library(netmeta)
## Loading required package: meta
## Loading 'meta' package (version 4.8-4).
## Type 'help(meta)' for a brief overview.
## Loading 'netmeta' package (version 0.9-6).
## Type 'help(netmeta-package)' for a brief overview.
data(Senn2013)
# Fixed effect model (default)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
as.data.frame(net1)
## studlab treat1 treat2 TE seTE seTE.adj TE.nma.fixed
## 1 DeFronzo1995 metf plac -1.90 0.1414 0.1414000 -1.11411019
## 2 Lewin2007 metf plac -0.82 0.0992 0.0992000 -1.11411019
## 3 Willms1999 acar metf 0.20 0.3579 0.3883779 0.28674122
## 4 Davidson2007 plac rosi 1.34 0.1435 0.1435000 1.20184201
## 5 Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141000 1.20184201
## 6 Kipnes2001 piog plac -1.30 0.1268 0.1268000 -1.06643311
## 7 Kerenyi2004 plac rosi 0.77 0.1078 0.1078000 1.20184201
## 8 Hanefeld2004 metf piog -0.16 0.0849 0.0849000 -0.04767708
## 9 Derosa2004 piog rosi 0.10 0.1831 0.1831000 0.13540890
## 10 Baksi2004 plac rosi 1.30 0.1014 0.1014000 1.20184201
## 11 Rosenstock2008 plac rosi 1.09 0.2263 0.2263000 1.20184201
## 12 Zhu2003 plac rosi 1.50 0.1624 0.1624000 1.20184201
## 13 Yang2003 metf rosi 0.14 0.2239 0.2239000 0.08773182
## 14 Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436000 -0.76234486
## 15 Oyama2008 acar sulf -0.40 0.1549 0.1549000 -0.38787182
## 16 Costa1997 acar plac -0.80 0.1432 0.1432000 -0.82736897
## 17 Hermansen2007 plac sita 0.57 0.1291 0.1291000 0.57000000
## 18 Garber2008 plac vild 0.70 0.1273 0.1273000 0.70000000
## 19 Alex1998 metf sulf -0.37 0.1184 0.1184000 -0.67461304
## 20 Johnston1994 migl plac -0.74 0.1839 0.1839000 -0.94393296
## 21 Johnston1998a migl plac -1.41 0.2235 0.2235000 -0.94393296
## 22 Kim2007 metf rosi 0.00 0.2339 0.2339000 0.08773182
## 23 Johnston1998b migl plac -0.68 0.2828 0.2828000 -0.94393296
## 24 Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356000 -1.11411019
## 25 Stucci1996 benf plac -0.23 0.3467 0.3467000 -0.90518645
## 26 Moulin2006 benf plac -1.01 0.1366 0.1366000 -0.90518645
## 27 Willms1999 metf plac -1.20 0.3758 0.4125252 -1.11411019
## 28 Willms1999 acar plac -1.00 0.4669 0.8241911 -0.82736897
## seTE.nma.fixed w.fixed Q.fixed TE.nma.random seTE.nma.random
## 1 0.05961148 50.015105 3.089047e+01 -1.126775351 0.1542666
## 2 0.05961148 101.619407 8.790160e+00 -1.126775351 0.1542666
## 3 0.11456703 6.629656 4.988179e-02 0.284990413 0.2580630
## 4 0.04766217 48.561959 9.269327e-01 1.233455753 0.1278016
## 5 0.04766217 76.811936 7.966777e-01 1.233455753 0.1278016
## 6 0.07584751 62.195862 3.393002e+00 -1.129091954 0.2196381
## 7 0.04766217 86.052299 1.604768e+01 1.233455753 0.1278016
## 8 0.06980214 138.734547 1.750336e+00 0.002316603 0.2255576
## 9 0.08177728 29.827943 3.739799e-02 0.104363799 0.2240281
## 10 0.04766217 97.257721 9.370773e-01 1.233455753 0.1278016
## 11 0.04766217 19.526792 2.442535e-01 1.233455753 0.1278016
## 12 0.04766217 37.916475 3.370706e+00 1.233455753 0.1278016
## 13 0.06766230 19.947653 5.449624e-02 0.106680402 0.1651709
## 14 0.09201201 48.494347 9.288705e+00 -0.816896953 0.2371470
## 15 0.11310215 41.677069 6.130393e-03 -0.425226138 0.2654975
## 16 0.10852085 48.765644 3.652842e-02 -0.841784938 0.2458203
## 17 0.12910000 59.999484 1.243185e-27 0.570000000 0.3540957
## 18 0.12730000 61.708245 7.304926e-27 0.700000000 0.3534434
## 19 0.08854114 71.334003 6.619018e+00 -0.710216551 0.2352538
## 20 0.12690612 29.568993 1.229735e+00 -0.949726378 0.2317585
## 21 0.12690612 20.019118 4.348523e+00 -0.949726378 0.2317585
## 22 0.06766230 18.278458 1.406870e-01 0.106680402 0.1651709
## 23 0.12690612 12.503776 8.710207e-01 -0.949726378 0.2317585
## 24 0.05961148 5.270166 2.687539e+00 -1.126775351 0.1542666
## 25 0.12709113 8.319406 3.792624e+00 -0.731147215 0.2860751
## 26 0.12709113 53.591832 5.887535e-01 -0.731147215 0.2860751
## 27 0.05961148 5.876234 4.334933e-02 -1.126775351 0.1542666
## 28 0.10852085 1.472123 4.387144e-02 -0.841784938 0.2458203
## w.random
## 1 7.769351
## 2 8.434721
## 3 3.219115
## 4 7.733403
## 5 8.214515
## 6 8.013131
## 7 8.309943
## 8 8.626272
## 9 7.030246
## 10 8.403440
## 11 6.252790
## 12 7.402435
## 13 6.295321
## 14 7.731687
## 15 7.535174
## 16 7.738551
## 17 7.975516
## 18 8.004981
## 19 8.147603
## 20 7.015765
## 21 6.302421
## 22 6.118972
## 23 5.299620
## 24 3.350485
## 25 4.368380
## 26 7.850743
## 27 2.970312
## 28 1.515938
as.data.frame(net1, details=TRUE)
## studlab treat1 treat2 TE seTE seTE.adj TE.nma.fixed
## 1 DeFronzo1995 metf plac -1.90 0.1414 0.1414000 -1.11411019
## 2 Lewin2007 metf plac -0.82 0.0992 0.0992000 -1.11411019
## 3 Willms1999 acar metf 0.20 0.3579 0.3883779 0.28674122
## 4 Davidson2007 plac rosi 1.34 0.1435 0.1435000 1.20184201
## 5 Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141000 1.20184201
## 6 Kipnes2001 piog plac -1.30 0.1268 0.1268000 -1.06643311
## 7 Kerenyi2004 plac rosi 0.77 0.1078 0.1078000 1.20184201
## 8 Hanefeld2004 metf piog -0.16 0.0849 0.0849000 -0.04767708
## 9 Derosa2004 piog rosi 0.10 0.1831 0.1831000 0.13540890
## 10 Baksi2004 plac rosi 1.30 0.1014 0.1014000 1.20184201
## 11 Rosenstock2008 plac rosi 1.09 0.2263 0.2263000 1.20184201
## 12 Zhu2003 plac rosi 1.50 0.1624 0.1624000 1.20184201
## 13 Yang2003 metf rosi 0.14 0.2239 0.2239000 0.08773182
## 14 Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436000 -0.76234486
## 15 Oyama2008 acar sulf -0.40 0.1549 0.1549000 -0.38787182
## 16 Costa1997 acar plac -0.80 0.1432 0.1432000 -0.82736897
## 17 Hermansen2007 plac sita 0.57 0.1291 0.1291000 0.57000000
## 18 Garber2008 plac vild 0.70 0.1273 0.1273000 0.70000000
## 19 Alex1998 metf sulf -0.37 0.1184 0.1184000 -0.67461304
## 20 Johnston1994 migl plac -0.74 0.1839 0.1839000 -0.94393296
## 21 Johnston1998a migl plac -1.41 0.2235 0.2235000 -0.94393296
## 22 Kim2007 metf rosi 0.00 0.2339 0.2339000 0.08773182
## 23 Johnston1998b migl plac -0.68 0.2828 0.2828000 -0.94393296
## 24 Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356000 -1.11411019
## 25 Stucci1996 benf plac -0.23 0.3467 0.3467000 -0.90518645
## 26 Moulin2006 benf plac -1.01 0.1366 0.1366000 -0.90518645
## 27 Willms1999 metf plac -1.20 0.3758 0.4125252 -1.11411019
## 28 Willms1999 acar plac -1.00 0.4669 0.8241911 -0.82736897
## seTE.nma.fixed lower.nma.fixed upper.nma.fixed leverage.fixed
## 1 0.05961148 -1.23094654 -0.99727384 0.17773009
## 2 0.05961148 -1.23094654 -0.99727384 0.36110743
## 3 0.11456703 0.06219397 0.51128847 0.08701825
## 4 0.04766217 1.10842588 1.29525814 0.11031733
## 5 0.04766217 1.10842588 1.29525814 0.17449229
## 6 0.07584751 -1.21509149 -0.91777473 0.35780311
## 7 0.04766217 1.10842588 1.29525814 0.19548346
## 8 0.06980214 -0.18448676 0.08913259 0.67596166
## 9 0.08177728 -0.02487163 0.29568943 0.19947509
## 10 0.04766217 1.10842588 1.29525814 0.22093862
## 11 0.04766217 1.10842588 1.29525814 0.04435866
## 12 0.04766217 1.10842588 1.29525814 0.08613417
## 13 0.06766230 -0.04488385 0.22034749 0.09132408
## 14 0.09201201 -0.94268508 -0.58200464 0.41056332
## 15 0.11310215 -0.60954796 -0.16619569 0.53313704
## 16 0.10852085 -1.04006592 -0.61467202 0.57430196
## 17 0.12910000 0.31696865 0.82303135 1.00000000
## 18 0.12730000 0.45049658 0.94950342 1.00000000
## 19 0.08854114 -0.84815049 -0.50107559 0.55922533
## 20 0.12690612 -1.19266439 -0.69520154 0.47621347
## 21 0.12690612 -1.19266439 -0.69520154 0.32241117
## 22 0.06766230 -0.04488385 0.22034749 0.08368220
## 23 0.12690612 -1.19266439 -0.69520154 0.20137536
## 24 0.05961148 -1.23094654 -0.99727384 0.01872768
## 25 0.12709113 -1.15428050 -0.65609240 0.13437635
## 26 0.12709113 -1.15428050 -0.65609240 0.86562365
## 27 0.05961148 -1.23094654 -0.99727384 0.02088136
## 28 0.10852085 -1.04006592 -0.61467202 0.01733686
## w.fixed Q.fixed TE.nma.random seTE.nma.random lower.nma.random
## 1 50.015105 3.089047e+01 -1.126775351 0.1542666 -1.429132296
## 2 101.619407 8.790160e+00 -1.126775351 0.1542666 -1.429132296
## 3 6.629656 4.988179e-02 0.284990413 0.2580630 -0.220803689
## 4 48.561959 9.269327e-01 1.233455753 0.1278016 0.982969138
## 5 76.811936 7.966777e-01 1.233455753 0.1278016 0.982969138
## 6 62.195862 3.393002e+00 -1.129091954 0.2196381 -1.559574779
## 7 86.052299 1.604768e+01 1.233455753 0.1278016 0.982969138
## 8 138.734547 1.750336e+00 0.002316603 0.2255576 -0.439768099
## 9 29.827943 3.739799e-02 0.104363799 0.2240281 -0.334723275
## 10 97.257721 9.370773e-01 1.233455753 0.1278016 0.982969138
## 11 19.526792 2.442535e-01 1.233455753 0.1278016 0.982969138
## 12 37.916475 3.370706e+00 1.233455753 0.1278016 0.982969138
## 13 19.947653 5.449624e-02 0.106680402 0.1651709 -0.217048707
## 14 48.494347 9.288705e+00 -0.816896953 0.2371470 -1.281696573
## 15 41.677069 6.130393e-03 -0.425226138 0.2654975 -0.945591609
## 16 48.765644 3.652842e-02 -0.841784938 0.2458203 -1.323583884
## 17 59.999484 1.243185e-27 0.570000000 0.3540957 -0.124014732
## 18 61.708245 7.304926e-27 0.700000000 0.3534434 0.007263732
## 19 71.334003 6.619018e+00 -0.710216551 0.2352538 -1.171305441
## 20 29.568993 1.229735e+00 -0.949726378 0.2317585 -1.403964765
## 21 20.019118 4.348523e+00 -0.949726378 0.2317585 -1.403964765
## 22 18.278458 1.406870e-01 0.106680402 0.1651709 -0.217048707
## 23 12.503776 8.710207e-01 -0.949726378 0.2317585 -1.403964765
## 24 5.270166 2.687539e+00 -1.126775351 0.1542666 -1.429132296
## 25 8.319406 3.792624e+00 -0.731147215 0.2860751 -1.291844014
## 26 53.591832 5.887535e-01 -0.731147215 0.2860751 -1.291844014
## 27 5.876234 4.334933e-02 -1.126775351 0.1542666 -1.429132296
## 28 1.472123 4.387144e-02 -0.841784938 0.2458203 -1.323583884
## upper.nma.random w.random treat1.pos treat2.pos
## 1 -0.82441841 7.769351 3 6
## 2 -0.82441841 8.434721 3 6
## 3 0.79078451 3.219115 1 3
## 4 1.48394237 7.733403 6 7
## 5 1.48394237 8.214515 6 7
## 6 -0.69860913 8.013131 5 6
## 7 1.48394237 8.309943 6 7
## 8 0.44440130 8.626272 3 5
## 9 0.54345087 7.030246 5 7
## 10 1.48394237 8.403440 6 7
## 11 1.48394237 6.252790 6 7
## 12 1.48394237 7.402435 6 7
## 13 0.43040951 6.295321 3 7
## 14 -0.35209733 7.731687 7 9
## 15 0.09513933 7.535174 1 9
## 16 -0.35998599 7.738551 1 6
## 17 1.26401473 7.975516 6 8
## 18 1.39273627 8.004981 6 10
## 19 -0.24912766 8.147603 3 9
## 20 -0.49548799 7.015765 4 6
## 21 -0.49548799 6.302421 4 6
## 22 0.43040951 6.118972 3 7
## 23 -0.49548799 5.299620 4 6
## 24 -0.82441841 3.350485 3 6
## 25 -0.17045042 4.368380 2 6
## 26 -0.17045042 7.850743 2 6
## 27 -0.82441841 2.970312 3 6
## 28 -0.35998599 1.515938 1 6
data(Senn2013)
# Generation of an object of class 'netmeta' with reference treatment 'plac', i.e. placebo
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD", reference="plac")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Decomposition of Cochran's Q
decomp.design(net1)
## Q statistics to assess homogeneity / consistency
##
## Q df p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
##
## Design-specific decomposition of within-designs Q statistic
##
## Design Q df p-value
## metf:rosi 0.19 1 0.6655
## plac:benf 4.38 1 0.0363
## plac:metf 42.16 2 < 0.0001
## plac:migl 6.45 2 0.0398
## plac:rosi 21.27 5 0.0007
##
## Between-designs Q statistic after detaching of single designs
##
## Detached design Q df p-value
## acar:sulf 22.52 6 0.0010
## metf:piog 17.13 6 0.0088
## metf:rosi 22.52 6 0.0010
## metf:sulf 7.51 6 0.2760
## piog:rosi 22.48 6 0.0010
## plac:acar 22.44 6 0.0010
## plac:metf 22.07 6 0.0012
## plac:piog 17.25 6 0.0084
## plac:rosi 16.29 6 0.0123
## rosi:sulf 6.77 6 0.3425
## plac:acar:metf 22.38 5 0.0004
##
## Q statistic to assess consistency under the assumption of
## a full design-by-treatment interaction random effects model
##
## Q df p-value tau.within tau2.within
## Between designs 2.19 7 0.9483 0.3797 0.1442
data(dietaryfat)
# Transform data from arm-based format to contrast-based format
#Using incidence rate ratios (sm="IRR") as effect measure.
# Note, the argument 'sm' is not necessary as this is the default
#in R function metainc called internally
p1 <- pairwise(list(treat1, treat2, treat3),
list(d1, d2, d3),
time=list(years1, years2, years3),
studlab=ID,
data=dietaryfat, sm="IRR")
p1
## TE seTE studlab treat1 treat2 event1
## 1 0.02202212 0.13363595 2 DART 1 2 113
## 2 -1.66363256 1.09544512 10 London Corn /Olive 1 2 1
## 3 -1.23608328 1.15470054 10 London Corn /Olive 1 3 1
## 4 0.42754928 0.73029674 10 London Corn /Olive 2 3 5
## 5 0.13122887 0.30276504 11 London Low Fat 1 2 24
## 6 -0.05863541 0.08803255 14 Minnesota Coronary 1 2 248
## 7 0.15090580 0.26071507 15 MRC Soya 1 2 31
## 8 0.31442233 0.19031014 18 Oslo Diet-Heart 1 2 65
## 9 1.13441029 1.15470054 22 STARS 1 2 3
## 10 -0.40523688 0.24770004 23 Sydney Diet-Heart 1 2 28
## 11 0.04519334 0.10675600 26 Veterans Administration 1 2 177
## 12 0.67701780 1.22474487 27 Veterans Diet & Skin CA 1 2 2
## time1 event2 time2 incr
## 1 1917.0 111 1925.0 0
## 2 43.6 5 41.3 0
## 3 43.6 3 38.0 0
## 4 41.3 3 38.0 0
## 5 393.5 20 373.9 0
## 6 4715.0 269 4823.0 0
## 7 715.0 28 751.0 0
## 8 885.0 48 895.0 0
## 9 87.8 1 91.0 0
## 10 1011.0 39 939.0 0
## 11 1544.0 174 1588.0 0
## 12 125.0 1 123.0 0
# Conduct network meta-analysis:
net1 <- netmeta(p1)
summary(net1)
## Number of studies: k = 10
## Number of treatments: n = 3
## Number of pairwise comparisons: m = 12
## Number of designs: d = 2
##
## Fixed effect model
##
## Treatment estimate (sm = 'IRR'):
## 1 2 3
## 1 . 1.0096 1.1714
## 2 0.9905 . 1.1603
## 3 0.8537 0.8618 .
##
## Lower 95%-confidence limit:
## 1 2 3
## 1 . 0.9084 0.2921
## 2 0.8913 . 0.2902
## 3 0.2129 0.2155 .
##
## Upper 95%-confidence limit:
## 1 2 3
## 1 . 1.1220 4.6969
## 2 1.1008 . 4.6400
## 3 3.4229 3.4464 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.0043; I^2 = 11.1%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 10.12 9 0.3405
## Within designs 7.79 8 0.4546
## Between designs 2.34 1 0.1262
# Conduct network meta-analysis using incidence rate differences
# (sm="IRD").
p2 <- pairwise(list(treat1, treat2, treat3),
list(d1, d2, d3),
time=list(years1, years2, years3),
studlab=ID,
data=dietaryfat, sm="IRD")
net2 <- netmeta(p2)
summary(net2)
