Context: Inconsistent outcome reporting of effectiveness outcomes in surgical trials of stress urinary incontinence (SUI) has hindered direct comparisons of various surgical treatments for SUI.Objective: To systematically review the verbatim outcome names; outcome definitions and tools used to measure the outcomes in surgical trials of SUI in women. Evidence acquisition: Trials of women with SUI who have undergone surgical interventions were included. We conducted a systematic review (SR) on outcomes reported in RCTs of surgical management published in 2014-2019, covering the following database: MEDLINE, EMBASE, CENTRAL and CDSR. Verbatim outcome names extracted from the included studies were categorized, and then grouped into domains using the Williamson-Clarke (W/C) outcome taxonomy. A matrix was also created to visualize and quantify the dimensions of outcome reporting heterogeneity in SUI trials.Evidence synthesis: A total of 844 verbatim outcome names was extracted, of which, 514 variated terms were reduced to 71 standardized outcome names. They were further categorized into 11 domains from the W/C taxonomy. There were 7.24 different terms on average to describe each outcome, and the four outcomes with the most heterogeneity evident in terms used to describe them were "Urinary retention", "Reoperation", "Subjective cure rate" and "Quality of Life". Each of them had 20 or more different terms. Only 28% of the outcome definitions were reported, and a variety of measuring tools was noted, particularly in subjective outcomes. High heterogeneity was found in the outcome names; outcome definitions; choice and number of measuring instruments of the outcomes; choice and number of outcomes reported across studies.
Conclusions:This SR provides objective evidence of heterogeneity in outcome reporting in SUI surgical trials. Our categorization of outcomes highlights the difficulties in summarising the current evidence base. A core outcome set, developed using the methods advocated by the COMET and COSMIN initiatives is required.