The Cochrane risk of bias assessment tool 2 (RoB 2) versus the original RoB: A perspective on the pros and cons.

Background and Aims

Critical appraisal or risk of bias assessment is a fundamental part of systematic reviews that clarifies the degree to which included research articles are qualified and reliable. Version 2 of the Cochrane tool for assessing the risk of bias in randomized trials (RoB 2), the updated version of the first tool, was released in 2019. Here, we have compared these two versions of Cochrane risk of bias assessment tools and highlighted the pros and cons of RoB 2.

Methods

Statistical analysis and methodology is not applicable to this article as no new data were created or analyzed in this study.

Results

The overall approach in RoB 2 is that by answering some signaling questions after the specification of results, effects of interest, and sources of information, an overall judgment for the quality of each study is reached. Accordingly, in the original version of the Cochrane RoB tool, the judgment can be in three different conclusions, including low, unclear, and high risk of bias. The most prominent difference in bias domains is the removal of "other bias" domain being replaced by "overall bias" judgment. Also, the most common presentation types of Cochrane risk of bias assessments are the "summary" and "graph" which are generated by Review Manager, web-based applications, or packages in R software.

Conclusion

The RoB 2 tool, compared to the original RoB, has improved and is the recommended version by the Cochrane Collaboration for quality assessment of randomized controlled trials. It is recommended to consider funding source, duration of follow-up, declaration of data availability, the status of baseline characteristics between groups, and sample size calculation methods in further revisions of the Cochrane risk of bias assessment tools.

Informations de copyright

© 2024 The Author(s). Health Science Reports published by Wiley Periodicals LLC.

Déclaration de conflit d'intérêts

The authors declare no conflict of interest.