如何使⽤R语⾔做meta分析中的敏感性分析敏感性分析对于
library(meta)
data(Fleiss93)
m1 <- metabin(event.e, n.e, event.c, n.c,
data = Fleiss93, studlab = study,
sm = "RR", method = "I")
m1
metainf(m1)
metainf(m1, pooled = "random")
forest(metainf(m1))
forest(metainf(m1), layout = "revman5")
forest(metainf(m1, pooled = "random"))
metainf(m1, sortvar = study)
metainf(m1, sortvar = 7:1)
m2 <- update(m1, title = "Fleiss93 meta-analysis",
backtransf = FALSE)
metainf(m2)
Loading 'meta' package (version 4.9-5).
Type 'help(meta)' for a brief overview.
RR 95%-CI %W(fixed) %W(random)
MRC-1 0.7420 [0.5223; 1.0543] 2.4 7.8
CDP 0.6993 [0.4828; 1.0129] 2.1 7.2
MRC-2 0.8270 [0.6487; 1.0545] 4.9 13.6
GASP 0.8209 [0.5269; 1.2789] 1.5 5.3
PARIS 0.8193 [0.5927; 1.1326] 2.8 8.9
AMIS 1.1183 [0.9411; 1.3289] 9.7 20.4
ISIS-2 0.9142 [0.8596; 0.9722] 76.6 36.8
Number of studies combined: k = 7
RR 95%-CI z p-value
Fixed effect model 0.9137 [0.8658; 0.9643] -3.28 0.0010 Random effects model 0.8929 [0.8006; 0.9959] -2.03 0.0419 Quantifying heterogeneity:
tau^2 = 0.0074; H = 1.29 [1.00; 1.98]; I^2 = 39.6% [0.0%; 74.6%] Test of heterogeneity:
Q d.f. p-value
9.93 6 0.1277
Details on meta-analytical method:
- Inver variance method
- DerSimonian-Laird estimator for tau^2
Influential analysis (Fixed effect model)
RR 95%-CI p-value tau^2 I^2
Omitting MRC-1 0.9183 [0.8696; 0.9698] 0.0022 0.0074 41.5% Omitting CDP 0.9190 [0.8703; 0.9705] 0.0024 0.0060 36.6% Omitting MRC-2 0.9185 [0.8691; 0.9706] 0.0025 0.0100 45.9% Omitting GASP 0.9152 [0.8669; 0.9663] 0.0014 0.0095 48.5% Omitting PARIS 0.9166 [0.8679; 0.9680] 0.0018 0.0095 47.3% Omitting AMIS 0.8940 [0.8447; 0.9462] 0.0001 0.0000 0.0% Omitting ISIS-2 0.9124 [0.8162; 1.0199] 0.1066 0.0214 49.6% Pooled estimate 0.9137 [0.8658; 0.9643] 0.0010 0.0074 39.6% Details on meta-analytical method:
- Inver variance method
Influential analysis (Random effects model)
RR 95%-CI p-value tau^2 I^2
RR 95%-CI p-value tau^2 I^2
Omitting MRC-1 0.9071 [0.8098; 1.0162] 0.0924 0.0074 41.5% Omitting CDP 0.9124 [0.8201; 1.0151] 0.0921 0.0060 36.6% Omitting MRC-2 0.8983 [0.7907; 1.0204] 0.0991 0.0100 45.9% Omitting GASP 0.8931 [0.7922; 1.0068] 0.0643 0.0095 48.5% Omitting PARIS 0.8963 [0.7929; 1.0131] 0.0799 0.0095 47.3% Omitting AMIS 0.8940 [0.8447; 0.9462] 0.0001 0.0000 0.0% Omitting ISIS-2 0.8596 [0.7249; 1.0194] 0.0820 0.0214 49.6% Pooled estimate 0.8929 [0.8006; 0.9959] 0.0419 0.0074 39.6% Details on meta-analytical method:
- Inver variance method
- DerSimonian-Laird estimator for tau^2
Influential analysis (Fixed effect model)
RR 95%-CI p-value tau^2 I^2
Omitting AMIS 0.8940 [0.8447; 0.9462] 0.0001 0.0000 0.0% Omitting CDP 0.9190 [0.8703; 0.9705] 0.