## Number of studies: k = 10
## Number of treatments: n = 3
## Number of pairwise comparisons: m = 12
## Number of designs: d = 2
##
## Fixed effect model
##
## Treatment estimate (sm = 'IRD'):
## 1 2 3
## 1 . 0.0000 -0.0411
## 2 -0.0000 . -0.0411
## 3 0.0411 0.0411 .
##
## Lower 95%-confidence limit:
## 1 2 3
## 1 . -0.0060 -0.1395
## 2 -0.0060 . -0.1397
## 3 -0.0574 -0.0575 .
##
## Upper 95%-confidence limit:
## 1 2 3
## 1 . 0.0060 0.0574
## 2 0.0060 . 0.0575
## 3 0.1395 0.1397 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 < 0.0001; I^2 = 16.3%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 10.75 9 0.2931
## Within designs 7.96 8 0.4374
## Between designs 2.79 1 0.0947
# Draw network graph
netgraph(net1, points=TRUE, cex.points=3, cex=1.25)
tname <- c("Control","Diet", "Diet 2")
netgraph(net1, points=TRUE, cex.points=3, cex=1.25, labels=tname)
data(Senn2013)
# Fixed effect model (default)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
forest(net1, ref="plac")
forest(net1, xlim=c(-1.5,1), ref="plac",
xlab="HbA1c difference", rightcols=FALSE)
# Random effects effect model
net2 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD", comb.random=TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
forest(net2, xlim=c(-1.5,1), ref="plac",
xlab="HbA1c difference")
# Add column with P-Scores on right side of forest plot
forest(net2, xlim=c(-1.5,1), ref="plac",
xlab="HbA1c difference",
rightcols=c("effect", "ci", "Pscore"),
rightlabs="P-Score",
just.addcols="right")
# Add column with P-Scores on left side of forest plot
forest(net2, xlim=c(-1.5,1), ref="plac",
xlab="HbA1c difference",
leftcols=c("studlab", "Pscore"),
leftlabs=c("Treatment", "P-Score"),
just.addcols="right")
# Sort forest plot by descending P-Score
forest(net2, xlim=c(-1.5,1), ref="plac",
xlab="HbA1c difference",
rightcols=c("effect", "ci", "Pscore"),
rightlabs="P-Score",just.addcols="right",
sortvar=-Pscore)
# Drop reference group and sort by and print number of studies with direct treatment comparisons
forest(net2, xlim=c(-1.5,1), ref="plac",
xlab="HbA1c difference",
leftcols=c("studlab", "k"),
leftlabs=c("Contrast\nto Placebo", "Direct\nComparisons"),
sortvar=-k,
drop=TRUE,
smlab="Random Effects Model")
# Use depression dataset
data(Linde2015)
# Define order of treatments
trts <- c("TCA", "SSRI", "SNRI", "NRI",
"Low-dose SARI", "NaSSa", "rMAO-A", "Hypericum",
"Placebo")
# Outcome labels
outcomes <- c("Early response", "Early remission")
# (1) Early response
p1 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(resp1, resp2, resp3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 14 TCA SSRI NA 10 NA 11 0 FALSE
## NA NA 18 TCA Placebo NA 73 NA 73 0 FALSE
## NA NA 21 TCA rMAO-A NA 46 NA 98 0 FALSE
## NA NA 27 SSRI Placebo NA 80 NA 81 0 FALSE
## NA NA 51 TCA rMAO-A NA 71 NA 71 0 FALSE
## NA NA 130 TCA NaSSa NA 35 NA 36 0 FALSE
## NA NA 131 SSRI SNRI NA 697 NA 688 0 FALSE
## NA NA 130 TCA Placebo NA 35 NA 34 0 FALSE
## NA NA 130 NaSSa Placebo NA 36 NA 34 0 FALSE
net1 <- netmeta(p1,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 21 TCA rMAO-A NA NA
## 27 SSRI Placebo NA NA
## 51 TCA rMAO-A NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## 131 SSRI SNRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# (2) Early remission
p2 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(remi1, remi2, remi3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr
## NA NA 1 TCA SNRI NA 75 NA 78 0.0
## NA NA 11 TCA SSRI NA 108 NA 99 0.0
## NA NA 14 TCA SSRI NA 10 NA 11 0.0
## NA NA 18 TCA Placebo NA 73 NA 73 0.0
## NA NA 20 TCA SSRI NA 55 NA 51 0.0
## NA NA 26 SSRI Placebo NA 314 NA 154 0.0
## NA NA 53 SSRI NaSSa NA 122 NA 121 0.0
## NA NA 56 TCA SSRI NA 92 NA 380 0.0
## NA NA 73 Hypericum Placebo NA 55 NA 57 0.0
## NA NA 90 TCA SSRI NA 42 NA 42 0.0
## NA NA 96 TCA SSRI 0 30 0 29 0.5
## NA NA 121 Low-dose SARI NaSSa NA 43 NA 40 0.0
## NA NA 130 TCA NaSSa NA 35 NA 36 0.0
## NA NA 1 TCA Placebo NA 75 NA 76 0.0
## NA NA 11 TCA Placebo NA 108 NA 101 0.0
## NA NA 53 SSRI Placebo NA 122 NA 129 0.0
## NA NA 130 TCA Placebo NA 35 NA 34 0.0
## NA NA 1 SNRI Placebo NA 78 NA 76 0.0
## NA NA 11 SSRI Placebo NA 99 NA 101 0.0
## NA NA 53 NaSSa Placebo NA 121 NA 129 0.0
## NA NA 130 NaSSa Placebo NA 36 NA 34 0.0
## allstudies
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
net2 <- netmeta(p2,
comb.fixed = FALSE, comb.random = TRUE, seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 1 TCA Placebo NA NA
## 1 TCA SNRI NA NA
## 1 SNRI Placebo NA NA
## 11 SSRI Placebo NA NA
## 11 TCA Placebo NA NA
## 11 TCA SSRI NA NA
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 20 TCA SSRI NA NA
## 26 SSRI Placebo NA NA
## 53 NaSSa Placebo NA NA
## 53 SSRI NaSSa NA NA
## 53 SSRI Placebo NA NA
## 56 TCA SSRI NA NA
## 73 Hypericum Placebo NA NA
## 90 TCA SSRI NA NA
## 96 TCA SSRI NA NA
## 121 Low-dose SARI NaSSa NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Partial order of treatment rankings (all five outcomes)
po <- netposet(netrank(net1, small.values = "bad"),
netrank(net2, small.values = "bad"),
outcomes = outcomes)
po
## Random effects model
## Hypericum Low-dose SARI NaSSa NRI Placebo rMAO-A SNRI SSRI
## Hypericum 0 1 0 0 0 0 0 0
## Low-dose SARI 0 0 0 1 0 0 0 1
## NaSSa 0 0 0 0 0 1 0 0
## NRI 0 0 1 0 0 0 0 0
## Placebo 0 0 0 0 0 0 0 0
## rMAO-A 0 0 0 0 1 0 0 0
## SNRI 0 0 0 0 0 0 0 0
## SSRI 0 0 1 0 0 0 0 0
## TCA 0 0 0 1 0 0 0 1
## TCA
## Hypericum 1
## Low-dose SARI 0
## NaSSa 0
## NRI 0
## Placebo 0
## rMAO-A 0
## SNRI 1
## SSRI 0
## TCA 0
data(Linde2015)
# Transform data from arm-based format to contrast-based format
#Outcome: early response
p1 <- pairwise(list(treatment1, treatment2, treatment3),
event = list(resp1, resp2, resp3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 14 TCA SSRI NA 10 NA 11 0 FALSE
## NA NA 18 TCA Placebo NA 73 NA 73 0 FALSE
## NA NA 21 TCA rMAO-A NA 46 NA 98 0 FALSE
## NA NA 27 SSRI Placebo NA 80 NA 81 0 FALSE
## NA NA 51 TCA rMAO-A NA 71 NA 71 0 FALSE
## NA NA 130 TCA NaSSa NA 35 NA 36 0 FALSE
## NA NA 131 SSRI SNRI NA 697 NA 688 0 FALSE
## NA NA 130 TCA Placebo NA 35 NA 34 0 FALSE
## NA NA 130 NaSSa Placebo NA 36 NA 34 0 FALSE
# Define order of treatments
trts <- c("TCA", "SSRI", "SNRI", "NRI",
"Low-dose SARI", "NaSSa", "rMAO-A", "Hypericum",
"Placebo")
# Conduct network meta-analysis
net1 <- netmeta(p1, comb.fixed = FALSE, comb.random = TRUE,
reference = "Placebo",
seq = trts)
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 21 TCA rMAO-A NA NA
## 27 SSRI Placebo NA NA
## 51 TCA rMAO-A NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## 131 SSRI SNRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
print(summary(net1), digits = 2)
## Number of studies: k = 59
## Number of treatments: n = 9
## Number of pairwise comparisons: m = 73
## Number of designs: d = 21
##
## Random effects model
##
## Treatment estimate (sm = 'OR', comparison: other treatments vs 'Placebo'):
## OR 95%-CI
## TCA 1.72 [1.42; 2.09]
## SSRI 1.68 [1.40; 2.01]
## SNRI 1.74 [1.25; 2.42]
## NRI 1.42 [0.84; 2.40]
## Low-dose SARI 1.78 [1.18; 2.70]
## NaSSa 1.14 [0.82; 1.60]
## rMAO-A 1.05 [0.69; 1.62]
## Hypericum 1.99 [1.58; 2.49]
## Placebo . .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.0352; I^2 = 26.9%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 79.37 58 0.0327
## Within designs 49.41 39 0.1226
## Between designs 29.95 19 0.0524
data(Senn2013)
nc1 <- netconnection(treat1, treat2, studlab, data = Senn2013)
nc1
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of networks: 1
##
## Distance matrix:
## acar benf metf migl piog plac rosi sita sulf vild
## acar . 2 1 2 2 1 2 2 1 2
## benf 2 . 2 2 2 1 2 2 3 2
## metf 1 2 . 2 1 1 1 2 1 2
## migl 2 2 2 . 2 1 2 2 3 2
## piog 2 2 1 2 . 1 1 2 2 2
## plac 1 1 1 1 1 . 1 1 2 1
## rosi 2 2 1 2 1 1 . 2 1 2
## sita 2 2 2 2 2 1 2 . 3 2
## sulf 1 3 1 3 2 2 1 3 . 3
## vild 2 2 2 2 2 1 2 2 3 .
# Extract number of (sub)networks
nc1$n.subnets
## [1] 1
# Extract distance matrix
nc1$D.matrix
## acar benf metf migl piog plac rosi sita sulf vild
## acar 0 2 1 2 2 1 2 2 1 2
## benf 2 0 2 2 2 1 2 2 3 2
## metf 1 2 0 2 1 1 1 2 1 2
## migl 2 2 2 0 2 1 2 2 3 2
## piog 2 2 1 2 0 1 1 2 2 2
## plac 1 1 1 1 1 0 1 1 2 1
## rosi 2 2 1 2 1 1 0 2 1 2
## sita 2 2 2 2 2 1 2 0 3 2
## sulf 1 3 1 3 2 2 1 3 0 3
## vild 2 2 2 2 2 1 2 2 3 0
# Conduct network meta-analysis
net1 <- netmeta(TE, seTE, treat1, treat2, studlab, data = Senn2013)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
net1
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.1414 0.1414 2
## Lewin2007 metf plac -0.82 0.0992 0.0992 2
## Willms1999 acar metf 0.20 0.3579 0.3884 3 *
## Davidson2007 plac rosi 1.34 0.1435 0.1435 2
## Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141 2
## Kipnes2001 piog plac -1.30 0.1268 0.1268 2
## Kerenyi2004 plac rosi 0.77 0.1078 0.1078 2
## Hanefeld2004 metf piog -0.16 0.0849 0.0849 2
## Derosa2004 piog rosi 0.10 0.1831 0.1831 2
## Baksi2004 plac rosi 1.30 0.1014 0.1014 2
## Rosenstock2008 plac rosi 1.09 0.2263 0.2263 2
## Zhu2003 plac rosi 1.50 0.1624 0.1624 2
## Yang2003 metf rosi 0.14 0.2239 0.2239 2
## Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436 2
## Oyama2008 acar sulf -0.40 0.1549 0.1549 2
## Costa1997 acar plac -0.80 0.1432 0.1432 2
## Hermansen2007 plac sita 0.57 0.1291 0.1291 2
## Garber2008 plac vild 0.70 0.1273 0.1273 2
## Alex1998 metf sulf -0.37 0.1184 0.1184 2
## Johnston1994 migl plac -0.74 0.1839 0.1839 2
## Johnston1998a migl plac -1.41 0.2235 0.2235 2
## Kim2007 metf rosi 0.00 0.2339 0.2339 2
## Johnston1998b migl plac -0.68 0.2828 0.2828 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356 2
## Stucci1996 benf plac -0.23 0.3467 0.3467 2
## Moulin2006 benf plac -1.01 0.1366 0.1366 2
## Willms1999 metf plac -1.20 0.3758 0.4125 3 *
## Willms1999 acar plac -1.00 0.4669 0.8242 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 95%-CI Q
## DeFronzo1995 metf plac -1.1141 [-1.2309; -0.9973] 30.89
## Lewin2007 metf plac -1.1141 [-1.2309; -0.9973] 8.79
## Willms1999 acar metf 0.2867 [ 0.0622; 0.5113] 0.05
## Davidson2007 plac rosi 1.2018 [ 1.1084; 1.2953] 0.93
## Wolffenbuttel1999 plac rosi 1.2018 [ 1.1084; 1.2953] 0.80
## Kipnes2001 piog plac -1.0664 [-1.2151; -0.9178] 3.39
## Kerenyi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 16.05
## Hanefeld2004 metf piog -0.0477 [-0.1845; 0.0891] 1.75
## Derosa2004 piog rosi 0.1354 [-0.0249; 0.2957] 0.04
## Baksi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 0.94
## Rosenstock2008 plac rosi 1.2018 [ 1.1084; 1.2953] 0.24
## Zhu2003 plac rosi 1.2018 [ 1.1084; 1.2953] 3.37
## Yang2003 metf rosi 0.0877 [-0.0449; 0.2203] 0.05
## Vongthavaravat2002 rosi sulf -0.7623 [-0.9427; -0.5820] 9.29
## Oyama2008 acar sulf -0.3879 [-0.6095; -0.1662] 0.01
## Costa1997 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## Hermansen2007 plac sita 0.5700 [ 0.3170; 0.8230] 0.00
## Garber2008 plac vild 0.7000 [ 0.4505; 0.9495] 0.00
## Alex1998 metf sulf -0.6746 [-0.8482; -0.5011] 6.62
## Johnston1994 migl plac -0.9439 [-1.1927; -0.6952] 1.23
## Johnston1998a migl plac -0.9439 [-1.1927; -0.6952] 4.35
## Kim2007 metf rosi 0.0877 [-0.0449; 0.2203] 0.14
## Johnston1998b migl plac -0.9439 [-1.1927; -0.6952] 0.87
## Gonzalez-Ortiz2004 metf plac -1.1141 [-1.2309; -0.9973] 2.69
## Stucci1996 benf plac -0.9052 [-1.1543; -0.6561] 3.79
## Moulin2006 benf plac -0.9052 [-1.1543; -0.6561] 0.59
## Willms1999 metf plac -1.1141 [-1.2309; -0.9973] 0.04
## Willms1999 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## leverage
## DeFronzo1995 0.18
## Lewin2007 0.36
## Willms1999 0.09
## Davidson2007 0.11
## Wolffenbuttel1999 0.17
## Kipnes2001 0.36
## Kerenyi2004 0.20
## Hanefeld2004 0.68
## Derosa2004 0.20
## Baksi2004 0.22
## Rosenstock2008 0.04
## Zhu2003 0.09
## Yang2003 0.09
## Vongthavaravat2002 0.41
## Oyama2008 0.53
## Costa1997 0.57
## Hermansen2007 1.00
## Garber2008 1.00
## Alex1998 0.56
## Johnston1994 0.48
## Johnston1998a 0.32
## Kim2007 0.08
## Johnston1998b 0.20
## Gonzalez-Ortiz2004 0.02
## Stucci1996 0.13
## Moulin2006 0.87
## Willms1999 0.02
## Willms1999 0.02
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = ''):
## acar benf metf migl piog plac rosi sita
## acar . 0.0778 0.2867 0.1166 0.2391 -0.8274 0.3745 -0.2574
## benf -0.0778 . 0.2089 0.0387 0.1612 -0.9052 0.2967 -0.3352
## metf -0.2867 -0.2089 . -0.1702 -0.0477 -1.1141 0.0877 -0.5441
## migl -0.1166 -0.0387 0.1702 . 0.1225 -0.9439 0.2579 -0.3739
## piog -0.2391 -0.1612 0.0477 -0.1225 . -1.0664 0.1354 -0.4964
## plac 0.8274 0.9052 1.1141 0.9439 1.0664 . 1.2018 0.5700
## rosi -0.3745 -0.2967 -0.0877 -0.2579 -0.1354 -1.2018 . -0.6318
## sita 0.2574 0.3352 0.5441 0.3739 0.4964 -0.5700 0.6318 .
## sulf 0.3879 0.4657 0.6746 0.5044 0.6269 -0.4395 0.7623 0.1305
## vild 0.1274 0.2052 0.4141 0.2439 0.3664 -0.7000 0.5018 -0.1300
## sulf vild
## acar -0.3879 -0.1274
## benf -0.4657 -0.2052
## metf -0.6746 -0.4141
## migl -0.5044 -0.2439
## piog -0.6269 -0.3664
## plac 0.4395 0.7000
## rosi -0.7623 -0.5018
## sita -0.1305 0.1300
## sulf . 0.2605
## vild -0.2605 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.2497 0.0622 -0.2107 -0.0094 -1.0401 0.1506 -0.5879
## benf -0.4054 . -0.0662 -0.3133 -0.1288 -1.1543 0.0306 -0.6903
## metf -0.5113 -0.4841 . -0.4450 -0.1845 -1.2309 -0.0449 -0.8228
## migl -0.4438 -0.3908 -0.1046 . -0.1673 -1.1927 -0.0078 -0.7287
## piog -0.4876 -0.4513 -0.0891 -0.4123 . -1.2151 -0.0249 -0.7899
## plac 0.6147 0.6561 0.9973 0.6952 0.9178 . 1.1084 0.3170
## rosi -0.5983 -0.5627 -0.2203 -0.5236 -0.2957 -1.2953 . -0.9016
## sita -0.0732 -0.0199 0.2654 0.0191 0.2030 -0.8230 0.3621 .
## sulf 0.1662 0.1588 0.5011 0.1978 0.4178 -0.6188 0.5820 -0.1796
## vild -0.2005 -0.1474 0.1386 -0.1084 0.0760 -0.9495 0.2354 -0.4854
## sulf vild
## acar -0.6095 -0.4552
## benf -0.7726 -0.5577
## metf -0.8482 -0.6896
## migl -0.8111 -0.5962
## piog -0.8361 -0.6569
## plac 0.2602 0.4505
## rosi -0.9427 -0.7683
## sita -0.4406 -0.2254
## sulf . -0.0467
## vild -0.5677 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.4054 0.5113 0.4438 0.4876 -0.6147 0.5983 0.0732 -0.1662
