Introduction:The pad test is an assessment tool for urinary incontinence (UI) severity classification and therapeutic response monitoring. However, the reliability and reproducibility of this test have been questioned.Objectives: To summarize the evidence regarding the accuracy measures and reproducibility of different pad test protocols for assessing UI. Methods: A systematic review of the diagnostic accuracy of this tool was performed (CRD42020219392). Eligibility criteria: Studies reporting data on the accuracy measures and reproducibility of the pad test when used for detecting UI in adult men and women. Data sources: MEDLINE, Science Direct, Cochrane, Web of Science, LILACS, and Pedro. Data extraction and synthesis: Two reviewers independently screened the articles, extracted the data, and evaluated the risk of bias (RoB) using the QUADAS-2 tool. Results: From 1048 studies, 18 studies were included. Eight of these reported accuracy data, and 12 reported reproducibility properties. A total of 1070 individuals were analyzed, whose mean age ranged from 20 to 90 years. The accuracy of the long-duration protocols was generally moderate to high (sensitivity, 60%-93%; specificity, 60%-84%). The 1-h protocols obtained higher accuracy values. The overall reproducibility was moderate to high (κ ≥ 0.66).
Limitations:The RoB was high and, due to different cutoff points adopted by studies, the bivariate model was not satisfied to perform a meta-analysis. Discussion: The 1-h pad test was more accurate but less reproducible when compared to the long-duration tests. Pad test results should be used with caution in clinical practice. K E Y W O R D S data accuracy, exposure, measures of association, reproducibility of results, risk or outcome, severity of illness index