A quantitative assessment of the quality of randomized controlled trial reporting in the urologic literature

ID: 

106

Session: 

Poster session 3

Date: 

Tuesday 25 October 2016 - 10:30 to 11:00

Location: 

All authors in correct order:

Narayan V1, Cone E2, Smith D1, Scales C2, Dahm P1
1 Minneapolis VAMC and University of Minnesota, Department of Urology, USA
2 Duke University, Duke Clinical Research Institute and Division of Urology, USA
Presenting author and contact person

Presenting author:

Vikram Narayan

Contact person:

Abstract text
Background: Randomized controlled trials (RCTs) are the gold standard format for clinical research, and clear reporting is vital to informing evidence-based practice. In 1996 the Consolidated Standards of Reporting Trials (CONSORT) statement was published to provide guidelines and standardization when reporting clinical trials.

Objectives: We systematically assessed the quality of RCT reporting in 2013 as compared to RCTs in 2004 and 1996. Our objective was to quantify any improvement in the intervening time period.

Methods: All RCTs published in four leading urology journals in 2013 were identified for formal review, and compared to a prior analysis of studies from 1996 and 2004 using the same inclusion criteria. Two reviewers abstracted data using a standardized evaluation form based on the CONSORT checklist. We calculated a summary reporting score (range 0 to 22) for each study and compared mean summary scores for 1996, 2004, and 2013. We settled disagreements by consensus and a third party referee. Chi-squared, Student’s T test, and ANOVA were used to analyze the results.

Results: A total of 82 RCTs published in 2013 met inclusion criteria and were compared to 65 and 87 studies from 1996 and 2004, respectively. The mean (± SD) CONSORT summary scores were significantly different between years, 15.6 (± 2.0) in 2013, 12.0 (± 2.5) in 2004, and 10.2 (± 2.3) in 1996 (P < 0.01). Provision of a flow diagram improved from 3.1% (1996) to 19.5% (2004) to 87.8% (2013; P < 0.001). Overall, reporting of important methodological criteria varied within journals, but improved substantially overall from 1996 to 2004 and from 2004 to 2013, with reporting of many key methodologic criteria appearing in more than 50% of RCTs for the first time in 2013. However, many items continue to be underreported, including blinding of study participants and team member roles.

Conclusions: The results of this systematic review suggest that RCT reporting in the urological literature has improved since the publication of the CONSORT statement, although many key methodologic criteria remain underreported. Further efforts are needed to continue to improve the urological literature.

Attachments: