Abstract 16310: Adherence of Randomized Controlled Trials in Heart Failure to Consort Reporting Standards

Abstract

Introduction: Randomized control trials (RCTs) have changed care in Heart Failure (HF), but the reliability of results is contingent on trial validity and transparent reporting. The Consolidated Standards of Reporting Trials (CONSORT) checklist was developed in 1996 and subsequently updated to guide the design and reporting of RCTs, but the adherence of HF RCTs to CONSORT remains poorly understood. This study sought to evaluate the adherence of HF RCTs to the CONSORT 2010 update and to examine temporal trends in trial reporting in HF trials. Methods: We searched MEDLINE, EMBASE and CINAHL for HF RCTs in high-impact journals between 2000 and 2020. The primary outcome measure was RCT CONSORT statement or CONSORT extension reporting presented as mean CONSORT score. Descriptive statistics were calculated for CONSORT scores. We used the Jonckheere-Terpstra test to examine temporal trends and employed multivariable linear regression to evaluate the association between prespecified trial characteristics and the mean CONSORT score. Results: There were 221 RCTs, and mean CONSORT score was 69.7% (SD 11.5). Adherence to CONSORT reporting standards improved from 61.5% (SD 11.1) in 2000-2003 to 76.5% (SD 8.8) in 2016-2020 ( P <.001). Adherence was greater in two-group parallel individual-level RCTs (β =6.22 95% CI: 3.14, 9.30; P <.001) than in other trial types. Drug (β= -3.91, 95% CI: -7.32, -0.50; P <.001) and device/surgical (β= -2.82 (95% CI: -7.56, 1.93; P <.001) interventions were associated with lower adherence to CONSORT reporting standards than other interventions, while publication after 2010 was associated with greater adherence (β=10.96, 95% CI: 8.37, 13.55; P <.001). Funding source did not appear to have an association with adherence (β=1.51, 95% CI: -1.53, 4.55; P =.33). Conclusions: Among HF RCTs published in high-impact journals, adherence to CONSORT reporting standards has improved over time, but remains variable and suboptimal overall.