## benf 0.2497 . 0.4841 0.3908 0.4513 -0.6561 0.5627 0.0199 -0.1588
## metf -0.0622 0.0662 . 0.1046 0.0891 -0.9973 0.2203 -0.2654 -0.5011
## migl 0.2107 0.3133 0.4450 . 0.4123 -0.6952 0.5236 -0.0191 -0.1978
## piog 0.0094 0.1288 0.1845 0.1673 . -0.9178 0.2957 -0.2030 -0.4178
## plac 1.0401 1.1543 1.2309 1.1927 1.2151 . 1.2953 0.8230 0.6188
## rosi -0.1506 -0.0306 0.0449 0.0078 0.0249 -1.1084 . -0.3621 -0.5820
## sita 0.5879 0.6903 0.8228 0.7287 0.7899 -0.3170 0.9016 . 0.1796
## sulf 0.6095 0.7726 0.8482 0.8111 0.8361 -0.2602 0.9427 0.4406 .
## vild 0.4552 0.5577 0.6896 0.5962 0.6569 -0.4505 0.7683 0.2254 0.0467
## vild
## acar 0.2005
## benf 0.1474
## metf -0.1386
## migl 0.1084
## piog -0.0760
## plac 0.9495
## rosi -0.2354
## sita 0.4854
## sulf 0.5677
## vild .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
data(Senn2013)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data = Senn2013, sm = "MD")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
netdistance(net1)
## acar benf metf migl piog plac rosi sita sulf vild
## acar 0 2 1 2 2 1 2 2 1 2
## benf 2 0 2 2 2 1 2 2 3 2
## metf 1 2 0 2 1 1 1 2 1 2
## migl 2 2 2 0 2 1 2 2 3 2
## piog 2 2 1 2 0 1 1 2 2 2
## plac 1 1 1 1 1 0 1 1 2 1
## rosi 2 2 1 2 1 1 0 2 1 2
## sita 2 2 2 2 2 1 2 0 3 2
## sulf 1 3 1 3 2 2 1 3 0 3
## vild 2 2 2 2 2 1 2 2 3 0
netdistance(net1$A.matrix)
## acar benf metf migl piog plac rosi sita sulf vild
## acar 0 2 1 2 2 1 2 2 1 2
## benf 2 0 2 2 2 1 2 2 3 2
## metf 1 2 0 2 1 1 1 2 1 2
## migl 2 2 2 0 2 1 2 2 3 2
## piog 2 2 1 2 0 1 1 2 2 2
## plac 1 1 1 1 1 0 1 1 2 1
## rosi 2 2 1 2 1 1 0 2 1 2
## sita 2 2 2 2 2 1 2 0 3 2
## sulf 1 3 1 3 2 2 1 3 0 3
## vild 2 2 2 2 2 1 2 2 3 0
data(Senn2013)
# Generation of an object of class 'netmeta' with reference treatment 'plac'
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data = Senn2013, sm = "MD", reference = "plac")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Network graph with default settings
netgraph(net1)
# Network graph with specified order of the treatments and one highlighted comparison
trts <- c("plac", "benf", "migl", "acar", "sulf",
"metf", "rosi", "piog", "sita", "vild")
netgraph(net1, highlight = "rosi:plac", seq = trts)
# Same network graph using argument 'seq' in netmeta function
net2 <- netmeta(TE, seTE, treat1, treat2, studlab,
data = Senn2013, sm = "MD", reference = "plac",
seq = trts)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
netgraph(net2, highlight = "rosi:plac")
# Network graph optimized, starting from a circle, with multi-arm
# study colored
netgraph(net1, start = "circle", iterate = TRUE, col.multiarm = "purple")
# Network graph optimized, starting from a circle, with multi-arm
# study colored and all intermediate iteration steps visible
netgraph(net1, start = "circle", iterate = TRUE, col.multiarm = "purple",
allfigures = TRUE)
# Network graph optimized, starting from Laplacian eigenvectors, with
# multi-arm study colored
netgraph(net1, start = "eigen", col.multiarm = "purple")
# Network graph optimized, starting from different Laplacian
# eigenvectors, with multi-arm study colored
netgraph(net1, start = "prcomp", col.multiarm = "purple")
# Network graph optimized, starting from random initial layout, with
# multi-arm study colored
netgraph(net1, start = "random", col.multiarm = "purple")
# Network graph without plastic look and one highlighted comparison
netgraph(net1, plastic = FALSE, highlight = "rosi:plac")
# Network graph without plastic look and comparisons with same thickness
netgraph(net1, plastic = FALSE, thickness = FALSE)
# Network graph with changed labels and specified order of the treatments
netgraph(net1, seq = c(1, 3, 5, 2, 9, 4, 7, 6, 8, 10),
labels = LETTERS[1:10])
data(Senn2013)
# Generation of an object of class 'netmeta' with reference treatment 'plac', i.e. placebo
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,data=Senn2013, sm="MD", reference="plac")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Generate a net heat plot based on a fixed effects model
netheat(net1)
# Generate a net heat plot based on a random effects model
netheat(net1, random=TRUE)
# Network meta-analysis of count mortality statistics
data(Woods2010)
p0 <- pairwise(treatment, event = r, n = N,
studlab = author, data = Woods2010, sm = "OR")
net0 <- netmeta(p0)
cilayout(bracket = "(", separator = " - ")
oldopts <- options(width = 100)
# League table for fixed effect model
netleague(net0, digits = 2)
## League table (fixed effect model):
##
## Fluticasone 0.55 (0.16 - 1.85) 1.31 (0.32 - 5.25) 1.92 (0.35 - 10.57)
## 1.81 (0.54 - 6.10) Placebo 2.37 (0.80 - 7.04) 3.49 (0.73 - 16.60)
## 0.77 (0.19 - 3.08) 0.42 (0.14 - 1.26) Salmeterol 1.47 (0.27 - 8.09)
## 0.52 (0.09 - 2.85) 0.29 (0.06 - 1.36) 0.68 (0.12 - 3.72) SFC
# League table for fixed effect and random effects model
netleague(net0, comb.random = TRUE, digits = 2)
## League table (fixed effect model):
##
## Fluticasone 0.55 (0.16 - 1.85) 1.31 (0.32 - 5.25) 1.92 (0.35 - 10.57)
## 1.81 (0.54 - 6.10) Placebo 2.37 (0.80 - 7.04) 3.49 (0.73 - 16.60)
## 0.77 (0.19 - 3.08) 0.42 (0.14 - 1.26) Salmeterol 1.47 (0.27 - 8.09)
## 0.52 (0.09 - 2.85) 0.29 (0.06 - 1.36) 0.68 (0.12 - 3.72) SFC
##
## League table (random effects model):
##
## Fluticasone 0.55 (0.16 - 1.85) 1.31 (0.32 - 5.25) 1.92 (0.35 - 10.57)
## 1.81 (0.54 - 6.10) Placebo 2.37 (0.80 - 7.04) 3.49 (0.73 - 16.60)
## 0.77 (0.19 - 3.08) 0.42 (0.14 - 1.26) Salmeterol 1.47 (0.27 - 8.09)
## 0.52 (0.09 - 2.85) 0.29 (0.06 - 1.36) 0.68 (0.12 - 3.72) SFC
# Change order of treatments according to treatment ranking
netleague(net0, comb.random = TRUE, digits = 2,
seq = netrank(net0))
## League table (fixed effect model):
##
## SFC 0.68 (0.12 - 3.72) 0.52 (0.09 - 2.85) 0.29 (0.06 - 1.36)
## 1.47 (0.27 - 8.09) Salmeterol 0.77 (0.19 - 3.08) 0.42 (0.14 - 1.26)
## 1.92 (0.35 - 10.57) 1.31 (0.32 - 5.25) Fluticasone 0.55 (0.16 - 1.85)
## 3.49 (0.73 - 16.60) 2.37 (0.80 - 7.04) 1.81 (0.54 - 6.10) Placebo
##
## League table (random effects model):
##
## SFC 0.68 (0.12 - 3.72) 0.52 (0.09 - 2.85) 0.29 (0.06 - 1.36)
## 1.47 (0.27 - 8.09) Salmeterol 0.77 (0.19 - 3.08) 0.42 (0.14 - 1.26)
## 1.92 (0.35 - 10.57) 1.31 (0.32 - 5.25) Fluticasone 0.55 (0.16 - 1.85)
## 3.49 (0.73 - 16.60) 2.37 (0.80 - 7.04) 1.81 (0.54 - 6.10) Placebo
print(netrank(net0), comb.random = TRUE)
## P-score (fixed) P-score (random)
## SFC 0.7963 0.7963
## Salmeterol 0.6377 0.6377
## Fluticasone 0.4706 0.4706
## Placebo 0.0954 0.0954
# Use depression dataset
data(Linde2015)
cilayout()
# Define order of treatments
trts <- c("TCA", "SSRI", "SNRI", "NRI",
"Low-dose SARI", "NaSSa", "rMAO-A", "Hypericum",
"Placebo")
# Outcome labels
outcomes <- c("Early response", "Early remission")
# (1) Early response
p1 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(resp1, resp2, resp3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 14 TCA SSRI NA 10 NA 11 0 FALSE
## NA NA 18 TCA Placebo NA 73 NA 73 0 FALSE
## NA NA 21 TCA rMAO-A NA 46 NA 98 0 FALSE
## NA NA 27 SSRI Placebo NA 80 NA 81 0 FALSE
## NA NA 51 TCA rMAO-A NA 71 NA 71 0 FALSE
## NA NA 130 TCA NaSSa NA 35 NA 36 0 FALSE
## NA NA 131 SSRI SNRI NA 697 NA 688 0 FALSE
## NA NA 130 TCA Placebo NA 35 NA 34 0 FALSE
## NA NA 130 NaSSa Placebo NA 36 NA 34 0 FALSE
net1 <- netmeta(p1,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 21 TCA rMAO-A NA NA
## 27 SSRI Placebo NA NA
## 51 TCA rMAO-A NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## 131 SSRI SNRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# (2) Early remission
p2 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(remi1, remi2, remi3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 1 TCA SNRI NA 75 NA 78 0.0 FALSE
## NA NA 11 TCA SSRI NA 108 NA 99 0.0 FALSE
## NA NA 14 TCA SSRI NA 10 NA 11 0.0 FALSE
## NA NA 18 TCA Placebo NA 73 NA 73 0.0 FALSE
## NA NA 20 TCA SSRI NA 55 NA 51 0.0 FALSE
## NA NA 26 SSRI Placebo NA 314 NA 154 0.0 FALSE
## NA NA 53 SSRI NaSSa NA 122 NA 121 0.0 FALSE
## NA NA 56 TCA SSRI NA 92 NA 380 0.0 FALSE
## NA NA 73 Hypericum Placebo NA 55 NA 57 0.0 FALSE
## NA NA 90 TCA SSRI NA 42 NA 42 0.0 FALSE
## NA NA 96 TCA SSRI 0 30 0 29 0.5 FALSE
## NA NA 121 Low-dose SARI NaSSa NA 43 NA 40 0.0 FALSE
## NA NA 130 TCA NaSSa NA 35 NA 36 0.0 FALSE
## NA NA 1 TCA Placebo NA 75 NA 76 0.0 FALSE
## NA NA 11 TCA Placebo NA 108 NA 101 0.0 FALSE
## NA NA 53 SSRI Placebo NA 122 NA 129 0.0 FALSE
## NA NA 130 TCA Placebo NA 35 NA 34 0.0 FALSE
## NA NA 1 SNRI Placebo NA 78 NA 76 0.0 FALSE
## NA NA 11 SSRI Placebo NA 99 NA 101 0.0 FALSE
## NA NA 53 NaSSa Placebo NA 121 NA 129 0.0 FALSE
## NA NA 130 NaSSa Placebo NA 36 NA 34 0.0 FALSE
net2 <- netmeta(p2,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 1 TCA Placebo NA NA
## 1 TCA SNRI NA NA
## 1 SNRI Placebo NA NA
## 11 SSRI Placebo NA NA
## 11 TCA Placebo NA NA
## 11 TCA SSRI NA NA
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 20 TCA SSRI NA NA
## 26 SSRI Placebo NA NA
## 53 NaSSa Placebo NA NA
## 53 SSRI NaSSa NA NA
## 53 SSRI Placebo NA NA
## 56 TCA SSRI NA NA
## 73 Hypericum Placebo NA NA
## 90 TCA SSRI NA NA
## 96 TCA SSRI NA NA
## 121 Low-dose SARI NaSSa NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
options(width = 200)
netleague(net1, digits = 2)
## League table (random effects model):
##
## TCA 1.03 [0.89; 1.18] 0.99 [0.73; 1.36] 1.22 [0.73; 2.04] 0.97 [0.66; 1.42] 1.51 [1.10; 2.08] 1.64 [1.11; 2.42] 0.87 [0.69; 1.10] 1.72 [1.42; 2.09]
## 0.97 [0.84; 1.12] SSRI 0.97 [0.72; 1.30] 1.18 [0.72; 1.94] 0.94 [0.63; 1.40] 1.47 [1.07; 2.02] 1.59 [1.06; 2.40] 0.84 [0.68; 1.05] 1.68 [1.40; 2.01]
## 1.01 [0.74; 1.38] 1.04 [0.77; 1.39] SNRI 1.23 [0.69; 2.18] 0.98 [0.60; 1.59] 1.52 [0.99; 2.34] 1.65 [1.01; 2.71] 0.88 [0.61; 1.25] 1.74 [1.25; 2.42]
## 0.82 [0.49; 1.37] 0.84 [0.52; 1.38] 0.81 [0.46; 1.45] NRI 0.79 [0.42; 1.50] 1.24 [0.69; 2.24] 1.35 [0.71; 2.56] 0.71 [0.42; 1.22] 1.42 [0.84; 2.40]
## 1.03 [0.70; 1.52] 1.06 [0.71; 1.58] 1.03 [0.63; 1.67] 1.26 [0.67; 2.37] Low-dose SARI 1.56 [1.06; 2.30] 1.69 [1.00; 2.88] 0.90 [0.58; 1.40] 1.78 [1.18; 2.70]
## 0.66 [0.48; 0.91] 0.68 [0.50; 0.93] 0.66 [0.43; 1.01] 0.81 [0.45; 1.45] 0.64 [0.43; 0.94] NaSSa 1.08 [0.68; 1.72] 0.57 [0.40; 0.83] 1.14 [0.82; 1.60]
## 0.61 [0.41; 0.90] 0.63 [0.42; 0.95] 0.61 [0.37; 0.99] 0.74 [0.39; 1.41] 0.59 [0.35; 1.00] 0.92 [0.58; 1.47] rMAO-A 0.53 [0.34; 0.83] 1.05 [0.69; 1.62]
## 1.15 [0.91; 1.46] 1.18 [0.95; 1.47] 1.14 [0.80; 1.64] 1.40 [0.82; 2.41] 1.11 [0.72; 1.73] 1.74 [1.20; 2.52] 1.89 [1.20; 2.96] Hypericum 1.99 [1.58; 2.49]
## 0.58 [0.48; 0.70] 0.60 [0.50; 0.71] 0.58 [0.41; 0.80] 0.71 [0.42; 1.19] 0.56 [0.37; 0.85] 0.88 [0.63; 1.23] 0.95 [0.62; 1.46] 0.50 [0.40; 0.63] Placebo
netleague(net1, digits = 2, ci = FALSE)
## League table (random effects model):
##
## TCA 1.03 0.99 1.22 0.97 1.51 1.64 0.87 1.72
## 0.97 SSRI 0.97 1.18 0.94 1.47 1.59 0.84 1.68
## 1.01 1.04 SNRI 1.23 0.98 1.52 1.65 0.88 1.74
## 0.82 0.84 0.81 NRI 0.79 1.24 1.35 0.71 1.42
## 1.03 1.06 1.03 1.26 Low-dose SARI 1.56 1.69 0.90 1.78
## 0.66 0.68 0.66 0.81 0.64 NaSSa 1.08 0.57 1.14
## 0.61 0.63 0.61 0.74 0.59 0.92 rMAO-A 0.53 1.05
## 1.15 1.18 1.14 1.40 1.11 1.74 1.89 Hypericum 1.99
## 0.58 0.60 0.58 0.71 0.56 0.88 0.95 0.50 Placebo
netleague(net2, digits = 2, ci = FALSE)
## League table (random effects model):
##
## TCA 1.05 0.93 1.05 1.01 1.22 1.22 0.95 1.92
## 0.95 SSRI 0.89 1.00 0.96 1.16 1.16 0.90 1.83
## 1.07 1.13 SNRI 1.13 1.09 1.31 1.31 1.02 2.07
## 0.95 1.00 0.89 NRI 0.96 1.16 1.16 0.90 1.83
## 0.99 1.04 0.92 1.04 Low-dose SARI 1.21 1.21 0.94 1.91
## 0.82 0.86 0.76 0.86 0.83 NaSSa 1.00 0.77 1.58
## 0.82 0.86 0.76 0.86 0.83 1.00 rMAO-A 0.77 1.57
## 1.06 1.11 0.98 1.11 1.07 1.29 1.29 Hypericum 2.04
## 0.52 0.55 0.48 0.55 0.52 0.63 0.64 0.49 Placebo
netleague(net1, net2, digits = 2, ci = FALSE)
## League table (random effects model):
##
## TCA 1.05 0.93 1.05 1.01 1.22 1.22 0.95 1.92
## 0.97 SSRI 0.89 1.00 0.96 1.16 1.16 0.90 1.83
## 1.01 1.04 SNRI 1.13 1.09 1.31 1.31 1.02 2.07
## 0.82 0.84 0.81 NRI 0.96 1.16 1.16 0.90 1.83
## 1.03 1.06 1.03 1.26 Low-dose SARI 1.21 1.21 0.94 1.91
## 0.66 0.68 0.66 0.81 0.64 NaSSa 1.00 0.77 1.58
## 0.61 0.63 0.61 0.74 0.59 0.92 rMAO-A 0.77 1.57
## 1.15 1.18 1.14 1.40 1.11 1.74 1.89 Hypericum 2.04
## 0.58 0.60 0.58 0.71 0.56 0.88 0.95 0.50 Placebo
netleague(net1, net2, seq = netrank(net1, small = "bad"), ci = FALSE)
## League table (random effects model):
##
## Hypericum 1.0682 0.9843 1.0579 1.1117 1.1119 1.2919 1.2936 2.0362
## 0.8975 Low-dose SARI 0.9214 0.9904 1.0408 1.0409 1.2094 1.2110 1.9062
## 0.8751 0.9750 SNRI 1.0749 1.1295 1.1297 1.3126 1.3143 2.0688
## 0.8680 0.9671 0.9919 TCA 1.0509 1.0510 1.2211 1.2228 1.9247
## 0.8445 0.9409 0.9650 0.9729 SSRI 1.0001 1.1620 1.1636 1.8315
## 0.7132 0.7946 0.8150 0.8217 0.8445 NRI 1.1619 1.1634 1.8313
## 0.5743 0.6398 0.6562 0.6616 0.6800 0.8052 NaSSa 1.0013 1.5761
## 0.5302 0.5908 0.6059 0.6109 0.6279 0.7435 0.9233 rMAO-A 1.5740
## 0.5033 0.5607 0.5751 0.5798 0.5960 0.7057 0.8764 0.9492 Placebo
netleague(net1, net2, seq = netrank(net2, small = "bad"), ci = FALSE)
## League table (random effects model):
##
## SNRI 1.0160 1.0749 1.0853 1.1297 1.1295 1.3126 1.3143 2.0688
## 1.1428 Hypericum 1.0579 1.0682 1.1119 1.1117 1.2919 1.2936 2.0362
## 0.9919 0.8680 TCA 1.0097 1.0510 1.0509 1.2211 1.2228 1.9247
## 1.0256 0.8975 1.0340 Low-dose SARI 1.0409 1.0408 1.2094 1.2110 1.9062
## 0.8150 0.7132 0.8217 0.7946 NRI 0.9999 1.1619 1.1634 1.8313
## 0.9650 0.8445 0.9729 0.9409 1.1841 SSRI 1.1620 1.1636 1.8315
## 0.6562 0.5743 0.6616 0.6398 0.8052 0.6800 NaSSa 1.0013 1.5761
## 0.6059 0.5302 0.6109 0.5908 0.7435 0.6279 0.9233 rMAO-A 1.5740
## 0.5751 0.5033 0.5798 0.5607 0.7057 0.5960 0.8764 0.9492 Placebo
print(netrank(net1, small = "bad"), comb.random = TRUE)
## P-score
## Hypericum 0.8939
## Low-dose SARI 0.7201
## SNRI 0.6892
## TCA 0.6802
## SSRI 0.6164
## NRI 0.4445
## NaSSa 0.2128
## rMAO-A 0.1521
## Placebo 0.0908
print(netrank(net2, small = "bad"), comb.random = TRUE)
## P-score
## SNRI 0.7681
## Hypericum 0.7453
## TCA 0.6481
## Low-dose SARI 0.6155
## NRI 0.5513
## SSRI 0.5111
## NaSSa 0.3346
## rMAO-A 0.3196
## Placebo 0.0063
options(oldopts)
# Generate a partial order of treatment rankings
np <- netposet(net1, net2, outcomes = outcomes, small.values = rep("bad",2))
plot(np)
data(Senn2013)
# Generation of an object of class 'netmeta' with
# reference treatment 'plac', i.e. placebo based
# on a fixed effects model
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD", reference="plac")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Calculate measures based on a fixed effects model
nm1 <- netmeasures(net1)
# Plot of minimal parallelism versus mean path length
plot(nm1$meanpath, nm1$minpar, pch="",
xlab="Mean path length", ylab="Minimal parallelism")
text(nm1$meanpath, nm1$minpar, names(nm1$meanpath), cex=0.8)
# Generation of an object of class 'netmeta' with
# reference treatment 'plac' based on a random
# effects model
net2 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD", reference="plac", comb.random=TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Calculate measures based on a random effects model
nm2 <- netmeasures(net2)
nm2