0024 0.0060 36.6% Omitting GASP 0.9152 [0.8669; 0.9663] 0.0014 0.0095 48.5% Omitting ISIS-2 0.9124 [0.8162; 1.0199] 0.1066 0.0214 49.6% Omitting MRC-1 0.9183 [0.8696; 0.9698] 0.0022 0.0074 41.5% Omitting MRC-2 0.9185 [0.8691; 0.9706] 0.0025 0.0100 45.9% Omitting PARIS 0.9166 [0.8679; 0.9680] 0.0018 0.0095 47.3% Pooled estimate 0.9137 [0.8658; 0.9643] 0.0010 0.0074 39.6% Details on meta-analytical method:
- Inver variance method
Influential analysis (Fixed effect model)
RR 95%-CI p-value tau^2 I^2
Omitting ISIS-2 0.9124 [0.8162; 1.0199] 0.1066 0.0214 49.6% Omitting AMIS 0.8940 [0.8447; 0.9462] 0.0001 0.0000 0.0%
Omitting AMIS 0.8940 [0.8447; 0.9462] 0.0001 0.0000 0.0% Omitting PARIS 0.9166 [0.8679; 0.9680] 0.0018 0.0095 47.3% Omitting GASP 0.9152 [0.8669; 0.9663] 0.0014 0.0095 48.5% Omitting MRC-2 0.9185 [0.8691; 0.9706] 0.0025 0.0100 45.9% Omitting CDP 0.9190 [0.8703; 0.9705] 0.0024 0.0060 36.6% Omitting MRC-1 0.9183 [0.8696; 0.9698] 0.0022 0.0074 41.5% Pooled estimate 0.9137 [0.8658; 0.9643] 0.0010 0.0074 39.6% Details on meta-analytical method:
- Inver variance method
Review: Fleiss93 meta-analysis
Influential analysis (Fixed effect model)
logRR 95%-CI p-value tau^2 I^2
Omitting MRC-1 -0.0852 [-0.1397; -0.0307] 0.0022 0.0074 41.5% Omitting CDP -0.0844 [-0.1389; -0.0300] 0.0024 0.0060 36.6% Omitting MRC-2 -0.0850 [-0.1403; -0.0298] 0.0025 0.0100 45.9% Omitting GASP -0.0886 [-0.1429; -0.0343] 0.0014 0.0095 48.5% Omitting PARIS -0.0871 [-0.1417; -0.0325] 0.0018 0.0095 47.3% Omitting AMIS -0.1120 [-0.1687; -0.0553] 0.0001 0.0000 0.0% Omitting ISIS-2 -0.0917 [-0.2031; 0.0197] 0.1066 0.0214 49.6% Pooled estimate -0.0902 [-0.1441; -0.0363] 0.0010 0.0074 39.6% Details on meta-analytical method:
- Inver variance method
Influential analysis (Fixed effect model)
SMD 95%-CI p-value tau^2 I^2
Omitting 1 -0.3439 [-0.6239; -0.0638] 0.0161 0.0200 18.4% Omitting 2 -0.2359 [-0.5567; 0.0848] 0.1494 0.0000 0.0% Omitting 3 -0.3632 [-0.6808; -0.0456] 0.0250 0.0247 17.3%
Omitting 4 -0.4464 [-0.7373; -0.1555] 0.0026 0.0000 0.0% Omitting 5 -0.3156 [-0.5882; -0.0431] 0.0232 0.0020 2.5% Pooled estimate -0.3434 [-0.6068; -0.0800] 0.0106 0.0000 0.0% Details on meta-analytical method:
- Inver variance metho
公众号推荐
“急重症抢救”是任何临床医⽣ &护⼠执业⽣涯都会遇到的情况！
微信公众号名⽚ 版权声明
平台原创
