Objectives: To introduce a proposed revision to the 'Risk of bias' tool for randomized trials and to obtain feedback.
Description: The Cochrane 'Risk of bias' tool for randomized trials was made available in 2008 and its use is now widespread in both Cochrane and non-Cochrane reviews. An evaluation of the tool led to minor modifications in 2011 to improve usability. An additional wish list of modifications was collated, including the need for an algorithm to help users reach domain-level judgments; clear guidance for reaching an overall, across-domain judgment; and extensions for cross-over and cluster designs. We have developed a draft version of a revised tool that addresses these issues, informed by recent developments in assessing non-randomized studies. The tool includes signalling questions within bias domains, and structured guidance for reaching overall risk of bias judgment.
The workshop will cover:
1. A brief overview of the development and structure of the revised tool.
2. A structured discussion with participants, focusing on:
- the role of signalling questions;
- extensions for cross-over and cluster trials;
- structures and guidance for reaching an overall risk of bias per outcome;
- conceptual continuity of bias assessment across the tools for non-randomized studies and trials;
- further development needs for guidance and software;
- implications for training;
- timelines for implementation;
- other feedback.