## $proportion
## acar:benf acar:metf acar:migl acar:piog acar:plac acar:rosi acar:sita
## 0.0000 0.2812 0.0000 0.0000 0.6526 0.0000 0.0000
## acar:sulf acar:vild benf:metf benf:migl benf:piog benf:plac benf:rosi
## 0.5311 0.0000 0.0000 0.0000 0.0000 1.0000 0.0000
## benf:sita benf:sulf benf:vild metf:migl metf:piog metf:plac metf:rosi
## 0.0000 0.0000 0.0000 0.0000 0.4389 0.5606 0.3387
## metf:sita metf:sulf metf:vild migl:piog migl:plac migl:rosi migl:sita
## 0.0000 0.4509 0.0000 0.0000 1.0000 0.0000 0.0000
## migl:sulf migl:vild piog:plac piog:rosi piog:sita piog:sulf piog:vild
## 0.0000 0.0000 0.3866 0.3528 0.0000 0.0000 0.0000
## plac:rosi plac:sita plac:sulf plac:vild rosi:sita rosi:sulf rosi:vild
## 0.7565 1.0000 0.0000 1.0000 0.0000 0.4348 0.0000
## sita:sulf sita:vild sulf:vild
## 0.0000 0.0000 0.0000
##
## $meanpath
## acar:benf acar:metf acar:migl acar:piog acar:plac acar:rosi acar:sita
## 2.7810 2.0082 2.7810 2.4711 1.7810 2.2137 2.7810
## acar:sulf acar:vild benf:metf benf:migl benf:piog benf:plac benf:rosi
## 1.8193 2.7810 2.5658 2.0000 2.7181 1.0000 2.3064
## benf:sita benf:sulf benf:vild metf:migl metf:piog metf:plac metf:rosi
## 2.0000 3.1121 2.0000 2.5658 1.6909 1.5658 1.7237
## metf:sita metf:sulf metf:vild migl:piog migl:plac migl:rosi migl:sita
## 2.5658 1.8531 2.5658 2.7181 1.0000 2.3064 2.0000
## migl:sulf migl:vild piog:plac piog:rosi piog:sita piog:sulf piog:vild
## 3.1121 2.0000 1.7181 1.8133 2.7181 2.3823 2.7181
## plac:rosi plac:sita plac:sulf plac:vild rosi:sita rosi:sulf rosi:vild
## 1.3064 1.0000 2.1121 1.0000 2.3064 1.9788 2.3064
## sita:sulf sita:vild sulf:vild
## 3.1121 2.0000 3.1121
##
## $minpar
## acar:benf acar:metf acar:migl acar:piog acar:plac acar:rosi acar:sita
## 1.0000 2.5037 1.0000 2.3119 2.1385 1.8162 1.0000
## acar:sulf acar:vild benf:metf benf:migl benf:piog benf:plac benf:rosi
## 1.8827 1.0000 1.0000 1.0000 1.0000 1.0000 1.0000
## benf:sita benf:sulf benf:vild metf:migl metf:piog metf:plac metf:rosi
## 1.0000 1.0000 1.0000 1.0000 2.2786 2.1489 2.1790
## metf:sita metf:sulf metf:vild migl:piog migl:plac migl:rosi migl:sita
## 1.0000 2.2177 1.0000 1.0000 1.0000 1.0000 1.0000
## migl:sulf migl:vild piog:plac piog:rosi piog:sita piog:sulf piog:vild
## 1.0000 1.0000 2.5869 2.3621 1.0000 2.5649 1.0000
## plac:rosi plac:sita plac:sulf plac:vild rosi:sita rosi:sulf rosi:vild
## 1.3219 1.0000 2.3831 1.0000 1.0000 2.2998 1.0000
## sita:sulf sita:vild sulf:vild
## 1.0000 1.0000 1.0000
##
## $minpar.study
## acar:benf acar:metf acar:migl acar:piog acar:plac acar:rosi acar:sita
## 1.5564 2.5037 2.1385 2.3119 2.1385 2.3083 1.0000
## acar:sulf acar:vild benf:metf benf:migl benf:piog benf:plac benf:rosi
## 1.8827 1.0000 1.5564 1.5564 1.5564 1.5564 1.5564
## benf:sita benf:sulf benf:vild metf:migl metf:piog metf:plac metf:rosi
## 1.0000 1.5564 1.0000 2.6537 2.2786 4.9818 5.8226
## metf:sita metf:sulf metf:vild migl:piog migl:plac migl:rosi migl:sita
## 1.0000 2.2177 1.0000 2.5869 2.6537 2.6537 1.0000
## migl:sulf migl:vild piog:plac piog:rosi piog:sita piog:sulf piog:vild
## 2.6414 1.0000 2.5869 2.8342 1.0000 2.5649 1.0000
## plac:rosi plac:sita plac:sulf plac:vild rosi:sita rosi:sulf rosi:vild
## 7.2857 1.0000 2.6414 1.0000 1.0000 2.2998 1.0000
## sita:sulf sita:vild sulf:vild
## 1.0000 1.0000 1.0000
##
## $H.tilde
## acar:metf:plac acar:plac acar:sulf benf:plac
## acar:benf 2.577526e-01 4.676222e-01 2.746251e-01 1.000000e+00
## acar:metf 2.926239e-01 3.994096e-01 3.079666e-01 3.391488e-16
## acar:migl 2.577526e-01 4.676222e-01 2.746251e-01 1.695744e-16
## acar:piog 2.703522e-01 4.325446e-01 2.971032e-01 2.543616e-16
## acar:plac 2.577526e-01 4.676222e-01 2.746251e-01 2.543616e-16
## acar:rosi 2.573856e-01 4.332202e-01 3.093942e-01 1.695744e-16
## acar:sita 2.577526e-01 4.676222e-01 2.746251e-01 1.695744e-16
## acar:sulf 1.868165e-01 2.820373e-01 5.311462e-01 8.478720e-17
## acar:vild 2.577526e-01 4.676222e-01 2.746251e-01 8.478720e-17
## benf:metf 1.189217e-01 6.821266e-02 3.334144e-02 1.000000e+00
## benf:migl 0.000000e+00 0.000000e+00 0.000000e+00 1.000000e+00
## benf:piog 5.090128e-02 3.507765e-02 2.247806e-02 1.000000e+00
## benf:plac 3.112961e-17 5.369698e-17 0.000000e+00 1.000000e+00
## benf:rosi 2.582341e-02 3.440204e-02 3.476910e-02 1.000000e+00
## benf:sita 8.589556e-17 0.000000e+00 5.228577e-17 1.000000e+00
## benf:sulf 7.568026e-02 1.855849e-01 2.565210e-01 1.000000e+00
## benf:vild 3.112961e-17 5.369698e-17 5.228577e-17 1.000000e+00
## metf:migl 1.189217e-01 6.821266e-02 3.334144e-02 1.695744e-16
## metf:piog 6.802043e-02 3.313501e-02 1.086339e-02 8.478720e-17
## metf:plac 1.189217e-01 6.821266e-02 3.334144e-02 8.478720e-17
## metf:rosi 9.346535e-02 3.381062e-02 1.427659e-03 1.695744e-16
## metf:sita 1.189217e-01 6.821266e-02 3.334144e-02 1.695744e-16
## metf:sulf 1.141776e-01 1.173722e-01 2.231796e-01 2.543616e-16
## metf:vild 1.189217e-01 6.821266e-02 3.334144e-02 2.543616e-16
## migl:piog 5.090128e-02 3.507765e-02 2.247806e-02 8.478720e-17
## migl:plac 3.112961e-17 5.369698e-17 0.000000e+00 8.478720e-17
## migl:rosi 2.582341e-02 3.440204e-02 3.476910e-02 0.000000e+00
## migl:sita 8.589556e-17 0.000000e+00 5.228577e-17 0.000000e+00
## migl:sulf 7.568026e-02 1.855849e-01 2.565210e-01 8.478720e-17
## migl:vild 3.112961e-17 5.369698e-17 5.228577e-17 8.478720e-17
## piog:plac 5.090128e-02 3.507765e-02 2.247806e-02 0.000000e+00
## piog:rosi 2.544493e-02 6.756137e-04 1.229105e-02 8.478720e-17
## piog:sita 5.090128e-02 3.507765e-02 2.247806e-02 8.478720e-17
## piog:sulf 8.353576e-02 1.505072e-01 2.340430e-01 1.695744e-16
## piog:vild 5.090128e-02 3.507765e-02 2.247806e-02 1.695744e-16
## plac:rosi 2.582341e-02 3.440204e-02 3.476910e-02 8.478720e-17
## plac:sita 1.065062e-16 5.369698e-17 5.228577e-17 8.478720e-17
## plac:sulf 7.568026e-02 1.855849e-01 2.565210e-01 1.695744e-16
## plac:vild 0.000000e+00 0.000000e+00 5.228577e-17 1.695744e-16
## rosi:sita 2.582341e-02 3.440204e-02 3.476910e-02 0.000000e+00
## rosi:sulf 7.056910e-02 1.511828e-01 2.217519e-01 8.478720e-17
## rosi:vild 2.582341e-02 3.440204e-02 3.476910e-02 8.478720e-17
## sita:sulf 7.568026e-02 1.855849e-01 2.565210e-01 8.478720e-17
## sita:vild 1.065062e-16 5.369698e-17 0.000000e+00 8.478720e-17
## sulf:vild 7.568026e-02 1.855849e-01 2.565210e-01 0.000000e+00
## metf:piog metf:plac metf:rosi metf:sulf migl:plac
## acar:benf 3.693606e-02 1.723667e-01 5.423948e-02 1.235764e-01 2.583740e-16
## acar:metf 7.707516e-02 2.929961e-01 1.614363e-01 2.096049e-01 5.167479e-16
## acar:migl 3.693606e-02 1.723667e-01 5.423948e-02 1.235764e-01 1.000000e+00
## acar:piog 3.617969e-01 3.454861e-02 1.217173e-02 1.485529e-01 3.875610e-16
## acar:plac 3.693606e-02 1.723667e-01 5.423948e-02 1.235764e-01 3.875610e-16
## acar:rosi 4.355917e-02 4.672869e-02 1.772435e-01 1.021092e-01 2.583740e-16
## acar:sita 3.693606e-02 1.723667e-01 5.423948e-02 1.235764e-01 2.583740e-16
## acar:sulf 1.243641e-02 8.652449e-02 2.352086e-03 2.413187e-01 1.291870e-16
## acar:vild 3.693606e-02 1.723667e-01 5.423948e-02 1.235764e-01 1.291870e-16
## benf:metf 1.140112e-01 4.653628e-01 2.156758e-01 8.602849e-02 2.583740e-16
## benf:migl 5.985679e-17 0.000000e+00 0.000000e+00 0.000000e+00 1.000000e+00
## benf:piog 3.248608e-01 2.069153e-01 4.206775e-02 2.497653e-02 1.291870e-16
## benf:plac 0.000000e+00 1.356870e-16 8.614147e-17 0.000000e+00 1.291870e-16
## benf:rosi 6.623109e-03 1.256380e-01 1.230041e-01 2.146716e-02 0.000000e+00
## benf:sita 0.000000e+00 2.713740e-16 1.722829e-16 5.653535e-17 0.000000e+00
## benf:sulf 4.937247e-02 2.588912e-01 5.188739e-02 3.648951e-01 1.291870e-16
## benf:vild 0.000000e+00 1.356870e-16 1.722829e-16 5.653535e-17 1.291870e-16
## metf:migl 1.140112e-01 4.653628e-01 2.156758e-01 8.602849e-02 1.000000e+00
## metf:piog 4.388721e-01 2.584475e-01 1.736080e-01 6.105196e-02 1.291870e-16
## metf:plac 1.140112e-01 4.653628e-01 2.156758e-01 8.602849e-02 1.291870e-16
## metf:rosi 1.206343e-01 3.397248e-01 3.386798e-01 1.074957e-01 2.583740e-16
## metf:sita 1.140112e-01 4.653628e-01 2.156758e-01 8.602849e-02 2.583740e-16
## metf:sulf 6.463874e-02 2.064717e-01 1.637884e-01 4.509236e-01 3.875610e-16
## metf:vild 1.140112e-01 4.653628e-01 2.156758e-01 8.602849e-02 3.875610e-16
## migl:piog 3.248608e-01 2.069153e-01 4.206775e-02 2.497653e-02 1.000000e+00
## migl:plac 5.985679e-17 1.356870e-16 8.614147e-17 0.000000e+00 1.000000e+00
## migl:rosi 6.623109e-03 1.256380e-01 1.230041e-01 2.146716e-02 1.000000e+00
## migl:sita 5.985679e-17 2.713740e-16 1.722829e-16 5.653535e-17 1.000000e+00
## migl:sulf 4.937247e-02 2.588912e-01 5.188739e-02 3.648951e-01 1.000000e+00
## migl:vild 5.985679e-17 1.356870e-16 1.722829e-16 5.653535e-17 1.000000e+00
## piog:plac 3.248608e-01 2.069153e-01 4.206775e-02 2.497653e-02 0.000000e+00
## piog:rosi 3.182377e-01 8.127730e-02 1.650718e-01 4.644369e-02 1.291870e-16
## piog:sita 3.248608e-01 2.069153e-01 4.206775e-02 2.497653e-02 1.291870e-16
## piog:sulf 3.742333e-01 5.197589e-02 9.819641e-03 3.898717e-01 2.583740e-16
## piog:vild 3.248608e-01 2.069153e-01 4.206775e-02 2.497653e-02 2.583740e-16
## plac:rosi 6.623109e-03 1.256380e-01 1.230041e-01 2.146716e-02 1.291870e-16
## plac:sita 0.000000e+00 4.070610e-16 2.584244e-16 5.653535e-17 1.291870e-16
## plac:sulf 4.937247e-02 2.588912e-01 5.188739e-02 3.648951e-01 2.583740e-16
## plac:vild 0.000000e+00 0.000000e+00 8.614147e-17 5.653535e-17 2.583740e-16
## rosi:sita 6.623109e-03 1.256380e-01 1.230041e-01 2.146716e-02 0.000000e+00
## rosi:sulf 5.599558e-02 1.332532e-01 1.748915e-01 3.434280e-01 1.291870e-16
## rosi:vild 6.623109e-03 1.256380e-01 1.230041e-01 2.146716e-02 1.291870e-16
## sita:sulf 4.937247e-02 2.588912e-01 5.188739e-02 3.648951e-01 1.291870e-16
## sita:vild 0.000000e+00 4.070610e-16 3.445659e-16 1.130707e-16 1.291870e-16
## sulf:vild 4.937247e-02 2.588912e-01 5.188739e-02 3.648951e-01 0.000000e+00
## piog:plac piog:rosi plac:rosi plac:sita plac:vild
## acar:benf 3.632228e-02 6.137752e-04 2.059020e-01 5.534126e-17 0.000000e+00
## acar:metf 4.846815e-02 2.860701e-02 9.168160e-02 1.106825e-16 0.000000e+00
## acar:migl 3.632228e-02 6.137752e-04 2.059020e-01 1.106825e-16 0.000000e+00
## acar:piog 3.502384e-01 2.879647e-01 1.272427e-01 1.660238e-16 0.000000e+00
## acar:plac 3.632228e-02 6.137752e-04 2.059020e-01 1.106825e-16 0.000000e+00
## acar:rosi 2.131442e-02 6.487359e-02 5.505979e-01 0.000000e+00 0.000000e+00
## acar:sita 3.632228e-02 6.137752e-04 2.059020e-01 1.000000e+00 0.000000e+00
## acar:sulf 2.390384e-02 1.146743e-02 2.137156e-01 0.000000e+00 5.554572e-17
## acar:vild 3.632228e-02 6.137752e-04 2.059020e-01 0.000000e+00 1.000000e+00
## benf:metf 8.479043e-02 2.922078e-02 2.975836e-01 5.534126e-17 0.000000e+00
## benf:migl 5.560227e-17 4.878213e-17 0.000000e+00 5.534126e-17 0.000000e+00
## benf:piog 3.865607e-01 2.885784e-01 3.331447e-01 1.106825e-16 0.000000e+00
## benf:plac 5.560227e-17 4.878213e-17 0.000000e+00 5.534126e-17 0.000000e+00
## benf:rosi 5.763670e-02 6.425981e-02 7.564999e-01 5.534126e-17 0.000000e+00
## benf:sita 1.112045e-16 9.756426e-17 0.000000e+00 1.000000e+00 0.000000e+00
## benf:sulf 6.022613e-02 1.085365e-02 4.196176e-01 5.534126e-17 5.554572e-17
## benf:vild 5.560227e-17 9.756426e-17 3.213855e-16 5.534126e-17 1.000000e+00
## metf:migl 8.479043e-02 2.922078e-02 2.975836e-01 0.000000e+00 0.000000e+00
## metf:piog 3.017703e-01 2.593576e-01 3.556107e-02 5.534126e-17 0.000000e+00
## metf:plac 8.479043e-02 2.922078e-02 2.975836e-01 0.000000e+00 0.000000e+00
## metf:rosi 2.715373e-02 9.348059e-02 4.589163e-01 1.106825e-16 0.000000e+00
## metf:sita 8.479043e-02 2.922078e-02 2.975836e-01 1.000000e+00 0.000000e+00
## metf:sulf 2.456431e-02 4.007444e-02 1.220340e-01 1.106825e-16 5.554572e-17
## metf:vild 8.479043e-02 2.922078e-02 2.975836e-01 1.106825e-16 1.000000e+00
## migl:piog 3.865607e-01 2.885784e-01 3.331447e-01 5.534126e-17 0.000000e+00
## migl:plac 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
## migl:rosi 5.763670e-02 6.425981e-02 7.564999e-01 1.106825e-16 0.000000e+00
## migl:sita 1.668068e-16 1.463464e-16 0.000000e+00 1.000000e+00 0.000000e+00
## migl:sulf 6.022613e-02 1.085365e-02 4.196176e-01 1.106825e-16 5.554572e-17
## migl:vild 0.000000e+00 4.878213e-17 3.213855e-16 1.106825e-16 1.000000e+00
## piog:plac 3.865607e-01 2.885784e-01 3.331447e-01 5.534126e-17 0.000000e+00
## piog:rosi 3.289240e-01 3.528382e-01 4.233552e-01 1.660238e-16 0.000000e+00
## piog:sita 3.865607e-01 2.885784e-01 3.331447e-01 1.000000e+00 0.000000e+00
## piog:sulf 3.263346e-01 2.994321e-01 8.647291e-02 1.660238e-16 5.554572e-17
## piog:vild 3.865607e-01 2.885784e-01 3.331447e-01 1.660238e-16 1.000000e+00
## plac:rosi 5.763670e-02 6.425981e-02 7.564999e-01 1.106825e-16 0.000000e+00
## plac:sita 1.668068e-16 1.463464e-16 0.000000e+00 1.000000e+00 0.000000e+00
## plac:sulf 6.022613e-02 1.085365e-02 4.196176e-01 1.106825e-16 5.554572e-17
## plac:vild 0.000000e+00 4.878213e-17 3.213855e-16 1.106825e-16 1.000000e+00
## rosi:sita 5.763670e-02 6.425981e-02 7.564999e-01 1.000000e+00 0.000000e+00
## rosi:sulf 2.589425e-03 5.340616e-02 3.368823e-01 0.000000e+00 5.554572e-17
## rosi:vild 5.763670e-02 6.425981e-02 7.564999e-01 0.000000e+00 1.000000e+00
## sita:sulf 6.022613e-02 1.085365e-02 4.196176e-01 1.000000e+00 5.554572e-17
## sita:vild 1.668068e-16 1.951285e-16 3.213855e-16 1.000000e+00 1.000000e+00
## sulf:vild 6.022613e-02 1.085365e-02 4.196176e-01 0.000000e+00 1.000000e+00
## rosi:sulf
## acar:benf 1.510487e-01
## acar:metf 9.836169e-02
## acar:migl 1.510487e-01
## acar:piog 1.485503e-01
## acar:plac 1.510487e-01
## acar:rosi 2.072850e-01
## acar:sita 1.510487e-01
## acar:sulf 2.275351e-01
## acar:vild 1.510487e-01
## benf:metf 5.268705e-02
## benf:migl 0.000000e+00
## benf:piog 2.498476e-03
## benf:plac 5.364935e-17
## benf:rosi 5.623626e-02
## benf:sita 5.364935e-17
## benf:sulf 3.785838e-01
## benf:vild 5.364935e-17
## metf:migl 5.268705e-02
## metf:piog 5.018857e-02
## metf:plac 5.268705e-02
## metf:rosi 1.089233e-01
## metf:sita 5.268705e-02
## metf:sulf 3.258968e-01
## metf:vild 5.268705e-02
## migl:piog 2.498476e-03
## migl:plac 5.364935e-17
## migl:rosi 5.623626e-02
## migl:sita 5.364935e-17
## migl:sulf 3.785838e-01
## migl:vild 5.364935e-17
## piog:plac 2.498476e-03
## piog:rosi 5.873474e-02
## piog:sita 2.498476e-03
## piog:sulf 3.760853e-01
## piog:vild 2.498476e-03
## plac:rosi 5.623626e-02
## plac:sita 1.072987e-16
## plac:sulf 3.785838e-01
## plac:vild 0.000000e+00
## rosi:sita 5.623626e-02
## rosi:sulf 4.348201e-01
## rosi:vild 5.623626e-02
## sita:sulf 3.785838e-01
## sita:vild 1.072987e-16
## sulf:vild 3.785838e-01
data(Senn2013)
# Fixed effect model (default)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
net1
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.1414 0.1414 2
## Lewin2007 metf plac -0.82 0.0992 0.0992 2
## Willms1999 acar metf 0.20 0.3579 0.3884 3 *
## Davidson2007 plac rosi 1.34 0.1435 0.1435 2
## Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141 2
## Kipnes2001 piog plac -1.30 0.1268 0.1268 2
## Kerenyi2004 plac rosi 0.77 0.1078 0.1078 2
## Hanefeld2004 metf piog -0.16 0.0849 0.0849 2
## Derosa2004 piog rosi 0.10 0.1831 0.1831 2
## Baksi2004 plac rosi 1.30 0.1014 0.1014 2
## Rosenstock2008 plac rosi 1.09 0.2263 0.2263 2
## Zhu2003 plac rosi 1.50 0.1624 0.1624 2
## Yang2003 metf rosi 0.14 0.2239 0.2239 2
## Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436 2
## Oyama2008 acar sulf -0.40 0.1549 0.1549 2
## Costa1997 acar plac -0.80 0.1432 0.1432 2
## Hermansen2007 plac sita 0.57 0.1291 0.1291 2
## Garber2008 plac vild 0.70 0.1273 0.1273 2
## Alex1998 metf sulf -0.37 0.1184 0.1184 2
## Johnston1994 migl plac -0.74 0.1839 0.1839 2
## Johnston1998a migl plac -1.41 0.2235 0.2235 2
## Kim2007 metf rosi 0.00 0.2339 0.2339 2
## Johnston1998b migl plac -0.68 0.2828 0.2828 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356 2
## Stucci1996 benf plac -0.23 0.3467 0.3467 2
## Moulin2006 benf plac -1.01 0.1366 0.1366 2
## Willms1999 metf plac -1.20 0.3758 0.4125 3 *
## Willms1999 acar plac -1.00 0.4669 0.8242 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q
## DeFronzo1995 metf plac -1.1141 [-1.2309; -0.9973] 30.89
## Lewin2007 metf plac -1.1141 [-1.2309; -0.9973] 8.79
## Willms1999 acar metf 0.2867 [ 0.0622; 0.5113] 0.05
## Davidson2007 plac rosi 1.2018 [ 1.1084; 1.2953] 0.93
## Wolffenbuttel1999 plac rosi 1.2018 [ 1.1084; 1.2953] 0.80
## Kipnes2001 piog plac -1.0664 [-1.2151; -0.9178] 3.39
## Kerenyi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 16.05
## Hanefeld2004 metf piog -0.0477 [-0.1845; 0.0891] 1.75
## Derosa2004 piog rosi 0.1354 [-0.0249; 0.2957] 0.04
## Baksi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 0.94
## Rosenstock2008 plac rosi 1.2018 [ 1.1084; 1.2953] 0.24
## Zhu2003 plac rosi 1.2018 [ 1.1084; 1.2953] 3.37
## Yang2003 metf rosi 0.0877 [-0.0449; 0.2203] 0.05
## Vongthavaravat2002 rosi sulf -0.7623 [-0.9427; -0.5820] 9.29
## Oyama2008 acar sulf -0.3879 [-0.6095; -0.1662] 0.01
## Costa1997 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## Hermansen2007 plac sita 0.5700 [ 0.3170; 0.8230] 0.00
## Garber2008 plac vild 0.7000 [ 0.4505; 0.9495] 0.00
## Alex1998 metf sulf -0.6746 [-0.8482; -0.5011] 6.62
## Johnston1994 migl plac -0.9439 [-1.1927; -0.6952] 1.23
## Johnston1998a migl plac -0.9439 [-1.1927; -0.6952] 4.35
## Kim2007 metf rosi 0.0877 [-0.0449; 0.2203] 0.14
## Johnston1998b migl plac -0.9439 [-1.1927; -0.6952] 0.87
## Gonzalez-Ortiz2004 metf plac -1.1141 [-1.2309; -0.9973] 2.69
## Stucci1996 benf plac -0.9052 [-1.1543; -0.6561] 3.79
## Moulin2006 benf plac -0.9052 [-1.1543; -0.6561] 0.59
## Willms1999 metf plac -1.1141 [-1.2309; -0.9973] 0.04
## Willms1999 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## leverage
## DeFronzo1995 0.18
## Lewin2007 0.36
## Willms1999 0.09
## Davidson2007 0.11
## Wolffenbuttel1999 0.17
## Kipnes2001 0.36
## Kerenyi2004 0.20
## Hanefeld2004 0.68
## Derosa2004 0.20
## Baksi2004 0.22
## Rosenstock2008 0.04
## Zhu2003 0.09
## Yang2003 0.09
## Vongthavaravat2002 0.41
## Oyama2008 0.53
## Costa1997 0.57
## Hermansen2007 1.00
## Garber2008 1.00
## Alex1998 0.56
## Johnston1994 0.48
## Johnston1998a 0.32
## Kim2007 0.08
## Johnston1998b 0.20
## Gonzalez-Ortiz2004 0.02
## Stucci1996 0.13
## Moulin2006 0.87
## Willms1999 0.02
## Willms1999 0.02
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## acar benf metf migl piog plac rosi sita
