Skip to main content

Table 3 Number (%) of (non-)significant meta-analyses under different scales for two-stage models (“binary” and “OEV” data)

From: Implications of analysing time-to-event outcomes as binary in meta-analysis: empirical evidence from the Cochrane Database of Systematic Reviews

  Outcome   OR HR (O-E & V)
“binary”
    Significant Non-significant Significant Non-Significant
HR
(clog-log)
All-cause mortality Significant 106 (15%) 2 (0.1%)   
Non-significant 4 (0.6%) 603 (84%)   
“OEV”
    Significant Non-significant Significant Non-Significant
HR
(clog-log)
Overall Survival Significant 20 (29%) 1 (0.2%) 18 (26%) 10 (14%)
Non-significant 1 (0.2%) 47 (68%) 3 (4%) 38 (55%)
Progression / Disease free Survival Significant 9 (29%) 0 (0%) 8 (26%) 6 (19%)
Non-significant 1 (3%) 21 (68%) 1 (3%) 16 (52%)
HR
(O-E &V)
Overall Survival Significant 18 (26%) 10 (14%)   
Non-significant 3 (4%) 38 (55%)   
Progression / Disease free Survival Significant 9 (29%) 5 (16%)   
Non-significant 1 (3%) 16 (52%)