Catalogue of Bias

< Back to search results

The Catalogue contains definitions, explanations, examples and preventive measures for biases affecting health evidence.

• Format Websites
• Target Audience Further education, Researchers
• EBM Stage 3 - Appraising evidence
• Duration >15 mins
• Difficulty Intermediate

Key Concepts addressed

• Beware of claims like these
• Too good to be true
• Check the evidence from treatment comparisons
• Don’t be misled by unfair comparisons
• 2-1a Comparison groups should be similar
• 2-1c Comparison groups should be treated equally
• 2-1d People should not know which treatment they get
• 2-1e Peoples' outcomes should be assessed similarly
• 2-1g All should be followed up
• 2-1h Peoples' outcomes should be analyzed in their original groups
• Don’t be misled by unreliable summaries of treatment comparisons
• 2-2a Reviews of fair comparisons should be systematic
• 2-2b All fair comparisons and outcomes should be reported
• Don’t be misled by how treatment effects are described
• 2-3g Statistical significance is not the same as importance
• 2-3b Relative measures of effects can be misleading
• 2-3c Average measures of effects can be misleading
• 2-3d Fair comparisons with few people or outcome events can be misleading
• 2-3e Subgroup analyses may be misleading
• 1-3d Peer-review and publication does not guarantee reliable information
• 1-2f Consider all of the relevant fair comparisons

Details

The Catalogue of Bias is a collaboration led by Oxford Centre for Evidence-Based Medicine.

The Catalogue contains definitions, explanations, examples and preventive measures for biases affecting health evidence.

The project takes as its starting point David Sackett’s 1979 paper “Bias in Analytical Research” (Chron Dis 1979;32:51-63).  The aim is to map all of the biases that affect health evidence.  New entries are being added all the time.  Users can sign up to join in the editorial effort or provide feedback on existing entries.