Quality and consistency of outcome reporting in clinical trials of immunosuppression in renal transplantation.

Abstract

Interpretation, comparison, and combination of results of clinical trials are reliant on accurate and complete reporting of outcomes. This study aimed to assess the quality and variability in outcome reporting in immunosuppression trials following renal transplantation. All randomized controlled trials comparing immunosuppressive interventions in renal transplant recipients published over a 5-year period were included. Outcomes reported in these studies were extracted, along with data regarding completeness of reporting and whether a clear definition of the method used to measure the outcome was provided. A total of 4760 outcomes were identified from 182 studies. Overall, 90.3% outcomes were completely reported; the remainder had missing data that would preclude use in meta-analysis; 31.5% manuscripts did not define a primary endpoint. Efficacy outcomes were more likely to be clearly defined than safety outcomes (OR: 0.022, P<.001) or patient-reported outcome measures (PROMs) (OR: 0.014, P<.001). PROMs were reported in less than half of manuscripts, and only five reported quality-of-life data using a validated tool. There was significant variability in the way that common efficacy and safety outcomes were defined. Variability in the way that endpoints are selected and reported in trials in renal transplantation makes interpretation and comparison between studies difficult.