic review for the Women's Preventive Services Initiative

Background: Urinary incontinence is infrequently addressed during routine health care despite its high prevalence and adverse effects on health. Purpose: To evaluate whether screening for urinary incontinence in women not previously diagnosed improves outcomes (symptoms, quality of life, and function) and to evaluate the accuracy of screening methods and potential harms of screening. Data Sources: English-language searches of Ovid MEDLINE, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews (1 January 1996 to 30 March 2018); ClinicalTrials.gov (April 2018); and reference lists of studies and reviews. Study Selection: Randomized trials, cohort studies, systematic reviews of studies that enrolled nonpregnant women without previously diagnosed urinary incontinence and compared clinical outcomes and adverse effects between women who were and were not screened, and diagnostic accuracy studies that reported performance measures of screening tests. Data Extraction: Dual extraction and quality assessment of individual studies. Data Synthesis: No studies evaluated the overall effectiveness or harms of screening. Seventeen studies evaluated the diagnostic accuracy of 18 screening questionnaires against a clinical diagnosis or results of diagnostic tests. Of these, 14 poor-quality studies were based in referral clinics, enrolled only symptomatic women, or had other limitations. One good-quality and 2 fair-quality studies (evaluating 4 methods) enrolled women not recruited on the basis of symptoms. Areas under the receiver-operating characteristic curve for stress, urge, and any type of incontinence in these studies were 0.79, 0.88, and 0.88 for the Michigan Incontinence Symptom Index; 0.85, 0.83, and 0.87 for the Bladder Control Self-Assessment Questionnaire; and 0.68, 0.82, and 0.75 for the Overactive Bladder Awareness Tool. The Incontinence Screening Questionnaire had a sensitivity of 66% and specificity of 80% for any type of incontinence. Limitation: Studies enrolled few participants, often from symptomatic referral populations; used various reference standards; and infrequently reported CIs. Conclusion: Evidence is insufficient on the overall effectiveness and harms of screening for urinary incontinence in women. Limited evidence in general populations suggests fairly high accuracy for some screening methods.