## acar . 0.0778 0.2867 0.1166 0.2391 -0.8274 0.3745 -0.2574
## benf -0.0778 . 0.2089 0.0387 0.1612 -0.9052 0.2967 -0.3352
## metf -0.2867 -0.2089 . -0.1702 -0.0477 -1.1141 0.0877 -0.5441
## migl -0.1166 -0.0387 0.1702 . 0.1225 -0.9439 0.2579 -0.3739
## piog -0.2391 -0.1612 0.0477 -0.1225 . -1.0664 0.1354 -0.4964
## plac 0.8274 0.9052 1.1141 0.9439 1.0664 . 1.2018 0.5700
## rosi -0.3745 -0.2967 -0.0877 -0.2579 -0.1354 -1.2018 . -0.6318
## sita 0.2574 0.3352 0.5441 0.3739 0.4964 -0.5700 0.6318 .
## sulf 0.3879 0.4657 0.6746 0.5044 0.6269 -0.4395 0.7623 0.1305
## vild 0.1274 0.2052 0.4141 0.2439 0.3664 -0.7000 0.5018 -0.1300
## sulf vild
## acar -0.3879 -0.1274
## benf -0.4657 -0.2052
## metf -0.6746 -0.4141
## migl -0.5044 -0.2439
## piog -0.6269 -0.3664
## plac 0.4395 0.7000
## rosi -0.7623 -0.5018
## sita -0.1305 0.1300
## sulf . 0.2605
## vild -0.2605 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.2497 0.0622 -0.2107 -0.0094 -1.0401 0.1506 -0.5879
## benf -0.4054 . -0.0662 -0.3133 -0.1288 -1.1543 0.0306 -0.6903
## metf -0.5113 -0.4841 . -0.4450 -0.1845 -1.2309 -0.0449 -0.8228
## migl -0.4438 -0.3908 -0.1046 . -0.1673 -1.1927 -0.0078 -0.7287
## piog -0.4876 -0.4513 -0.0891 -0.4123 . -1.2151 -0.0249 -0.7899
## plac 0.6147 0.6561 0.9973 0.6952 0.9178 . 1.1084 0.3170
## rosi -0.5983 -0.5627 -0.2203 -0.5236 -0.2957 -1.2953 . -0.9016
## sita -0.0732 -0.0199 0.2654 0.0191 0.2030 -0.8230 0.3621 .
## sulf 0.1662 0.1588 0.5011 0.1978 0.4178 -0.6188 0.5820 -0.1796
## vild -0.2005 -0.1474 0.1386 -0.1084 0.0760 -0.9495 0.2354 -0.4854
## sulf vild
## acar -0.6095 -0.4552
## benf -0.7726 -0.5577
## metf -0.8482 -0.6896
## migl -0.8111 -0.5962
## piog -0.8361 -0.6569
## plac 0.2602 0.4505
## rosi -0.9427 -0.7683
## sita -0.4406 -0.2254
## sulf . -0.0467
## vild -0.5677 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.4054 0.5113 0.4438 0.4876 -0.6147 0.5983 0.0732 -0.1662
## benf 0.2497 . 0.4841 0.3908 0.4513 -0.6561 0.5627 0.0199 -0.1588
## metf -0.0622 0.0662 . 0.1046 0.0891 -0.9973 0.2203 -0.2654 -0.5011
## migl 0.2107 0.3133 0.4450 . 0.4123 -0.6952 0.5236 -0.0191 -0.1978
## piog 0.0094 0.1288 0.1845 0.1673 . -0.9178 0.2957 -0.2030 -0.4178
## plac 1.0401 1.1543 1.2309 1.1927 1.2151 . 1.2953 0.8230 0.6188
## rosi -0.1506 -0.0306 0.0449 0.0078 0.0249 -1.1084 . -0.3621 -0.5820
## sita 0.5879 0.6903 0.8228 0.7287 0.7899 -0.3170 0.9016 . 0.1796
## sulf 0.6095 0.7726 0.8482 0.8111 0.8361 -0.2602 0.9427 0.4406 .
## vild 0.4552 0.5577 0.6896 0.5962 0.6569 -0.4505 0.7683 0.2254 0.0467
## vild
## acar 0.2005
## benf 0.1474
## metf -0.1386
## migl 0.1084
## piog -0.0760
## plac 0.9495
## rosi -0.2354
## sita 0.4854
## sulf 0.5677
## vild .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
net1$Q.decomp
## treat1 treat2 Q df pval.Q
## 1 acar metf 0.00000000 0 1.000000e+00
## 2 acar plac 0.05715942 1 8.110433e-01
## 3 acar sulf 0.00000000 0 1.000000e+00
## 4 benf plac 4.38137713 1 3.633363e-02
## 5 metf piog 0.00000000 0 1.000000e+00
## 6 metf plac 42.17838677 3 3.677205e-09
## 7 metf rosi 0.18695080 1 6.654667e-01
## 8 metf sulf 0.00000000 0 1.000000e+00
## 9 migl plac 6.44927778 2 3.977014e-02
## 10 piog plac 0.00000000 0 1.000000e+00
## 11 piog rosi 0.00000000 0 1.000000e+00
## 12 plac rosi 21.27336195 5 7.191616e-04
## 13 plac sita 0.00000000 0 1.000000e+00
## 14 plac vild 0.00000000 0 1.000000e+00
## 15 rosi sulf 0.00000000 0 1.000000e+00
# Comparison with reference group
print(net1, reference="plac")
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.1414 0.1414 2
## Lewin2007 metf plac -0.82 0.0992 0.0992 2
## Willms1999 acar metf 0.20 0.3579 0.3884 3 *
## Davidson2007 plac rosi 1.34 0.1435 0.1435 2
## Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141 2
## Kipnes2001 piog plac -1.30 0.1268 0.1268 2
## Kerenyi2004 plac rosi 0.77 0.1078 0.1078 2
## Hanefeld2004 metf piog -0.16 0.0849 0.0849 2
## Derosa2004 piog rosi 0.10 0.1831 0.1831 2
## Baksi2004 plac rosi 1.30 0.1014 0.1014 2
## Rosenstock2008 plac rosi 1.09 0.2263 0.2263 2
## Zhu2003 plac rosi 1.50 0.1624 0.1624 2
## Yang2003 metf rosi 0.14 0.2239 0.2239 2
## Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436 2
## Oyama2008 acar sulf -0.40 0.1549 0.1549 2
## Costa1997 acar plac -0.80 0.1432 0.1432 2
## Hermansen2007 plac sita 0.57 0.1291 0.1291 2
## Garber2008 plac vild 0.70 0.1273 0.1273 2
## Alex1998 metf sulf -0.37 0.1184 0.1184 2
## Johnston1994 migl plac -0.74 0.1839 0.1839 2
## Johnston1998a migl plac -1.41 0.2235 0.2235 2
## Kim2007 metf rosi 0.00 0.2339 0.2339 2
## Johnston1998b migl plac -0.68 0.2828 0.2828 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356 2
## Stucci1996 benf plac -0.23 0.3467 0.3467 2
## Moulin2006 benf plac -1.01 0.1366 0.1366 2
## Willms1999 metf plac -1.20 0.3758 0.4125 3 *
## Willms1999 acar plac -1.00 0.4669 0.8242 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q
## DeFronzo1995 metf plac -1.1141 [-1.2309; -0.9973] 30.89
## Lewin2007 metf plac -1.1141 [-1.2309; -0.9973] 8.79
## Willms1999 acar metf 0.2867 [ 0.0622; 0.5113] 0.05
## Davidson2007 plac rosi 1.2018 [ 1.1084; 1.2953] 0.93
## Wolffenbuttel1999 plac rosi 1.2018 [ 1.1084; 1.2953] 0.80
## Kipnes2001 piog plac -1.0664 [-1.2151; -0.9178] 3.39
## Kerenyi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 16.05
## Hanefeld2004 metf piog -0.0477 [-0.1845; 0.0891] 1.75
## Derosa2004 piog rosi 0.1354 [-0.0249; 0.2957] 0.04
## Baksi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 0.94
## Rosenstock2008 plac rosi 1.2018 [ 1.1084; 1.2953] 0.24
## Zhu2003 plac rosi 1.2018 [ 1.1084; 1.2953] 3.37
## Yang2003 metf rosi 0.0877 [-0.0449; 0.2203] 0.05
## Vongthavaravat2002 rosi sulf -0.7623 [-0.9427; -0.5820] 9.29
## Oyama2008 acar sulf -0.3879 [-0.6095; -0.1662] 0.01
## Costa1997 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## Hermansen2007 plac sita 0.5700 [ 0.3170; 0.8230] 0.00
## Garber2008 plac vild 0.7000 [ 0.4505; 0.9495] 0.00
## Alex1998 metf sulf -0.6746 [-0.8482; -0.5011] 6.62
## Johnston1994 migl plac -0.9439 [-1.1927; -0.6952] 1.23
## Johnston1998a migl plac -0.9439 [-1.1927; -0.6952] 4.35
## Kim2007 metf rosi 0.0877 [-0.0449; 0.2203] 0.14
## Johnston1998b migl plac -0.9439 [-1.1927; -0.6952] 0.87
## Gonzalez-Ortiz2004 metf plac -1.1141 [-1.2309; -0.9973] 2.69
## Stucci1996 benf plac -0.9052 [-1.1543; -0.6561] 3.79
## Moulin2006 benf plac -0.9052 [-1.1543; -0.6561] 0.59
## Willms1999 metf plac -1.1141 [-1.2309; -0.9973] 0.04
## Willms1999 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## leverage
## DeFronzo1995 0.18
## Lewin2007 0.36
## Willms1999 0.09
## Davidson2007 0.11
## Wolffenbuttel1999 0.17
## Kipnes2001 0.36
## Kerenyi2004 0.20
## Hanefeld2004 0.68
## Derosa2004 0.20
## Baksi2004 0.22
## Rosenstock2008 0.04
## Zhu2003 0.09
## Yang2003 0.09
## Vongthavaravat2002 0.41
## Oyama2008 0.53
## Costa1997 0.57
## Hermansen2007 1.00
## Garber2008 1.00
## Alex1998 0.56
## Johnston1994 0.48
## Johnston1998a 0.32
## Kim2007 0.08
## Johnston1998b 0.20
## Gonzalez-Ortiz2004 0.02
## Stucci1996 0.13
## Moulin2006 0.87
## Willms1999 0.02
## Willms1999 0.02
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD', comparison: other treatments vs 'plac'):
## MD 95%-CI
## acar -0.8274 [-1.0401; -0.6147]
## benf -0.9052 [-1.1543; -0.6561]
## metf -1.1141 [-1.2309; -0.9973]
## migl -0.9439 [-1.1927; -0.6952]
## piog -1.0664 [-1.2151; -0.9178]
## plac . .
## rosi -1.2018 [-1.2953; -1.1084]
## sita -0.5700 [-0.8230; -0.3170]
## sulf -0.4395 [-0.6188; -0.2602]
## vild -0.7000 [-0.9495; -0.4505]
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
# Random effects model
net2 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD", comb.random=TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
net2
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.1414 0.1414 2
## Lewin2007 metf plac -0.82 0.0992 0.0992 2
## Willms1999 acar metf 0.20 0.3579 0.3884 3 *
## Davidson2007 plac rosi 1.34 0.1435 0.1435 2
## Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141 2
## Kipnes2001 piog plac -1.30 0.1268 0.1268 2
## Kerenyi2004 plac rosi 0.77 0.1078 0.1078 2
## Hanefeld2004 metf piog -0.16 0.0849 0.0849 2
## Derosa2004 piog rosi 0.10 0.1831 0.1831 2
## Baksi2004 plac rosi 1.30 0.1014 0.1014 2
## Rosenstock2008 plac rosi 1.09 0.2263 0.2263 2
## Zhu2003 plac rosi 1.50 0.1624 0.1624 2
## Yang2003 metf rosi 0.14 0.2239 0.2239 2
## Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436 2
## Oyama2008 acar sulf -0.40 0.1549 0.1549 2
## Costa1997 acar plac -0.80 0.1432 0.1432 2
## Hermansen2007 plac sita 0.57 0.1291 0.1291 2
## Garber2008 plac vild 0.70 0.1273 0.1273 2
## Alex1998 metf sulf -0.37 0.1184 0.1184 2
## Johnston1994 migl plac -0.74 0.1839 0.1839 2
## Johnston1998a migl plac -1.41 0.2235 0.2235 2
## Kim2007 metf rosi 0.00 0.2339 0.2339 2
## Johnston1998b migl plac -0.68 0.2828 0.2828 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356 2
## Stucci1996 benf plac -0.23 0.3467 0.3467 2
## Moulin2006 benf plac -1.01 0.1366 0.1366 2
## Willms1999 metf plac -1.20 0.3758 0.4125 3 *
## Willms1999 acar plac -1.00 0.4669 0.8242 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q
## DeFronzo1995 metf plac -1.1141 [-1.2309; -0.9973] 30.89
## Lewin2007 metf plac -1.1141 [-1.2309; -0.9973] 8.79
## Willms1999 acar metf 0.2867 [ 0.0622; 0.5113] 0.05
## Davidson2007 plac rosi 1.2018 [ 1.1084; 1.2953] 0.93
## Wolffenbuttel1999 plac rosi 1.2018 [ 1.1084; 1.2953] 0.80
## Kipnes2001 piog plac -1.0664 [-1.2151; -0.9178] 3.39
## Kerenyi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 16.05
## Hanefeld2004 metf piog -0.0477 [-0.1845; 0.0891] 1.75
## Derosa2004 piog rosi 0.1354 [-0.0249; 0.2957] 0.04
## Baksi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 0.94
## Rosenstock2008 plac rosi 1.2018 [ 1.1084; 1.2953] 0.24
## Zhu2003 plac rosi 1.2018 [ 1.1084; 1.2953] 3.37
## Yang2003 metf rosi 0.0877 [-0.0449; 0.2203] 0.05
## Vongthavaravat2002 rosi sulf -0.7623 [-0.9427; -0.5820] 9.29
## Oyama2008 acar sulf -0.3879 [-0.6095; -0.1662] 0.01
## Costa1997 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## Hermansen2007 plac sita 0.5700 [ 0.3170; 0.8230] 0.00
## Garber2008 plac vild 0.7000 [ 0.4505; 0.9495] 0.00
## Alex1998 metf sulf -0.6746 [-0.8482; -0.5011] 6.62
## Johnston1994 migl plac -0.9439 [-1.1927; -0.6952] 1.23
## Johnston1998a migl plac -0.9439 [-1.1927; -0.6952] 4.35
## Kim2007 metf rosi 0.0877 [-0.0449; 0.2203] 0.14
## Johnston1998b migl plac -0.9439 [-1.1927; -0.6952] 0.87
## Gonzalez-Ortiz2004 metf plac -1.1141 [-1.2309; -0.9973] 2.69
## Stucci1996 benf plac -0.9052 [-1.1543; -0.6561] 3.79
## Moulin2006 benf plac -0.9052 [-1.1543; -0.6561] 0.59
## Willms1999 metf plac -1.1141 [-1.2309; -0.9973] 0.04
## Willms1999 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## leverage
## DeFronzo1995 0.18
## Lewin2007 0.36
## Willms1999 0.09
## Davidson2007 0.11
## Wolffenbuttel1999 0.17
## Kipnes2001 0.36
## Kerenyi2004 0.20
## Hanefeld2004 0.68
## Derosa2004 0.20
## Baksi2004 0.22
## Rosenstock2008 0.04
## Zhu2003 0.09
## Yang2003 0.09
## Vongthavaravat2002 0.41
## Oyama2008 0.53
## Costa1997 0.57
## Hermansen2007 1.00
## Garber2008 1.00
## Alex1998 0.56
## Johnston1994 0.48
## Johnston1998a 0.32
## Kim2007 0.08
## Johnston1998b 0.20
## Gonzalez-Ortiz2004 0.02
## Stucci1996 0.13
## Moulin2006 0.87
## Willms1999 0.02
## Willms1999 0.02
##
## Results (random effects model):
##
## treat1 treat2 MD 95%-CI
## DeFronzo1995 metf plac -1.1268 [-1.4291; -0.8244]
## Lewin2007 metf plac -1.1268 [-1.4291; -0.8244]
## Willms1999 acar metf 0.2850 [-0.2208; 0.7908]
## Davidson2007 plac rosi 1.2335 [ 0.9830; 1.4839]
## Wolffenbuttel1999 plac rosi 1.2335 [ 0.9830; 1.4839]
## Kipnes2001 piog plac -1.1291 [-1.5596; -0.6986]
## Kerenyi2004 plac rosi 1.2335 [ 0.9830; 1.4839]
## Hanefeld2004 metf piog 0.0023 [-0.4398; 0.4444]
## Derosa2004 piog rosi 0.1044 [-0.3347; 0.5435]
## Baksi2004 plac rosi 1.2335 [ 0.9830; 1.4839]
## Rosenstock2008 plac rosi 1.2335 [ 0.9830; 1.4839]
## Zhu2003 plac rosi 1.2335 [ 0.9830; 1.4839]
## Yang2003 metf rosi 0.1067 [-0.2170; 0.4304]
## Vongthavaravat2002 rosi sulf -0.8169 [-1.2817; -0.3521]
## Oyama2008 acar sulf -0.4252 [-0.9456; 0.0951]
## Costa1997 acar plac -0.8418 [-1.3236; -0.3600]
## Hermansen2007 plac sita 0.5700 [-0.1240; 1.2640]
## Garber2008 plac vild 0.7000 [ 0.0073; 1.3927]
## Alex1998 metf sulf -0.7102 [-1.1713; -0.2491]
## Johnston1994 migl plac -0.9497 [-1.4040; -0.4955]
## Johnston1998a migl plac -0.9497 [-1.4040; -0.4955]
## Kim2007 metf rosi 0.1067 [-0.2170; 0.4304]
## Johnston1998b migl plac -0.9497 [-1.4040; -0.4955]
## Gonzalez-Ortiz2004 metf plac -1.1268 [-1.4291; -0.8244]
## Stucci1996 benf plac -0.7311 [-1.2918; -0.1705]
## Moulin2006 benf plac -0.7311 [-1.2918; -0.1705]
## Willms1999 metf plac -1.1268 [-1.4291; -0.8244]
## Willms1999 acar plac -0.8418 [-1.3236; -0.3600]
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## acar benf metf migl piog plac rosi sita
## acar . 0.0778 0.2867 0.1166 0.2391 -0.8274 0.3745 -0.2574
## benf -0.0778 . 0.2089 0.0387 0.1612 -0.9052 0.2967 -0.3352
## metf -0.2867 -0.2089 . -0.1702 -0.0477 -1.1141 0.0877 -0.5441
## migl -0.1166 -0.0387 0.1702 . 0.1225 -0.9439 0.2579 -0.3739
## piog -0.2391 -0.1612 0.0477 -0.1225 . -1.0664 0.1354 -0.4964
## plac 0.8274 0.9052 1.1141 0.9439 1.0664 . 1.2018 0.5700
## rosi -0.3745 -0.2967 -0.0877 -0.2579 -0.1354 -1.2018 . -0.6318
## sita 0.2574 0.3352 0.5441 0.3739 0.4964 -0.5700 0.6318 .
## sulf 0.3879 0.4657 0.6746 0.5044 0.6269 -0.4395 0.7623 0.1305
## vild 0.1274 0.2052 0.4141 0.2439 0.3664 -0.7000 0.5018 -0.1300
## sulf vild
## acar -0.3879 -0.1274
## benf -0.4657 -0.2052
## metf -0.6746 -0.4141
## migl -0.5044 -0.2439
## piog -0.6269 -0.3664
## plac 0.4395 0.7000
## rosi -0.7623 -0.5018
## sita -0.1305 0.1300
## sulf . 0.2605
## vild -0.2605 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.2497 0.0622 -0.2107 -0.0094 -1.0401 0.1506 -0.5879
## benf -0.4054 . -0.0662 -0.3133 -0.1288 -1.1543 0.0306 -0.6903
## metf -0.5113 -0.4841 . -0.4450 -0.1845 -1.2309 -0.0449 -0.8228
## migl -0.4438 -0.3908 -0.1046 . -0.1673 -1.1927 -0.0078 -0.7287
## piog -0.4876 -0.4513 -0.0891 -0.4123 . -1.2151 -0.0249 -0.7899
## plac 0.6147 0.6561 0.9973 0.6952 0.9178 . 1.1084 0.3170
## rosi -0.5983 -0.5627 -0.2203 -0.5236 -0.2957 -1.2953 . -0.9016
## sita -0.0732 -0.0199 0.2654 0.0191 0.2030 -0.8230 0.3621 .
## sulf 0.1662 0.1588 0.5011 0.1978 0.4178 -0.6188 0.5820 -0.1796
## vild -0.2005 -0.1474 0.1386 -0.1084 0.0760 -0.9495 0.2354 -0.4854
## sulf vild
## acar -0.6095 -0.4552
## benf -0.7726 -0.5577
## metf -0.8482 -0.6896
## migl -0.8111 -0.5962
## piog -0.8361 -0.6569
## plac 0.2602 0.4505
## rosi -0.9427 -0.7683
## sita -0.4406 -0.2254
## sulf . -0.0467
## vild -0.5677 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.4054 0.5113 0.4438 0.4876 -0.6147 0.5983 0.0732 -0.1662
