The Rational Clinical Examination
March 26, 2008

What Type of Urinary Incontinence Does This Woman Have?

Author Affiliations

Author Affiliations: Divisions of Geriatrics and General Internal Medicine, and Knowledge Translation Program, University of Calgary, Calgary, Alberta, Canada (Drs Holroyd-Leduc and Straus); Institut Universitaire de Geriatrie de Montreal, Montreal, Quebec, Canada (Dr Tannenbaum); Knowledge Translation Program, University of Toronto/St Michael's Hospital, Toronto, Ontario, Canada (Dr Straus and Mr Thorpe); and Department of Public Health Sciences, University of Toronto, Toronto, Ontario (Mr Thorpe).

JAMA. 2008;299(12):1446-1456. doi:10.1001/jama.299.12.1446

Context Urinary incontinence is a prevalent condition and treatment options can depend on what type of incontinence is present.

Objective To systematically review the evidence about the most accurate way to determine the type of urinary incontinence during an office assessment.

Data Sources A search of MEDLINE using Ovid (1966-July 2007) and EMBASE (1980-July 2007), and the bibliographies of retrieved articles to identify relevant studies. Search terms included urinary incontinence, diagnostic tests, medical history taking, physical examination, cough stress test, and urodynamics.

Study Selection English-language articles were identified that addressed the office diagnosis of urinary incontinence in adults, in which data was not limited to case reports. Cohort studies of patients undergoing history, physical examination, and/or office procedures (excluding urodynamics) for diagnosing the type of urinary incontinence were included. Case-control studies were considered when there was insufficient data available from cohort studies. The accepted reference standard for categorization of incontinence type was diagnosis confirmed by an expert, urodynamic studies, or both.

Data Extraction Two investigators independently appraised study quality and extracted relevant data. Minimum inclusion criteria were completion of an appropriate reference standard in all patients and the ability to extract relevant data.

Data Synthesis Forty articles were identified for inclusion. A random-effects model was used for quantitative synthesis. Minimal data was available for men. In women, simple questions modestly helped diagnose stress urinary incontinence (summary positive likelihood ratio [LR], 2.2; 95% confidence interval [CI], 1.6-3.2; summary negative LR, 0.39; 95% CI, 0.25-0.61) but are more helpful in diagnosing urge urinary incontinence (summary positive LR, 4.2; 95% CI, 2.3-7.6; summary negative LR, 0.48; 95% CI, 0.36-0.62). A positive bladder stress test may help diagnose stress urinary incontinence (summary LR, 3.1; 95% CI, 1.7-5.5); however, a negative test is not as useful (summary LR, 0.36; 95% CI, 0.21-0.60). A systematic assessment combining the history, physical examination, and results of bedside tests to establish a clinical diagnosis appears to be of modest value in diagnosing stress urinary incontinence (summary positive LR, 3.7; 95% CI, 2.6-5.2; summary negative LR, 0.20; 95% CI, 0.08-0.51). The systematic assessment is less helpful in diagnosing urge urinary incontinence (summary positive LR, 2.2; 95% CI, 0.55-8.7; summary negative LR, 0.63; 95% CI, 0.34-1.17).

Conclusions The most helpful component for diagnosing urge urinary incontinence is a history of urine loss associated with urgency. A bladder stress test may be helpful for diagnosing stress urinary incontinence.