Relationship between variation baseline risk and selection of effect size of binary data
FU Wen-jie
HE Qian
XU Yang-peng
TAO Yuan
ZHANG Chao
NIU Yu-ming
Abstract：Objective To explore the selection of effect size of binary data in Meta-analysis.Methods Combining with the reported data, and considering the pattern of variable baseline risk, four indexes of effect size, including odds ratios (OR); risk differences (RD), risk ratios of benefit [RR(B)]; and relative risk of harmful [RR(H)] were selected and compared in binary data.Results When the predicted baseline risk changes, the results of RR(B) andRR(H) will change in absolute benefit indexes of treatment effect. There will be deviations between observed results and actual results when the size of heterogeneity is taken alone as the selective criteria of effect size.Conclusion The selection of effect size is influenced by the changes of baseline risk to a large extent, and heterogeneity is depended alone will lead to result deviations. At the same time, empirical evidence and clinical data features should be taken into consideration.
Keywords：Meta-analysisEffect sizeBaseline riskBinary data
Publication Date：2016-01-01
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 1291-1295 )