This lesson is the seventh in the series of online learning materials designed to help guideline developers and authors of systematic reviews learn how to use the GRADE approach to grade the evidence in systematic reviews, to create Summary of Findings Tables and GRADE Evidence Profiles, and move from evidence to making recommendations.
This lesson points out that publication bias leads to a systematic under- or over-estimate of the effect due to selective publication of studies. It explains when we should suspect publication bias and how publication bias affects the quality of evidence in the GRADE approach. Funnel plots are presented as a way to explore publication bias.
This presentation requires Adobe Flash Player to run.
- understand the concept of publication bias and how it affects the quality of evidence
- interpret funnel plots
- decide whether or not to downgrade the quality of evidence based on publication bias
Prepared by Holger Schünemann and Nancy Santesso from the Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Canada. Funding for the development of this lesson was provided by the Cochrane Collaboration through the grant 'Training and Capacity Building for Summary of Findings Tables' and the World Health Organization.