Discrepancies in meta-analyses answering the same clinical question were hard to explain: a meta-epidemiological study

Ref ID 588
First Author C. Hacke
Journal JOURNAL OF CLINICAL EPIDEMIOLOGY
Year Of Publishing 2020
URL https://www.jclinepi.com/article/S0895-4356(19)30074-5/fulltext
Keywords • Cochrane
• Statistical
• General medical
• Inference
• Overlapping reviews/redundancy
• Non-Cochrane reviews
• Missing data
Problem(s) • Errors in study inclusion or omission of relevant studies
• Cochrane reviews more rigorous/higher quality than non-Cochrane reviews
• Redundant / overlapping / duplicated review question; leads to research waste
Article Type Empirical
Article Subtype Meta-epidemiological analysis
First Author Country Germany
Aim To assess quantitative concordance in effect estimates between meta-analyses from 24 Cochrane and 20 non-Cochrane systematic reviews matched on population, intervention, condition, and outcome. To explore reasons for discordances between matched-review conclusions.
Level of Investigation Analytical
Summary of Findings On average, meta-analyses from non-Cochrane reviews reported higher effect estimates compared with meta-analyses from Cochrane reviews answering the same clinical question. Disagreements in the interpretation of eligibility criteria were identified as reasons underpinning discrepant findings in 14 pairs
Number of systematic reviews included 44
Number of eligible systematic reviews assessed 44
Treatment impacted Yes
Treatment impacted description
Interpretation impacted Yes
Interpretation impacted description Forest plot demonstrates one matched pair of reviews was found to be at opposing sides of the line of no effect