## benf 0.2497 . 0.4841 0.3908 0.4513 -0.6561 0.5627 0.0199 -0.1588
## metf -0.0622 0.0662 . 0.1046 0.0891 -0.9973 0.2203 -0.2654 -0.5011
## migl 0.2107 0.3133 0.4450 . 0.4123 -0.6952 0.5236 -0.0191 -0.1978
## piog 0.0094 0.1288 0.1845 0.1673 . -0.9178 0.2957 -0.2030 -0.4178
## plac 1.0401 1.1543 1.2309 1.1927 1.2151 . 1.2953 0.8230 0.6188
## rosi -0.1506 -0.0306 0.0449 0.0078 0.0249 -1.1084 . -0.3621 -0.5820
## sita 0.5879 0.6903 0.8228 0.7287 0.7899 -0.3170 0.9016 . 0.1796
## sulf 0.6095 0.7726 0.8482 0.8111 0.8361 -0.2602 0.9427 0.4406 .
## vild 0.4552 0.5577 0.6896 0.5962 0.6569 -0.4505 0.7683 0.2254 0.0467
## vild
## acar 0.2005
## benf 0.1474
## metf -0.1386
## migl 0.1084
## piog -0.0760
## plac 0.9495
## rosi -0.2354
## sita 0.4854
## sulf 0.5677
## vild .
##
## Random effects model
##
## Treatment estimate (sm = 'MD'):
## acar benf metf migl piog plac rosi sita
## acar . -0.1106 0.2850 0.1079 0.2873 -0.8418 0.3917 -0.2718
## benf 0.1106 . 0.3956 0.2186 0.3979 -0.7311 0.5023 -0.1611
## metf -0.2850 -0.3956 . -0.1770 0.0023 -1.1268 0.1067 -0.5568
## migl -0.1079 -0.2186 0.1770 . 0.1794 -0.9497 0.2837 -0.3797
## piog -0.2873 -0.3979 -0.0023 -0.1794 . -1.1291 0.1044 -0.5591
## plac 0.8418 0.7311 1.1268 0.9497 1.1291 . 1.2335 0.5700
## rosi -0.3917 -0.5023 -0.1067 -0.2837 -0.1044 -1.2335 . -0.6635
## sita 0.2718 0.1611 0.5568 0.3797 0.5591 -0.5700 0.6635 .
## sulf 0.4252 0.3146 0.7102 0.5332 0.7125 -0.4166 0.8169 0.1534
## vild 0.1418 0.0311 0.4268 0.2497 0.4291 -0.7000 0.5335 -0.1300
## sulf vild
## acar -0.4252 -0.1418
## benf -0.3146 -0.0311
## metf -0.7102 -0.4268
## migl -0.5332 -0.2497
## piog -0.7125 -0.4291
## plac 0.4166 0.7000
## rosi -0.8169 -0.5335
## sita -0.1534 0.1300
## sulf . 0.2834
## vild -0.2834 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.8499 -0.2208 -0.5542 -0.3313 -1.3236 -0.1189 -1.1166
## benf -0.6286 . -0.2414 -0.5030 -0.3089 -1.2918 -0.1118 -1.0534
## metf -0.7908 -1.0327 . -0.7227 -0.4398 -1.4291 -0.2170 -1.3138
## migl -0.7701 -0.9402 -0.3686 . -0.4465 -1.4040 -0.2350 -1.2092
## piog -0.9059 -1.1048 -0.4444 -0.8052 . -1.5596 -0.3347 -1.3758
## plac 0.3600 0.1705 0.8244 0.4955 0.6986 . 0.9830 -0.1240
## rosi -0.9023 -1.1164 -0.4304 -0.8025 -0.5435 -1.4839 . -1.4013
## sita -0.5731 -0.7311 -0.2002 -0.4497 -0.2576 -1.2640 -0.0744 .
## sulf -0.0951 -0.4184 0.2491 -0.1220 0.1205 -0.8887 0.3521 -0.6859
## vild -0.7020 -0.8601 -0.3291 -0.5787 -0.3865 -1.3927 -0.2032 -1.1106
## sulf vild
## acar -0.9456 -0.9856
## benf -1.0476 -0.9224
## metf -1.1713 -1.1826
## migl -1.1883 -1.0781
## piog -1.3045 -1.2447
## plac -0.0556 0.0073
## rosi -1.2817 -1.2701
## sita -0.9928 -0.8506
## sulf . -0.5549
## vild -1.1218 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.6286 0.7908 0.7701 0.9059 -0.3600 0.9023 0.5731 0.0951
## benf 0.8499 . 1.0327 0.9402 1.1048 -0.1705 1.1164 0.7311 0.4184
## metf 0.2208 0.2414 . 0.3686 0.4444 -0.8244 0.4304 0.2002 -0.2491
## migl 0.5542 0.5030 0.7227 . 0.8052 -0.4955 0.8025 0.4497 0.1220
## piog 0.3313 0.3089 0.4398 0.4465 . -0.6986 0.5435 0.2576 -0.1205
## plac 1.3236 1.2918 1.4291 1.4040 1.5596 . 1.4839 1.2640 0.8887
## rosi 0.1189 0.1118 0.2170 0.2350 0.3347 -0.9830 . 0.0744 -0.3521
## sita 1.1166 1.0534 1.3138 1.2092 1.3758 0.1240 1.4013 . 0.6859
## sulf 0.9456 1.0476 1.1713 1.1883 1.3045 0.0556 1.2817 0.9928 .
## vild 0.9856 0.9224 1.1826 1.0781 1.2447 -0.0073 1.2701 0.8506 0.5549
## vild
## acar 0.7020
## benf 0.8601
## metf 0.3291
## migl 0.5787
## piog 0.3865
## plac 1.3927
## rosi 0.2032
## sita 1.1106
## sulf 1.1218
## vild .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
# Change printing order of treatments (placebo first)
trts <- c("plac", "acar", "benf", "metf", "migl", "piog",
"rosi", "sita", "sulf", "vild")
net3 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD",
seq=trts)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
print(summary(net3), digits=2)
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## plac acar benf metf migl piog rosi sita sulf vild
## plac . 0.83 0.91 1.11 0.94 1.07 1.20 0.57 0.44 0.70
## acar -0.83 . 0.08 0.29 0.12 0.24 0.37 -0.26 -0.39 -0.13
## benf -0.91 -0.08 . 0.21 0.04 0.16 0.30 -0.34 -0.47 -0.21
## metf -1.11 -0.29 -0.21 . -0.17 -0.05 0.09 -0.54 -0.67 -0.41
## migl -0.94 -0.12 -0.04 0.17 . 0.12 0.26 -0.37 -0.50 -0.24
## piog -1.07 -0.24 -0.16 0.05 -0.12 . 0.14 -0.50 -0.63 -0.37
## rosi -1.20 -0.37 -0.30 -0.09 -0.26 -0.14 . -0.63 -0.76 -0.50
## sita -0.57 0.26 0.34 0.54 0.37 0.50 0.63 . -0.13 0.13
## sulf -0.44 0.39 0.47 0.67 0.50 0.63 0.76 0.13 . 0.26
## vild -0.70 0.13 0.21 0.41 0.24 0.37 0.50 -0.13 -0.26 .
##
## Lower 95%-confidence limit:
## plac acar benf metf migl piog rosi sita sulf vild
## plac . 0.61 0.66 1.00 0.70 0.92 1.11 0.32 0.26 0.45
## acar -1.04 . -0.25 0.06 -0.21 -0.01 0.15 -0.59 -0.61 -0.46
## benf -1.15 -0.41 . -0.07 -0.31 -0.13 0.03 -0.69 -0.77 -0.56
## metf -1.23 -0.51 -0.48 . -0.44 -0.18 -0.04 -0.82 -0.85 -0.69
## migl -1.19 -0.44 -0.39 -0.10 . -0.17 -0.01 -0.73 -0.81 -0.60
## piog -1.22 -0.49 -0.45 -0.09 -0.41 . -0.02 -0.79 -0.84 -0.66
## rosi -1.30 -0.60 -0.56 -0.22 -0.52 -0.30 . -0.90 -0.94 -0.77
## sita -0.82 -0.07 -0.02 0.27 0.02 0.20 0.36 . -0.44 -0.23
## sulf -0.62 0.17 0.16 0.50 0.20 0.42 0.58 -0.18 . -0.05
## vild -0.95 -0.20 -0.15 0.14 -0.11 0.08 0.24 -0.49 -0.57 .
##
## Upper 95%-confidence limit:
## plac acar benf metf migl piog rosi sita sulf vild
## plac . 1.04 1.15 1.23 1.19 1.22 1.30 0.82 0.62 0.95
## acar -0.61 . 0.41 0.51 0.44 0.49 0.60 0.07 -0.17 0.20
## benf -0.66 0.25 . 0.48 0.39 0.45 0.56 0.02 -0.16 0.15
## metf -1.00 -0.06 0.07 . 0.10 0.09 0.22 -0.27 -0.50 -0.14
## migl -0.70 0.21 0.31 0.44 . 0.41 0.52 -0.02 -0.20 0.11
## piog -0.92 0.01 0.13 0.18 0.17 . 0.30 -0.20 -0.42 -0.08
## rosi -1.11 -0.15 -0.03 0.04 0.01 0.02 . -0.36 -0.58 -0.24
## sita -0.32 0.59 0.69 0.82 0.73 0.79 0.90 . 0.18 0.49
## sulf -0.26 0.61 0.77 0.85 0.81 0.84 0.94 0.44 . 0.57
## vild -0.45 0.46 0.56 0.69 0.60 0.66 0.77 0.23 0.05 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
data(Linde2015)
# Define order of treatments
trts <- c("TCA", "SSRI", "SNRI", "NRI",
"Low-dose SARI", "NaSSa", "rMAO-A", "Hypericum",
"Placebo")
# Outcome labels
outcomes <- c("Early response", "Early remission")
# (1) Early response
p1 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(resp1, resp2, resp3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 14 TCA SSRI NA 10 NA 11 0 FALSE
## NA NA 18 TCA Placebo NA 73 NA 73 0 FALSE
## NA NA 21 TCA rMAO-A NA 46 NA 98 0 FALSE
## NA NA 27 SSRI Placebo NA 80 NA 81 0 FALSE
## NA NA 51 TCA rMAO-A NA 71 NA 71 0 FALSE
## NA NA 130 TCA NaSSa NA 35 NA 36 0 FALSE
## NA NA 131 SSRI SNRI NA 697 NA 688 0 FALSE
## NA NA 130 TCA Placebo NA 35 NA 34 0 FALSE
## NA NA 130 NaSSa Placebo NA 36 NA 34 0 FALSE
net1 <- netmeta(p1,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 21 TCA rMAO-A NA NA
## 27 SSRI Placebo NA NA
## 51 TCA rMAO-A NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## 131 SSRI SNRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# (2) Early remission
p2 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(remi1, remi2, remi3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr
## NA NA 1 TCA SNRI NA 75 NA 78 0.0
## NA NA 11 TCA SSRI NA 108 NA 99 0.0
## NA NA 14 TCA SSRI NA 10 NA 11 0.0
## NA NA 18 TCA Placebo NA 73 NA 73 0.0
## NA NA 20 TCA SSRI NA 55 NA 51 0.0
## NA NA 26 SSRI Placebo NA 314 NA 154 0.0
## NA NA 53 SSRI NaSSa NA 122 NA 121 0.0
## NA NA 56 TCA SSRI NA 92 NA 380 0.0
## NA NA 73 Hypericum Placebo NA 55 NA 57 0.0
## NA NA 90 TCA SSRI NA 42 NA 42 0.0
## NA NA 96 TCA SSRI 0 30 0 29 0.5
## NA NA 121 Low-dose SARI NaSSa NA 43 NA 40 0.0
## NA NA 130 TCA NaSSa NA 35 NA 36 0.0
## NA NA 1 TCA Placebo NA 75 NA 76 0.0
## NA NA 11 TCA Placebo NA 108 NA 101 0.0
## NA NA 53 SSRI Placebo NA 122 NA 129 0.0
## NA NA 130 TCA Placebo NA 35 NA 34 0.0
## NA NA 1 SNRI Placebo NA 78 NA 76 0.0
## NA NA 11 SSRI Placebo NA 99 NA 101 0.0
## NA NA 53 NaSSa Placebo NA 121 NA 129 0.0
## NA NA 130 NaSSa Placebo NA 36 NA 34 0.0
## allstudies
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
net2 <- netmeta(p2,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 1 TCA Placebo NA NA
## 1 TCA SNRI NA NA
## 1 SNRI Placebo NA NA
## 11 SSRI Placebo NA NA
## 11 TCA Placebo NA NA
## 11 TCA SSRI NA NA
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 20 TCA SSRI NA NA
## 26 SSRI Placebo NA NA
## 53 NaSSa Placebo NA NA
## 53 SSRI NaSSa NA NA
## 53 SSRI Placebo NA NA
## 56 TCA SSRI NA NA
## 73 Hypericum Placebo NA NA
## 90 TCA SSRI NA NA
## 96 TCA SSRI NA NA
## 121 Low-dose SARI NaSSa NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Partial order of treatment rankings (all five outcomes)
po <- netposet(netrank(net1, small.values = "bad"),
netrank(net2, small.values = "bad"),
outcomes = outcomes)
# Outcome labels
outcomes <- c("Early response", "Early remission",
"Lost to follow-up", "Lost to follow-up due to AEs",
"Adverse events (AEs)")
# (3) Loss to follow-up
p3 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(loss1, loss2, loss3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 9 TCA SSRI NA 363 NA 173 0 FALSE
## NA NA 14 TCA SSRI NA 10 NA 11 0 FALSE
## NA NA 48 SSRI Placebo 31 191 NA 189 0 FALSE
## NA NA 53 SSRI NaSSa NA 122 NA 121 0 FALSE
## NA NA 90 TCA SSRI NA 42 NA 42 0 FALSE
## NA NA 116 SSRI NaSSa NA 31 NA 31 0 FALSE
## NA NA 120 TCA SSRI NA 218 NA 109 0 FALSE
## NA NA 53 SSRI Placebo NA 122 NA 129 0 FALSE
## NA NA 53 NaSSa Placebo NA 121 NA 129 0 FALSE
net3 <- netmeta(p3,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 9 TCA SSRI NA NA
## 14 TCA SSRI NA NA
## 48 SSRI Placebo NA NA
## 53 NaSSa Placebo NA NA
## 53 SSRI NaSSa NA NA
## 53 SSRI Placebo NA NA
## 90 TCA SSRI NA NA
## 116 SSRI NaSSa NA NA
## 120 TCA SSRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# (4) Loss to follow-up due to adverse events
p4 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(loss.ae1, loss.ae2, loss.ae3),
n = list(n1, n2, n3),
studlab = id, data = subset(Linde2015, id != 55),
sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr
## NA NA 7 TCA Placebo NA 32 NA 31 0.0
## NA NA 9 TCA SSRI NA 363 NA 173 0.0
## NA NA 14 TCA SSRI 0 10 0 11 0.5
## NA NA 30 NaSSa rMAO-A NA 38 NA 79 0.0
## NA NA 31 TCA Placebo NA 90 NA 88 0.0
## NA NA 66 Hypericum Placebo 0 20 0 20 0.5
## NA NA 68 Hypericum Placebo 0 20 0 20 0.5
## NA NA 71 Hypericum Placebo 0 48 0 49 0.5
## NA NA 90 TCA SSRI NA 42 NA 42 0.0
## NA NA 114 Low-dose SARI NaSSa 0 20 0 20 0.5
## NA NA 120 TCA SSRI NA 218 NA 109 0.0
## NA NA 128 TCA SSRI NA 50 NA 53 0.0
## NA NA 130 TCA NaSSa NA 35 NA 36 0.0
## NA NA 130 TCA Placebo NA 35 NA 34 0.0
## NA NA 130 NaSSa Placebo NA 36 NA 34 0.0
## allstudies
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
net4 <- netmeta(p4,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 7 TCA Placebo NA NA
## 9 TCA SSRI NA NA
## 14 TCA SSRI NA NA
## 30 NaSSa rMAO-A NA NA
## 31 TCA Placebo NA NA
## 66 Hypericum Placebo NA NA
## 68 Hypericum Placebo NA NA
## 71 Hypericum Placebo NA NA
## 90 TCA SSRI NA NA
## 114 Low-dose SARI NaSSa NA NA
## 120 TCA SSRI NA NA
## 128 TCA SSRI NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# (5) Adverse events
p5 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(ae1, ae2, ae3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr
## NA NA 1 TCA SNRI NA 75 NA 78 0.0
## NA NA 4 TCA Placebo NA 35 NA 23 0.0
## NA NA 7 TCA Placebo NA 32 NA 31 0.0
## NA NA 9 TCA SSRI NA 363 NA 173 0.0
## NA NA 22 TCA SSRI NA 73 NA 71 0.0
## NA NA 24 TCA Placebo NA 31 NA 30 0.0
## NA NA 27 SSRI Placebo NA 80 NA 81 0.0
## NA NA 30 NaSSa rMAO-A NA 38 NA 79 0.0
## NA NA 31 TCA Placebo NA 90 NA 88 0.0
## NA NA 39 NaSSa Placebo NA 27 NA 25 0.0
## NA NA 49 SSRI Placebo NA 137 NA 140 0.0
## NA NA 51 TCA rMAO-A NA 71 NA 71 0.0
## NA NA 52 TCA rMAO-A NA 23 NA 30 0.0
## NA NA 53 SSRI NaSSa NA 122 NA 121 0.0
## NA NA 56 TCA SSRI NA 92 NA 380 0.0
## NA NA 66 Hypericum Placebo 0 20 0 20 0.5
## NA NA 68 Hypericum Placebo 0 20 0 20 0.5
## NA NA 83 TCA Low-dose SARI NA 79 NA 112 0.0
## NA NA 88 Low-dose SARI NaSSa NA 61 NA 64 0.0
## NA NA 90 TCA SSRI NA 42 NA 42 0.0
## NA NA 91 TCA rMAO-A NA 64 NA 66 0.0
## NA NA 112 SSRI SNRI NA 178 NA 183 0.0
## NA NA 113 TCA Low-dose SARI NA 114 NA 114 0.0
## NA NA 120 TCA SSRI NA 218 NA 109 0.0
## NA NA 128 TCA SSRI NA 50 NA 53 0.0
## NA NA 130 TCA NaSSa NA 35 NA 36 0.0
## NA NA 132 TCA SNRI NA 43 NA 45 0.0
## NA NA 135 SSRI NRI NA 298 NA 303 0.0
## NA NA 1 TCA Placebo NA 75 NA 76 0.0
## NA NA 53 SSRI Placebo NA 122 NA 129 0.0
## NA NA 83 TCA NaSSa NA 79 NA 36 0.0
## NA NA 130 TCA Placebo NA 35 NA 34 0.0
## NA NA 1 SNRI Placebo NA 78 NA 76 0.0
## NA NA 53 NaSSa Placebo NA 121 NA 129 0.0
## NA NA 83 Low-dose SARI NaSSa NA 112 NA 36 0.0
## NA NA 130 NaSSa Placebo NA 36 NA 34 0.0
## allstudies
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
net5 <- netmeta(p5,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 1 TCA Placebo NA NA
## 1 TCA SNRI NA NA
## 1 SNRI Placebo NA NA
## 4 TCA Placebo NA NA
## 7 TCA Placebo NA NA
## 9 TCA SSRI NA NA
## 22 TCA SSRI NA NA
## 24 TCA Placebo NA NA
## 27 SSRI Placebo NA NA
## 30 NaSSa rMAO-A NA NA
## 31 TCA Placebo NA NA
## 39 NaSSa Placebo NA NA
## 49 SSRI Placebo NA NA
## 51 TCA rMAO-A NA NA
## 52 TCA rMAO-A NA NA
## 53 NaSSa Placebo NA NA
## 53 SSRI NaSSa NA NA
## 53 SSRI Placebo NA NA
## 56 TCA SSRI NA NA
## 66 Hypericum Placebo NA NA
## 68 Hypericum Placebo NA NA
## 83 Low-dose SARI NaSSa NA NA
## 83 TCA Low-dose SARI NA NA
## 83 TCA NaSSa NA NA
## 88 Low-dose SARI NaSSa NA NA
## 90 TCA SSRI NA NA
## 91 TCA rMAO-A NA NA
## 112 SSRI SNRI NA NA
## 113 TCA Low-dose SARI NA NA
## 120 TCA SSRI NA NA
## 128 TCA SSRI NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## 132 TCA SNRI NA NA
## 135 SSRI NRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Partial order of treatment rankings (all five outcomes)
po.ranks <- netposet(netrank(net1, small.values = "bad"),
netrank(net2, small.values = "bad"),
netrank(net3, small.values = "good"),
netrank(net4, small.values = "good"),
netrank(net5, small.values = "good"),
outcomes = outcomes)
# Same result
po.nets <- netposet(net1, net2, net3, net4, net5,
small.values = c("bad", "bad", "good", "good", "good"),
outcomes = outcomes)
all.equal(po.ranks, po.nets)
## [1] "Component \"small.values\": target is NULL, current is character"
## [2] "Component \"call\": target, current do not match when deparsed"
# Print matrix with P-scores (random effects model)
po.nets$P.random
## Early response Early remission Lost to follow-up
## Hypericum 0.89387883 0.745274702 0.84825521
## Low-dose SARI 0.72014426 0.615516715 0.90565348
## NaSSa 0.21275817 0.334605556 0.32869707
## NRI 0.44453371 0.551304613 0.05634402
## Placebo 0.09078155 0.006304086 0.41048763
## rMAO-A 0.15210079 0.319624853 0.59842793
## SNRI 0.68917883 0.768103792 0.46848342
## SSRI 0.61643623 0.511123290 0.44722161
## TCA 0.68018764 0.648142391 0.43642963
## Lost to follow-up due to AEs Adverse events (AEs)
## Hypericum 0.8207662 0.74877289
## Low-dose SARI 0.7153234 0.88020103
## NaSSa 0.1416704 0.69852753
## NRI 0.0230363 NA
## Placebo 0.8354985 0.64096019
## rMAO-A 0.8568193 0.47913934
## SNRI 0.2770870 0.19361360
## SSRI 0.4842401 0.32477927
## TCA 0.3455589 0.03400613
po12 <- netposet(netrank(net1, small.values = "bad"),
netrank(net2, small.values = "bad"),
outcomes = outcomes[1:2])
# Scatter plot
oldpar <- par(pty = "s")
plot(po12)
par(oldpar)
