A case study of binary outcome data extraction across three systematic reviews of hip arthroplasty: errors and differences of selection

Ref ID 256
First Author C. Carroll
Journal BMC RESEARCH NOTES
Year Of Publishing 2013
URL https://pubmed.ncbi.nlm.nih.gov/24344873/
Keywords • Surgery
• Inference
• Error
• Single reviewer
Problem(s) • Single reviewer / lack of double checking
• Data extraction errors and double counting
Article Type Empirical
Article Subtype Small analytical study
First Author Country United Kingdom
Aim To assess differences in the binary event data extracted and analysed by three systematic reviews on the same topic: a comparison of total hip arthroplasty and hemiarthroplasty
Level of Investigation Analytical
Summary of Findings Assessment of differences in the binary event data extracted and analysed by three systematic reviews of total hip arthroplasty and hemiarthroplasty found differences in selection accounted for between 8% and 42% of the data differences between reviews. Errors in data extraction accounted for between 8% and 17% of the differences. These differences did lead to small differences in meta-analysed relative risks between the two treatments in the three reviews, but none was significant.
Number of systematic reviews included 3
Number of eligible systematic reviews assessed 3