Published in:
Open Access 01-12-2016 | Erratum
Erratum to: Bias and precision of methods for estimating the difference in restricted mean survival time from an individual patient data meta-analysis
Published in: BMC Medical Research Methodology | Issue 1/2016
Login to get accessExcerpt
After publication of the original article [1], it came to the authors’ attention that there were errors affecting the equations and Table 3. These corrections were mentioned by the authors during the proofing process, but unfortunately not implemented before final publication.
Table 3
Results for comparisons of methods in estimating the difference in restricted mean survival time (rmstD) in MAC-NPC and MAC-NPC2 meta-analyses
|
Meta-analysis model
|
Methods
|
MAC-NPC
|
MAC-NPC2
|
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
|
t * = 5 years
|
t * = 10 years
|
t * = 5 years
|
t * = 10 years
|
||||||||||
|
rmstD
|
SE
|
p-value
|
rmstD
|
SE
|
p-value
|
rmstD
|
SE
|
p-value
|
rmstD
|
SE
|
p-value
|
||
|
Naïve Kaplan-Meier
|
0.20
|
0.08
|
0.008
|
0.51
|
0.19
|
0.006
|
0.17
|
0.04
|
<0.001
|
0.54
|
0.11
|
<0.001
|
|
|
Random effects
|
Pooled Kaplan-Meier
|
0.17
|
0.11
|
0.106
|
0.49
|
0.28
|
0.081
|
0.20
|
0.05
|
<0.001
|
0.59
|
0.13
|
<0.001
|
|
Pooled Exponential
|
0.17
|
0.09
|
0.076
|
0.51
|
0.29
|
0.078
|
0.17
|
0.03
|
<0.001
|
0.55
|
0.11
|
<0.001
|
|
|
Peto-quintile
|
0.23
|
0.09
|
0.007
|
0.55
|
0.22
|
0.011
|
0.21
|
0.04
|
<0.001
|
0.59
|
0.12
|
<0.001
|
|
|
Fixed effect
|
Pooled Kaplan-Meier
|
0.20
|
0.07
|
0.005
|
0.52
|
0.18
|
0.004
|
0.18
|
0.04
|
<0.001
|
0.59
|
0.10
|
<0.001
|
|
Pooled Exponential
|
0.18
|
0.06
|
0.003
|
0.55
|
0.18
|
0.002
|
0.17
|
0.03
|
<0.001
|
0.56
|
0.09
|
<0.001
|
|
|
Peto-quintile
|
0.20
|
0.07
|
0.006
|
0.46
|
0.16
|
0.004
|
0.18
|
0.04
|
<0.001
|
0.53
|
0.09
|
<0.001
|
|