# Example using ranking matrix with P-scores
# Ribassin-Majed L, Marguet S, Lee A.W., et al. (2017),
# What is the best treatment of locally advanced nasopharyngeal
# carcinoma? An individual patient data network meta-analysis.
# Journal of Clinical Oncology.
# 35, 498-505, DOI:10.1200/JCO.2016.67.4119
outcomes <- c("OS", "PFS", "LC", "DC")
treatments <- c("RT", "IC-RT", "IC-CRT", "CRT",
"CRT-AC", "RT-AC", "IC-RT-AC")
# P-scores (from Table 1)
pscore.os <- c(15, 33, 63, 70, 96, 28, 45) / 100
pscore.pfs <- c( 4, 46, 79, 52, 94, 36, 39) / 100
pscore.lc <- c( 9, 27, 47, 37, 82, 58, 90) / 100
pscore.dc <- c(16, 76, 95, 48, 72, 32, 10) / 100
pscore.matrix <- data.frame(pscore.os, pscore.pfs, pscore.lc, pscore.dc)
rownames(pscore.matrix) <- treatments
colnames(pscore.matrix) <- outcomes
pscore.matrix
## OS PFS LC DC
## RT 0.15 0.04 0.09 0.16
## IC-RT 0.33 0.46 0.27 0.76
## IC-CRT 0.63 0.79 0.47 0.95
## CRT 0.70 0.52 0.37 0.48
## CRT-AC 0.96 0.94 0.82 0.72
## RT-AC 0.28 0.36 0.58 0.32
## IC-RT-AC 0.45 0.39 0.90 0.10
po <- netposet(pscore.matrix)
po12 <- netposet(pscore.matrix[, 1:2])
po
## RT IC-RT IC-CRT CRT CRT-AC RT-AC IC-RT-AC
## RT 0 0 0 0 0 0 0
## IC-RT 1 0 0 0 0 0 0
## IC-CRT 0 1 0 0 0 0 0
## CRT 1 0 0 0 0 0 0
## CRT-AC 0 0 0 1 0 1 0
## RT-AC 1 0 0 0 0 0 0
## IC-RT-AC 0 0 0 0 0 0 0
po12
## RT IC-RT IC-CRT CRT CRT-AC RT-AC IC-RT-AC
## RT 0 0 0 0 0 0 0
## IC-RT 0 0 0 0 0 1 0
## IC-CRT 0 1 0 0 0 0 1
## CRT 0 1 0 0 0 0 1
## CRT-AC 0 0 1 1 0 0 0
## RT-AC 1 0 0 0 0 0 0
## IC-RT-AC 0 0 0 0 0 1 0
oldpar <- par(pty = "s")
plot(po12)
par(oldpar)
data(Senn2013)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
net2 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD",
comb.fixed=FALSE, comb.random=TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
net3 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD",comb.random=TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
nr1 <- netrank(net1)
nr1
## P-score
## rosi 0.9789
## metf 0.8513
## piog 0.7686
## migl 0.6200
## benf 0.5727
## acar 0.4792
## vild 0.3512
## sita 0.2386
## sulf 0.1395
## plac 0.0000
print(nr1, sort=FALSE)
## P-score
## acar 0.4792
## benf 0.5727
## metf 0.8513
## migl 0.6200
## piog 0.7686
## plac 0.0000
## rosi 0.9789
## sita 0.2386
## sulf 0.1395
## vild 0.3512
nr2 <- netrank(net2)
nr2
## P-score
## rosi 0.8934
## metf 0.7818
## piog 0.7746
## migl 0.6137
## acar 0.5203
## benf 0.4358
## vild 0.4232
## sita 0.3331
## sulf 0.2103
## plac 0.0139
print(nr2, sort=FALSE)
## P-score
## acar 0.5203
## benf 0.4358
## metf 0.7818
## migl 0.6137
## piog 0.7746
## plac 0.0139
## rosi 0.8934
## sita 0.3331
## sulf 0.2103
## vild 0.4232
nr3 <- netrank(net3)
nr3
## P-score (fixed) P-score (random)
## rosi 0.9789 0.8934
## metf 0.8513 0.7818
## piog 0.7686 0.7746
## migl 0.6200 0.6137
## acar 0.4792 0.5203
## benf 0.5727 0.4358
## vild 0.3512 0.4232
## sita 0.2386 0.3331
## sulf 0.1395 0.2103
## plac 0.0000 0.0139
print(nr3, sort="fixed")
## P-score (fixed) P-score (random)
## rosi 0.9789 0.8934
## metf 0.8513 0.7818
## piog 0.7686 0.7746
## migl 0.6200 0.6137
## benf 0.5727 0.4358
## acar 0.4792 0.5203
## vild 0.3512 0.4232
## sita 0.2386 0.3331
## sulf 0.1395 0.2103
## plac 0.0000 0.0139
print(nr3, sort=FALSE)
## P-score (fixed) P-score (random)
## acar 0.4792 0.5203
## benf 0.5727 0.4358
## metf 0.8513 0.7818
## migl 0.6200 0.6137
## piog 0.7686 0.7746
## plac 0.0000 0.0139
## rosi 0.9789 0.8934
## sita 0.2386 0.3331
## sulf 0.1395 0.2103
## vild 0.3512 0.4232
data(Woods2010)
p1 <- pairwise(treatment, event = r, n = N,
studlab = author, data = Woods2010, sm = "OR")
net1 <- netmeta(p1)
print(netsplit(net1), digits = 2)
## Fixed effect model:
##
## comparison k prop nma direct indir. RoR z p-value
## Fluticasone:Placebo 1 0.96 1.81 1.83 1.49 1.23 0.06 0.9486
## Fluticasone:Salmeterol 1 0.86 0.77 0.75 0.86 0.88 -0.06 0.9485
## Fluticasone:SFC 1 1.00 0.52 0.52 . . . .
## Placebo:Salmeterol 3 1.00 0.42 0.42 . . . .
## Placebo:SFC 1 0.98 0.29 0.28 0.40 0.72 -0.06 0.9486
## Salmeterol:SFC 1 0.90 0.68 0.69 0.57 1.21 0.06 0.9485
##
## Legend:
## comparison - Treatment comparison
## k - Number of studies providing direct evidence
## prop - Direct evidence proportion
## nma - Estimated treatment effect (OR) in network meta-analysis
## direct - Estimated treatment effect (OR) derived from direct evidence
## indir. - Estimated treatment effect (OR) derived from indirect evidence
## RoR - Ratio of Ratios (direct versus indirect)
## z - z-value of test for disagreement (direct versus indirect)
## p-value - p-value of test for disagreement (direct versus indirect)
print(netsplit(net1), digits = 2,
backtransf = FALSE, comb.random = TRUE)
## Fixed effect model:
##
## comparison k prop nma direct indir. Diff z p-value
## Fluticasone:Placebo 1 0.96 0.60 0.60 0.40 0.20 0.06 0.9486
## Fluticasone:Salmeterol 1 0.86 -0.27 -0.28 -0.15 -0.13 -0.06 0.9485
## Fluticasone:SFC 1 1.00 -0.65 -0.65 . . . .
## Placebo:Salmeterol 3 1.00 -0.86 -0.86 . . . .
## Placebo:SFC 1 0.98 -1.25 -1.26 -0.93 -0.33 -0.06 0.9486
## Salmeterol:SFC 1 0.90 -0.39 -0.37 -0.56 0.19 0.06 0.9485
##
## Random effects model:
##
## comparison k prop nma direct indir. Diff z p-value
## Fluticasone:Placebo 1 0.96 0.60 0.60 0.40 0.20 0.06 0.9486
## Fluticasone:Salmeterol 1 0.86 -0.27 -0.28 -0.15 -0.13 -0.06 0.9485
## Fluticasone:SFC 1 1.00 -0.65 -0.65 . . . .
## Placebo:Salmeterol 3 1.00 -0.86 -0.86 . . . .
## Placebo:SFC 1 0.98 -1.25 -1.26 -0.93 -0.33 -0.06 0.9486
## Salmeterol:SFC 1 0.90 -0.39 -0.37 -0.56 0.19 0.06 0.9485
##
## Legend:
## comparison - Treatment comparison
## k - Number of studies providing direct evidence
## prop - Direct evidence proportion
## nma - Estimated treatment effect (logOR) in network meta-analysis
## direct - Estimated treatment effect (logOR) derived from direct evidence
## indir. - Estimated treatment effect (logOR) derived from indirect evidence
## Diff - Difference between direct and indirect treatment estimates
## z - z-value of test for disagreement (direct versus indirect)
## p-value - p-value of test for disagreement (direct versus indirect)
data(Senn2013)
net2 <- netmeta(TE, seTE, treat1, treat2, studlab, data = Senn2013,
comb.random = TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
print(netsplit(net2), digits = 2)
## Fixed effect model:
##
## comparison k prop nma direct indir. Diff z p-value
## acar:benf 0 0.00 -0.08 . -0.08 . . .
## acar:metf 1 0.10 -0.29 -0.20 -0.30 0.10 0.26 0.7981
## acar:migl 0 0.00 -0.12 . -0.12 . . .
## acar:piog 0 0.00 -0.24 . -0.24 . . .
## acar:plac 2 0.63 0.83 0.82 0.84 -0.03 -0.12 0.9030
## acar:rosi 0 0.00 -0.37 . -0.37 . . .
## acar:sita 0 0.00 0.26 . 0.26 . . .
## acar:sulf 1 0.53 0.39 0.40 0.37 0.03 0.11 0.9088
## acar:vild 0 0.00 0.13 . 0.13 . . .
## benf:metf 0 0.00 -0.21 . -0.21 . . .
## benf:migl 0 0.00 -0.04 . -0.04 . . .
## benf:piog 0 0.00 -0.16 . -0.16 . . .
## benf:plac 2 1.00 0.91 0.91 . . . .
## benf:rosi 0 0.00 -0.30 . -0.30 . . .
## benf:sita 0 0.00 0.34 . 0.34 . . .
## benf:sulf 0 0.00 0.47 . 0.47 . . .
## benf:vild 0 0.00 0.21 . 0.21 . . .
## metf:migl 0 0.00 0.17 . 0.17 . . .
## metf:piog 1 0.68 0.05 0.16 -0.19 0.35 2.32 0.0201
## metf:plac 4 0.58 1.11 1.15 1.06 0.09 0.76 0.4489
## metf:rosi 2 0.18 -0.09 -0.07 -0.09 0.02 0.10 0.9204
## metf:sita 0 0.00 0.54 . 0.54 . . .
## metf:sulf 1 0.56 0.67 0.37 1.06 -0.69 -3.87 0.0001
## metf:vild 0 0.00 0.41 . 0.41 . . .
## migl:piog 0 0.00 -0.12 . -0.12 . . .
## migl:plac 3 1.00 0.94 0.94 . . . .
## migl:rosi 0 0.00 -0.26 . -0.26 . . .
## migl:sita 0 0.00 0.37 . 0.37 . . .
## migl:sulf 0 0.00 0.50 . 0.50 . . .
## migl:vild 0 0.00 0.24 . 0.24 . . .
## piog:plac 1 0.36 1.07 1.30 0.94 0.36 2.30 0.0215
## piog:rosi 1 0.20 -0.14 -0.10 -0.14 0.04 0.22 0.8289
## piog:sita 0 0.00 0.50 . 0.50 . . .
## piog:sulf 0 0.00 0.63 . 0.63 . . .
## piog:vild 0 0.00 0.37 . 0.37 . . .
## plac:rosi 6 0.83 -1.20 -1.15 -1.47 0.32 2.50 0.0125
## plac:sita 1 1.00 -0.57 -0.57 . . . .
## plac:sulf 0 0.00 -0.44 . -0.44 . . .
## plac:vild 1 1.00 -0.70 -0.70 . . . .
## rosi:sita 0 0.00 0.63 . 0.63 . . .
## rosi:sulf 1 0.41 0.76 1.20 0.46 0.74 3.97 < 0.0001
## rosi:vild 0 0.00 0.50 . 0.50 . . .
## sita:sulf 0 0.00 0.13 . 0.13 . . .
## sita:vild 0 0.00 -0.13 . -0.13 . . .
## sulf:vild 0 0.00 -0.26 . -0.26 . . .
##
## Random effects model:
##
## comparison k prop nma direct indir. Diff z p-value
## acar:benf 0 0.00 0.11 . 0.11 . . .
## acar:metf 1 0.28 -0.28 -0.20 -0.32 0.12 0.21 0.8368
## acar:migl 0 0.00 -0.11 . -0.11 . . .
## acar:piog 0 0.00 -0.29 . -0.29 . . .
## acar:plac 2 0.65 0.84 0.86 0.81 0.04 0.08 0.9338
## acar:rosi 0 0.00 -0.39 . -0.39 . . .
## acar:sita 0 0.00 0.27 . 0.27 . . .
## acar:sulf 1 0.53 0.43 0.40 0.45 -0.05 -0.10 0.9195
## acar:vild 0 0.00 0.14 . 0.14 . . .
## benf:metf 0 0.00 -0.40 . -0.40 . . .
## benf:migl 0 0.00 -0.22 . -0.22 . . .
## benf:piog 0 0.00 -0.40 . -0.40 . . .
## benf:plac 2 1.00 0.73 0.73 . . . .
## benf:rosi 0 0.00 -0.50 . -0.50 . . .
## benf:sita 0 0.00 0.16 . 0.16 . . .
## benf:sulf 0 0.00 0.31 . 0.31 . . .
## benf:vild 0 0.00 0.03 . 0.03 . . .
## metf:migl 0 0.00 0.18 . 0.18 . . .
## metf:piog 1 0.44 -0.00 0.16 -0.13 0.29 0.64 0.5245
## metf:plac 4 0.56 1.13 1.18 1.06 0.12 0.40 0.6912
## metf:rosi 2 0.34 -0.11 -0.07 -0.12 0.05 0.15 0.8771
## metf:sita 0 0.00 0.56 . 0.56 . . .
## metf:sulf 1 0.45 0.71 0.37 0.99 -0.62 -1.31 0.1900
## metf:vild 0 0.00 0.43 . 0.43 . . .
## migl:piog 0 0.00 -0.18 . -0.18 . . .
## migl:plac 3 1.00 0.95 0.95 . . . .
## migl:rosi 0 0.00 -0.28 . -0.28 . . .
## migl:sita 0 0.00 0.38 . 0.38 . . .
## migl:sulf 0 0.00 0.53 . 0.53 . . .
## migl:vild 0 0.00 0.25 . 0.25 . . .
## piog:plac 1 0.39 1.13 1.30 1.02 0.28 0.62 0.5368
## piog:rosi 1 0.35 -0.10 -0.10 -0.11 0.01 0.01 0.9885
## piog:sita 0 0.00 0.56 . 0.56 . . .
## piog:sulf 0 0.00 0.71 . 0.71 . . .
## piog:vild 0 0.00 0.43 . 0.43 . . .
## plac:rosi 6 0.76 -1.23 -1.18 -1.40 0.22 0.74 0.4587
## plac:sita 1 1.00 -0.57 -0.57 . . . .
## plac:sulf 0 0.00 -0.42 . -0.42 . . .
## plac:vild 1 1.00 -0.70 -0.70 . . . .
## rosi:sita 0 0.00 0.66 . 0.66 . . .
## rosi:sulf 1 0.43 0.82 1.20 0.52 0.68 1.42 0.1565
## rosi:vild 0 0.00 0.53 . 0.53 . . .
## sita:sulf 0 0.00 0.15 . 0.15 . . .
## sita:vild 0 0.00 -0.13 . -0.13 . . .
## sulf:vild 0 0.00 -0.28 . -0.28 . . .
##
## Legend:
## comparison - Treatment comparison
## k - Number of studies providing direct evidence
## prop - Direct evidence proportion
## nma - Estimated treatment effect in network meta-analysis
## direct - Estimated treatment effect derived from direct evidence
## indir. - Estimated treatment effect derived from indirect evidence
## Diff - Difference between direct and indirect treatment estimates
## z - z-value of test for disagreement (direct versus indirect)
## p-value - p-value of test for disagreement (direct versus indirect)
# Layout of Puhan et al. (2014), Table 1
print(netsplit(net2), digits = 2, ci = TRUE, test = FALSE)
