Adequacy of trial registration and consistency in outcome reporting in rheumatology RCTs: A meta-research study

Introduction Transparent reporting of randomized controlled trials (RCTs) is essential to ensure research integrity. While registration practices of RCTs are improving, concerns remain regarding the adequacy of outcome registration and the consistency of outcome reporting in publications. Objectives To evaluate the adequacy of primary outcome registration and the consistency of outcome reporting between registry entries and publications in rheumatology RCTs. Methods We conducted a meta-research study including primary reports of RCTs in rheumatology published from 2009 to 2022. We assessed whether primary outcomes were adequately registered (presence of SPIRIT-based criteria: measurement, analysis metric, and time points). For adequately registered outcomes, we evaluated consistency between the registry and the published report. Logistic regression was used to explore factors associated with inadequate registration and inconsistent reporting. Results We analyzed 947 RCTs involving 1679 primary outcomes. Only 38% of trials adequately described all their primary outcomes in the registry. Of these adequately registered trials, 67% reported their primary outcomes consistently in the corresponding publication, meaning just 25% of trials met both criteria. The most common gap in outcome description was missing participant-level analysis metric, while the most prevalent inconsistency was related to changes in the timing of outcome assessment. Registration of adequately described primary outcome improved between 2009 and 2013 before plateauing, while consistency in reporting showed no improvement over time. Conclusion Despite improvements in trial registration practices, major gaps persist in outcome specification and reporting consistency in rheumatology RCTs.