Significant variation in outcome reporting in pelvic floor research trials has been confirmed. This results in significant research heterogeneity and limited comparability between primary studies, hindering meta-analysis and synthesis of data. This study aimed to systematically review outcome reporting in randomised controlled trials (RCTs) evaluating conservative interventions for stress urinary incontinence (SUI) in women to assess for variation and create an inventory of outcomes utilised. 
This is the first step in development of a core outcome set (COS), a consensus-based agreed minimum set of outcomes recommended to be measured and reported in all clinical trials. The quality of outcome reporting and the methodological quality of primary studies was also evaluated using the Jadad (1) and MOMENT (2) scoring systems respectively.

Using Medical Subject Headings (MeSH) terms, the following databases were searched for RCTs evaluating conservative treatments for female SUI: PubMed, Medline, EMBASE and the Cochrane Central Register of Controlled trials (CENTRAL). MeSH terms included: ‘stress urinary incontinence’, ‘conservative’, ‘non-surgical’, ‘pelvic floor muscle training’, ‘pharmacological’, ‘duloxetine’, ‘biofeedback’, ‘vaginal cones’, ‘lifestyle intervention’, ‘electrical stimulation’, ‘alternative treatment’, ‘randomised controlled trials’, ‘systematic review’, and ‘outcomes’. An inventory of reported outcomes and outcome measures from eligible RCTs was developed before being further categorised into outcome domains and themes according to a medical outcome taxonomy. Quality of primary studies was independently assessed by two researchers with any discrepancies resolved by discussion with a third researcher. Statistical correlations between study characteristics and quality parameter scores were also assessed using Spearman’s rho analysis.

Initial literature search yielded 711 RCTs. n total, 74 eligible studies were eventually included, involving 20,668 participants. 317 outcomes were reported using 134 outcome measures. These were classified into 4 main themes and subsequently 11 domains. The most frequently reported outcome domains were ‘SUI treatment success’ (90.5%), ‘patient reported outcomes’ (87.8%) and ‘lower urinary tract symptoms’ (74.3%).  Objective and subjective cure were the most reported outcomes in the ‘SUI treatment success’ domain; quality of life, experience of symptoms and overall improvement in the ‘patient-reported outcomes’ domain; and symptom severity and symptom frequency in the ‘lower urinary tract symptoms’ domain. The majority of studies reported the use of at least one validated questionnaire (86.5%), the most frequently used were Patient Global Impression of Improvement scale (PGI-I; 24.3%) and the Incontinence Quality of Life instrument (I-QOL; 23.0%). Several domains were underreported, with the least reported domains being cost-effectiveness (1.4%), treatment pathway (4.1%) and sexual dysfunction (5.4%). The median Jadad score was 3, with an interquartile range of 2-4.75, and the median MOMENT score was 5, with an interquartile range of 3-5. Statistically significant correlations identified were between Jadad and MOMENT scores, and between the type of journal (general, specialty or subspecialty) and the year of publication.

This systematic review demonstrates variations in reported outcomes and outcome measures utilised in studies evaluating conservative treatments for female SUI. The least reported outcome was cost-effectiveness, with only one study reporting on outcomes relating to the cost of treatment. There is currently no predefined, minimum set of outcomes for use in research evaluating conservative treatments for SUI. We propose the interim use of the most commonly reported outcomes within the three most reported domains: subjective and objective cure rates in the ‘SUI treatment success’ domain; quality of life, symptom experience and overall improvement in the ‘patient-reported outcomes’ domain; and symptom frequency and symptom severity in the ‘lower urinary tract symptoms’ domain (Table 1). Validated questionnaires should be used wherever possible, with PGI-I, I-QOL and International Consultation on Incontinence Questionnaire-Urinary Incontinence Short Form (ICIQ-UI SF) the most reported questionnaires identified. These can be regarded as guidance for a minimum set of outcomes for use in the interim before the development and dissemination of a COS. The positive correlation between MOMENT and Jadad scores validates these as effective scoring systems.

Quality parameters and outcome reporting varies significantly in RCTs assessing conservative treatments for SUI. The most reported outcomes, in the three most reported domains, are recommended for interim use in future trials on the conservative treatments of SUI until a core outcome set is developed and disseminated.

Jadad A, Moore A, Carroll D, Jenkinson C, Reynolds J, Gavaghan D, McQuay H. Assessing the quality of clinical trials: Is blinding necessary? Controlled Clinical Trials. 1996 Feb; 17(1):1-12.
Harman NL, Bruce IA, Callery P, Tierney S, Sharif MO, O'Brien K, Williamson PR. MOMENT - Management of Otitis Media with Effusion in Cleft Palate: protocol for a systematic review of the literature and identification of a core outcome set using a Delphi survey. Trials. 2013 Mar; 14(1):70.