## Fixed effect model:
##
## comparison k prop nma 95%-CI direct 95%-CI indir. 95%-CI
## acar:benf 0 0.00 -0.08 [-0.41; 0.25] . . -0.08 [-0.41; 0.25]
## acar:metf 1 0.10 -0.29 [-0.51; -0.06] -0.20 [-0.90; 0.50] -0.30 [-0.53; -0.06]
## acar:migl 0 0.00 -0.12 [-0.44; 0.21] . . -0.12 [-0.44; 0.21]
## acar:piog 0 0.00 -0.24 [-0.49; 0.01] . . -0.24 [-0.49; 0.01]
## acar:plac 2 0.63 0.83 [ 0.61; 1.04] 0.82 [ 0.55; 1.09] 0.84 [ 0.50; 1.19]
## acar:rosi 0 0.00 -0.37 [-0.60; -0.15] . . -0.37 [-0.60; -0.15]
## acar:sita 0 0.00 0.26 [-0.07; 0.59] . . 0.26 [-0.07; 0.59]
## acar:sulf 1 0.53 0.39 [ 0.17; 0.61] 0.40 [ 0.10; 0.70] 0.37 [ 0.05; 0.70]
## acar:vild 0 0.00 0.13 [-0.20; 0.46] . . 0.13 [-0.20; 0.46]
## benf:metf 0 0.00 -0.21 [-0.48; 0.07] . . -0.21 [-0.48; 0.07]
## benf:migl 0 0.00 -0.04 [-0.39; 0.31] . . -0.04 [-0.39; 0.31]
## benf:piog 0 0.00 -0.16 [-0.45; 0.13] . . -0.16 [-0.45; 0.13]
## benf:plac 2 1.00 0.91 [ 0.66; 1.15] 0.91 [ 0.66; 1.15] . .
## benf:rosi 0 0.00 -0.30 [-0.56; -0.03] . . -0.30 [-0.56; -0.03]
## benf:sita 0 0.00 0.34 [-0.02; 0.69] . . 0.34 [-0.02; 0.69]
## benf:sulf 0 0.00 0.47 [ 0.16; 0.77] . . 0.47 [ 0.16; 0.77]
## benf:vild 0 0.00 0.21 [-0.15; 0.56] . . 0.21 [-0.15; 0.56]
## metf:migl 0 0.00 0.17 [-0.10; 0.44] . . 0.17 [-0.10; 0.44]
## metf:piog 1 0.68 0.05 [-0.09; 0.18] 0.16 [-0.01; 0.33] -0.19 [-0.43; 0.05]
## metf:plac 4 0.58 1.11 [ 1.00; 1.23] 1.15 [ 1.00; 1.31] 1.06 [ 0.88; 1.24]
## metf:rosi 2 0.18 -0.09 [-0.22; 0.04] -0.07 [-0.39; 0.24] -0.09 [-0.24; 0.06]
## metf:sita 0 0.00 0.54 [ 0.27; 0.82] . . 0.54 [ 0.27; 0.82]
## metf:sulf 1 0.56 0.67 [ 0.50; 0.85] 0.37 [ 0.14; 0.60] 1.06 [ 0.80; 1.32]
## metf:vild 0 0.00 0.41 [ 0.14; 0.69] . . 0.41 [ 0.14; 0.69]
## migl:piog 0 0.00 -0.12 [-0.41; 0.17] . . -0.12 [-0.41; 0.17]
## migl:plac 3 1.00 0.94 [ 0.70; 1.19] 0.94 [ 0.70; 1.19] . .
## migl:rosi 0 0.00 -0.26 [-0.52; 0.01] . . -0.26 [-0.52; 0.01]
## migl:sita 0 0.00 0.37 [ 0.02; 0.73] . . 0.37 [ 0.02; 0.73]
## migl:sulf 0 0.00 0.50 [ 0.20; 0.81] . . 0.50 [ 0.20; 0.81]
## migl:vild 0 0.00 0.24 [-0.11; 0.60] . . 0.24 [-0.11; 0.60]
## piog:plac 1 0.36 1.07 [ 0.92; 1.22] 1.30 [ 1.05; 1.55] 0.94 [ 0.75; 1.12]
## piog:rosi 1 0.20 -0.14 [-0.30; 0.02] -0.10 [-0.46; 0.26] -0.14 [-0.32; 0.03]
## piog:sita 0 0.00 0.50 [ 0.20; 0.79] . . 0.50 [ 0.20; 0.79]
## piog:sulf 0 0.00 0.63 [ 0.42; 0.84] . . 0.63 [ 0.42; 0.84]
## piog:vild 0 0.00 0.37 [ 0.08; 0.66] . . 0.37 [ 0.08; 0.66]
## plac:rosi 6 0.83 -1.20 [-1.30; -1.11] -1.15 [-1.25; -1.05] -1.47 [-1.69; -1.24]
## plac:sita 1 1.00 -0.57 [-0.82; -0.32] -0.57 [-0.82; -0.32] . .
## plac:sulf 0 0.00 -0.44 [-0.62; -0.26] . . -0.44 [-0.62; -0.26]
## plac:vild 1 1.00 -0.70 [-0.95; -0.45] -0.70 [-0.95; -0.45] . .
## rosi:sita 0 0.00 0.63 [ 0.36; 0.90] . . 0.63 [ 0.36; 0.90]
## rosi:sulf 1 0.41 0.76 [ 0.58; 0.94] 1.20 [ 0.92; 1.48] 0.46 [ 0.22; 0.69]
## rosi:vild 0 0.00 0.50 [ 0.24; 0.77] . . 0.50 [ 0.24; 0.77]
## sita:sulf 0 0.00 0.13 [-0.18; 0.44] . . 0.13 [-0.18; 0.44]
## sita:vild 0 0.00 -0.13 [-0.49; 0.23] . . -0.13 [-0.49; 0.23]
## sulf:vild 0 0.00 -0.26 [-0.57; 0.05] . . -0.26 [-0.57; 0.05]
##
## Random effects model:
##
## comparison k prop nma 95%-CI direct 95%-CI indir. 95%-CI
## acar:benf 0 0.00 0.11 [-0.63; 0.85] . . 0.11 [-0.63; 0.85]
## acar:metf 1 0.28 -0.28 [-0.79; 0.22] -0.20 [-1.15; 0.75] -0.32 [-0.91; 0.28]
## acar:migl 0 0.00 -0.11 [-0.77; 0.55] . . -0.11 [-0.77; 0.55]
## acar:piog 0 0.00 -0.29 [-0.91; 0.33] . . -0.29 [-0.91; 0.33]
## acar:plac 2 0.65 0.84 [ 0.36; 1.32] 0.86 [ 0.26; 1.45] 0.81 [ 0.00; 1.63]
## acar:rosi 0 0.00 -0.39 [-0.90; 0.12] . . -0.39 [-0.90; 0.12]
## acar:sita 0 0.00 0.27 [-0.57; 1.12] . . 0.27 [-0.57; 1.12]
## acar:sulf 1 0.53 0.43 [-0.10; 0.95] 0.40 [-0.31; 1.11] 0.45 [-0.31; 1.21]
## acar:vild 0 0.00 0.14 [-0.70; 0.99] . . 0.14 [-0.70; 0.99]
## benf:metf 0 0.00 -0.40 [-1.03; 0.24] . . -0.40 [-1.03; 0.24]
## benf:migl 0 0.00 -0.22 [-0.94; 0.50] . . -0.22 [-0.94; 0.50]
## benf:piog 0 0.00 -0.40 [-1.10; 0.31] . . -0.40 [-1.10; 0.31]
## benf:plac 2 1.00 0.73 [ 0.17; 1.29] 0.73 [ 0.17; 1.29] . .
## benf:rosi 0 0.00 -0.50 [-1.12; 0.11] . . -0.50 [-1.12; 0.11]
## benf:sita 0 0.00 0.16 [-0.73; 1.05] . . 0.16 [-0.73; 1.05]
## benf:sulf 0 0.00 0.31 [-0.42; 1.05] . . 0.31 [-0.42; 1.05]
## benf:vild 0 0.00 0.03 [-0.86; 0.92] . . 0.03 [-0.86; 0.92]
## metf:migl 0 0.00 0.18 [-0.37; 0.72] . . 0.18 [-0.37; 0.72]
## metf:piog 1 0.44 -0.00 [-0.44; 0.44] 0.16 [-0.51; 0.83] -0.13 [-0.72; 0.46]
## metf:plac 4 0.56 1.13 [ 0.82; 1.43] 1.18 [ 0.78; 1.58] 1.06 [ 0.60; 1.51]
## metf:rosi 2 0.34 -0.11 [-0.43; 0.22] -0.07 [-0.63; 0.49] -0.12 [-0.52; 0.27]
## metf:sita 0 0.00 0.56 [-0.20; 1.31] . . 0.56 [-0.20; 1.31]
## metf:sulf 1 0.45 0.71 [ 0.25; 1.17] 0.37 [-0.32; 1.06] 0.99 [ 0.37; 1.61]
## metf:vild 0 0.00 0.43 [-0.33; 1.18] . . 0.43 [-0.33; 1.18]
## migl:piog 0 0.00 -0.18 [-0.81; 0.45] . . -0.18 [-0.81; 0.45]
## migl:plac 3 1.00 0.95 [ 0.50; 1.40] 0.95 [ 0.50; 1.40] . .
## migl:rosi 0 0.00 -0.28 [-0.80; 0.23] . . -0.28 [-0.80; 0.23]
## migl:sita 0 0.00 0.38 [-0.45; 1.21] . . 0.38 [-0.45; 1.21]
## migl:sulf 0 0.00 0.53 [-0.12; 1.19] . . 0.53 [-0.12; 1.19]
## migl:vild 0 0.00 0.25 [-0.58; 1.08] . . 0.25 [-0.58; 1.08]
## piog:plac 1 0.39 1.13 [ 0.70; 1.56] 1.30 [ 0.61; 1.99] 1.02 [ 0.47; 1.57]
## piog:rosi 1 0.35 -0.10 [-0.54; 0.33] -0.10 [-0.84; 0.64] -0.11 [-0.65; 0.44]
## piog:sita 0 0.00 0.56 [-0.26; 1.38] . . 0.56 [-0.26; 1.38]
## piog:sulf 0 0.00 0.71 [ 0.12; 1.30] . . 0.71 [ 0.12; 1.30]
## piog:vild 0 0.00 0.43 [-0.39; 1.24] . . 0.43 [-0.39; 1.24]
## plac:rosi 6 0.76 -1.23 [-1.48; -0.98] -1.18 [-1.47; -0.89] -1.40 [-1.91; -0.89]
## plac:sita 1 1.00 -0.57 [-1.26; 0.12] -0.57 [-1.26; 0.12] . .
## plac:sulf 0 0.00 -0.42 [-0.89; 0.06] . . -0.42 [-0.89; 0.06]
## plac:vild 1 1.00 -0.70 [-1.39; -0.01] -0.70 [-1.39; -0.01] . .
## rosi:sita 0 0.00 0.66 [-0.07; 1.40] . . 0.66 [-0.07; 1.40]
## rosi:sulf 1 0.43 0.82 [ 0.35; 1.28] 1.20 [ 0.50; 1.90] 0.52 [-0.10; 1.14]
## rosi:vild 0 0.00 0.53 [-0.20; 1.27] . . 0.53 [-0.20; 1.27]
## sita:sulf 0 0.00 0.15 [-0.69; 0.99] . . 0.15 [-0.69; 0.99]
## sita:vild 0 0.00 -0.13 [-1.11; 0.85] . . -0.13 [-1.11; 0.85]
## sulf:vild 0 0.00 -0.28 [-1.12; 0.55] . . -0.28 [-1.12; 0.55]
##
## Legend:
## comparison - Treatment comparison
## k - Number of studies providing direct evidence
## prop - Direct evidence proportion
## nma - Estimated treatment effect in network meta-analysis
## direct - Estimated treatment effect derived from direct evidence
## indir. - Estimated treatment effect derived from indirect evidence
# Example using continuous outcomes (internal call of function metacont)
data(parkinson)
# Transform data from arm-based format to contrast-based format
p1 <- pairwise(list(Treatment1, Treatment2, Treatment3),
n=list(n1, n2, n3),
mean=list(y1, y2, y3),
sd=list(sd1, sd2, sd3),
data=parkinson, studlab=Study)
p1
## TE seTE studlab treat1 treat2 n1 mean1 sd1 n2 mean2 sd2
## 1 0.31 0.6680897 1 1 3 54 -1.22 3.70 95 -1.53 4.28
## 2 1.70 0.3826406 2 1 2 172 -0.70 3.70 173 -2.40 3.40
## 3 2.30 0.7177460 3 1 2 76 -0.30 4.40 71 -2.60 4.30
## 4 0.90 0.6949881 3 1 4 76 -0.30 4.40 81 -1.20 4.30
## 5 -1.40 0.6990666 3 2 4 71 -2.60 4.30 81 -1.20 4.30
## 6 0.35 0.4419417 4 3 4 128 -0.24 3.00 72 -0.59 3.00
## 7 -0.55 0.5551146 5 3 4 80 -0.73 3.00 46 -0.18 3.00
## 8 0.30 0.2742763 6 4 5 137 -2.20 2.31 131 -2.50 2.18
## 9 0.30 0.3200872 7 4 5 154 -1.80 2.48 143 -2.10 2.99
# Conduct network meta-analysis
net1 <- netmeta(p1)
net1
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## 1 1 3 0.31 0.6681 0.6681 2
## 2 1 2 1.70 0.3826 0.3826 2
## 3 1 2 2.30 0.7177 0.8976 3 *
## 3 1 4 0.90 0.6950 0.8404 3 *
## 3 2 4 -1.40 0.6991 0.8497 3 *
## 4 3 4 0.35 0.4419 0.4419 2
## 5 3 4 -0.55 0.5551 0.5551 2
## 6 4 5 0.30 0.2743 0.2743 2
## 7 4 5 0.30 0.3201 0.3201 2
##
## Number of treatment arms (by study):
## narms
## 1 2
## 2 2
## 3 3
## 4 2
## 5 2
## 6 2
## 7 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q leverage
## 1 1 3 0.4781 [-0.4757; 1.4319] 0.06 0.53
## 2 1 2 1.8116 [ 1.1595; 2.4636] 0.08 0.76
## 3 1 2 1.8116 [ 1.1595; 2.4636] 0.30 0.14
## 3 1 4 0.5240 [-0.4141; 1.4621] 0.20 0.32
## 3 2 4 -1.2876 [-2.3110; -0.2641] 0.02 0.38
## 4 3 4 0.0459 [-0.5877; 0.6795] 0.47 0.54
## 5 3 4 0.0459 [-0.5877; 0.6795] 1.15 0.34
## 6 4 5 0.3000 [-0.1082; 0.7082] 0.00 0.58
## 7 4 5 0.3000 [-0.1082; 0.7082] 0.00 0.42
##
## Number of studies: k = 7
## Number of treatments: n = 5
## Number of pairwise comparisons: m = 9
## Number of designs: d = 5
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## 1 2 3 4 5
## 1 . 1.8116 0.4781 0.5240 0.8240
## 2 -1.8116 . -1.3334 -1.2876 -0.9876
## 3 -0.4781 1.3334 . 0.0459 0.3459
## 4 -0.5240 1.2876 -0.0459 . 0.3000
## 5 -0.8240 0.9876 -0.3459 -0.3000 .
##
## Lower 95%-confidence limit:
## 1 2 3 4 5
## 1 . 1.1595 -0.4757 -0.4141 -0.1991
## 2 -2.4636 . -2.3972 -2.3110 -2.0894
## 3 -1.4319 0.2697 . -0.5877 -0.4079
## 4 -1.4621 0.2641 -0.6795 . -0.1082
## 5 -1.8471 -0.1143 -1.0996 -0.7082 .
##
## Upper 95%-confidence limit:
## 1 2 3 4 5
## 1 . 2.4636 1.4319 1.4621 1.8471
## 2 -1.1595 . -0.2697 -0.2641 0.1143
## 3 0.4757 2.3972 . 0.6795 1.0996
## 4 0.4141 2.3110 0.5877 . 0.7082
## 5 0.1991 2.0894 0.4079 0.1082 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0; I^2 = 0%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 2.29 4 0.6830
## Within designs 1.61 2 0.4473
## Between designs 0.68 2 0.7121
# Draw network graphs
netgraph(net1, points=TRUE, cex.points=3, cex=1.5,
thickness="se.fixed")
netgraph(net1, points=TRUE, cex.points=3, cex = 1.5,
plastic=TRUE, thickness="se.fixed",
iterate=TRUE)
netgraph(net1, points=TRUE, cex.points=3, cex = 1.5,
plastic=TRUE, thickness="se.fixed",
iterate=TRUE, start="eigen")
# Example using generic outcomes (internal call of function metagen)
# Calculate standard error for means y1, y2, y3
parkinson$se1 <- with(parkinson, sqrt(sd1^2/n1))
parkinson$se2 <- with(parkinson, sqrt(sd2^2/n2))
parkinson$se3 <- with(parkinson, sqrt(sd3^2/n3))
# Transform data from arm-based format to contrast-based format using
# means and standard errors (note, argument 'sm' has to be used to
# specify that argument 'TE' is a mean difference)
p2 <- pairwise(list(Treatment1, Treatment2, Treatment3),
TE=list(y1, y2, y3),
seTE=list(se1, se2, se3),
data=parkinson, studlab=Study,
sm="MD")
p2
## TE seTE studlab treat1 treat2 TE1 seTE1 TE2 seTE2
## 1 0.31 0.6680897 1 1 3 -1.22 0.5035062 -1.53 0.4391187
## 2 1.70 0.3826406 2 1 2 -0.70 0.2821224 -2.40 0.2584972
## 3 2.30 0.7177460 3 1 2 -0.30 0.5047146 -2.60 0.5103161
## 4 0.90 0.6949881 3 1 4 -0.30 0.5047146 -1.20 0.4777778
## 5 -1.40 0.6990666 3 2 4 -2.60 0.5103161 -1.20 0.4777778
## 6 0.35 0.4419417 4 3 4 -0.24 0.2651650 -0.59 0.3535534
## 7 -0.55 0.5551146 5 3 4 -0.73 0.3354102 -0.18 0.4423259
## 8 0.30 0.2742763 6 4 5 -2.20 0.1973566 -2.50 0.1904675
## 9 0.30 0.3200872 7 4 5 -1.80 0.1998441 -2.10 0.2500364
# Compare pairwise objects p1 (based on continuous outcomes) and p2
# (based on generic outcomes)
all.equal(p1[, c("TE", "seTE", "studlab", "treat1", "treat2")],
p2[, c("TE", "seTE", "studlab", "treat1", "treat2")])
## [1] TRUE
# Same result as network meta-analysis based on continuous outcomes
# (object net1)
net2 <- netmeta(p2)
net2
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## 1 1 3 0.31 0.6681 0.6681 2
## 2 1 2 1.70 0.3826 0.3826 2
## 3 1 2 2.30 0.7177 0.8976 3 *
## 3 1 4 0.90 0.6950 0.8404 3 *
## 3 2 4 -1.40 0.6991 0.8497 3 *
## 4 3 4 0.35 0.4419 0.4419 2
## 5 3 4 -0.55 0.5551 0.5551 2
## 6 4 5 0.30 0.2743 0.2743 2
## 7 4 5 0.30 0.3201 0.3201 2
##
## Number of treatment arms (by study):
## narms
## 1 2
## 2 2
## 3 3
## 4 2
## 5 2
## 6 2
## 7 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q leverage
## 1 1 3 0.4781 [-0.4757; 1.4319] 0.06 0.53
## 2 1 2 1.8116 [ 1.1595; 2.4636] 0.08 0.76
## 3 1 2 1.8116 [ 1.1595; 2.4636] 0.30 0.14
## 3 1 4 0.5240 [-0.4141; 1.4621] 0.20 0.32
## 3 2 4 -1.2876 [-2.3110; -0.2641] 0.02 0.38
## 4 3 4 0.0459 [-0.5877; 0.6795] 0.47 0.54
## 5 3 4 0.0459 [-0.5877; 0.6795] 1.15 0.34
## 6 4 5 0.3000 [-0.1082; 0.7082] 0.00 0.58
## 7 4 5 0.3000 [-0.1082; 0.7082] 0.00 0.42
##
## Number of studies: k = 7
## Number of treatments: n = 5
## Number of pairwise comparisons: m = 9
## Number of designs: d = 5
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## 1 2 3 4 5
## 1 . 1.8116 0.4781 0.5240 0.8240
## 2 -1.8116 . -1.3334 -1.2876 -0.9876
## 3 -0.4781 1.3334 . 0.0459 0.3459
## 4 -0.5240 1.2876 -0.0459 . 0.3000
## 5 -0.8240 0.9876 -0.3459 -0.3000 .
##
## Lower 95%-confidence limit:
## 1 2 3 4 5
## 1 . 1.1595 -0.4757 -0.4141 -0.1991
## 2 -2.4636 . -2.3972 -2.3110 -2.0894
## 3 -1.4319 0.2697 . -0.5877 -0.4079
## 4 -1.4621 0.2641 -0.6795 . -0.1082
## 5 -1.8471 -0.1143 -1.0996 -0.7082 .
##
## Upper 95%-confidence limit:
## 1 2 3 4 5
## 1 . 2.4636 1.4319 1.4621 1.8471
## 2 -1.1595 . -0.2697 -0.2641 0.1143
## 3 0.4757 2.3972 . 0.6795 1.0996
## 4 0.4141 2.3110 0.5877 . 0.7082
## 5 0.1991 2.0894 0.4079 0.1082 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0; I^2 = 0%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 2.29 4 0.6830
## Within designs 1.61 2 0.4473
## Between designs 0.68 2 0.7121
# Example with binary data
data(smokingcessation)
# Transform data from arm-based format to contrast-based format
# (interal call of metabin function). Argument 'sm' has to be used for
# odds ratio as risk ratio (sm="RR") is default of metabin function.
p3 <- pairwise(list(treat1, treat2, treat3),
list(event1, event2, event3),
list(n1, n2, n3),
data=smokingcessation,
sm="OR")
p3
## TE seTE studlab treat1 treat2 event1 n1 event2 n2
## 1 -1.051293027 0.4132432 1 A C 9 140 23 140
## 2 -0.128527575 0.4759803 1 A D 9 140 10 138
## 3 0.922765452 0.3997972 1 C D 23 140 10 138
## 4 -0.001244555 0.4504070 2 B C 11 78 12 85
## 5 -0.225333286 0.3839393 2 B D 11 78 29 170
## 6 -0.224088731 0.3722995 2 C D 12 85 29 170
## 7 -2.202289286 0.1430439 3 A C 75 731 363 714
## 8 -0.870353637 0.7910933 4 A C 2 106 9 205
## 9 -0.415648522 0.1557329 5 A C 58 549 237 1561
## 10 -2.779683746 1.4698402 6 A C 0 33 9 48
## 11 -2.705393275 0.6251608 7 A C 3 100 31 98
## 12 -2.425187415 1.0422512 8 A C 1 31 26 95
## 13 -0.443616874 0.5219769 9 A C 6 39 17 77
## 14 0.015964936 0.1699150 10 A B 79 702 77 694
## 15 -0.393504544 0.3265754 11 A B 18 671 21 535
## 16 -0.390412268 0.1680177 12 A C 64 642 107 761
## 17 -0.106335650 0.5955997 13 A C 5 62 8 90
## 18 -0.583398287 0.2983467 14 A C 20 234 34 237
## 19 -3.522516830 1.4969970 15 A D 0 20 9 20
## 20 -0.679594214 0.4411158 16 A B 8 116 19 149
## 21 -0.539679846 0.1401199 17 A C 95 1107 143 1031
## 22 0.125505082 0.3199924 18 A C 15 187 36 504
## 23 0.239970162 0.1736564 19 A C 78 584 73 675
## 24 -0.038956007 0.1873842 20 A C 69 1177 54 888
## 25 0.151684587 0.4289753 21 B C 20 49 16 43
## 26 -1.043486306 0.4489795 22 B D 7 66 32 127
## 27 -0.680724661 0.4092394 23 C D 12 76 20 74
## 28 0.405465108 0.7139060 24 C D 9 55 3 26
## incr allstudies
## 1 0.0 FALSE
## 2 0.0 FALSE
## 3 0.0 FALSE
## 4 0.0 FALSE
## 5 0.0 FALSE
## 6 0.0 FALSE
## 7 0.0 FALSE
## 8 0.0 FALSE
## 9 0.0 FALSE
## 10 0.5 FALSE
## 11 0.0 FALSE
## 12 0.0 FALSE
## 13 0.0 FALSE
## 14 0.0 FALSE
## 15 0.0 FALSE
## 16 0.0 FALSE
## 17 0.0 FALSE
## 18 0.0 FALSE
## 19 0.5 FALSE
## 20 0.0 FALSE
## 21 0.0 FALSE
## 22 0.0 FALSE
## 23 0.0 FALSE
## 24 0.0 FALSE
## 25 0.0 FALSE
## 26 0.0 FALSE
## 27 0.0 FALSE
## 28 0.0 FALSE
# Conduct network meta-analysis
net3 <- netmeta(p3)
net3
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## 1 A C -1.0513 0.4132 0.4776 3 *
## 1 A D -0.1285 0.4760 0.6875 3 *
## 1 C D 0.9228 0.3998 0.4551 3 *
## 2 B C -0.0012 0.4504 0.6707 3 *
## 2 B D -0.2253 0.3839 0.4390 3 *
## 2 C D -0.2241 0.3723 0.4204 3 *
## 3 A C -2.2023 0.1430 0.1430 2
## 4 A C -0.8704 0.7911 0.7911 2
## 5 A C -0.4156 0.1557 0.1557 2
## 6 A C -2.7797 1.4698 1.4698 2
## 7 A C -2.7054 0.6252 0.6252 2
## 8 A C -2.4252 1.0423 1.0423 2
## 9 A C -0.4436 0.5220 0.5220 2
## 10 A B 0.0160 0.1699 0.1699 2
## 11 A B -0.3935 0.3266 0.3266 2
## 12 A C -0.3904 0.1680 0.1680 2
## 13 A C -0.1063 0.5956 0.5956 2
## 14 A C -0.5834 0.2983 0.2983 2
## 15 A D -3.5225 1.4970 1.4970 2
## 16 A B -0.6796 0.4411 0.4411 2
## 17 A C -0.5397 0.1401 0.1401 2
## 18 A C 0.1255 0.3200 0.3200 2
## 19 A C 0.2400 0.1737 0.1737 2
## 20 A C -0.0390 0.1874 0.1874 2
## 21 B C 0.1517 0.4290 0.4290 2
## 22 B D -1.0435 0.4490 0.4490 2
## 23 C D -0.6807 0.4092 0.4092 2
## 24 C D 0.4055 0.7139 0.7139 2
##
## Number of treatment arms (by study):
## narms
## 1 3
## 10 2
## 11 2
## 12 2
## 13 2
## 14 2
## 15 2
## 16 2
## 17 2
## 18 2
## 19 2
## 2 3
## 20 2
## 21 2
## 22 2
## 23 2
## 24 2
## 3 2
## 4 2
## 5 2
## 6 2
## 7 2
## 8 2
## 9 2
##
## Results (fixed effect model):
##
## treat1 treat2 OR 95%-CI Q leverage
## 1 A C 0.5208 [0.4640; 0.5846] 0.70 0.02
## 1 A D 0.4883 [0.3379; 0.7057] 0.73 0.07
## 1 C D 0.9376 [0.6526; 1.3472] 4.71 0.17
## 2 B C 0.6359 [0.4894; 0.8263] 0.45 0.04
## 2 B D 0.5963 [0.4030; 0.8822] 0.44 0.21
## 2 C D 0.9376 [0.6526; 1.3472] 0.14 0.19
## 3 A C 0.5208 [0.4640; 0.5846] 117.39 0.17
## 4 A C 0.5208 [0.4640; 0.5846] 0.08 0.01
## 5 A C 0.5208 [0.4640; 0.5846] 2.31 0.14
## 6 A C 0.5208 [0.4640; 0.5846] 2.09 0.00
## 7 A C 0.5208 [0.4640; 0.5846] 10.78 0.01
## 8 A C 0.5208 [0.4640; 0.5846] 2.89 0.00
## 9 A C 0.5208 [0.4640; 0.5846] 0.16 0.01
## 10 A B 0.8189 [0.6397; 1.0483] 1.61 0.55
## 11 A B 0.8189 [0.6397; 1.0483] 0.35 0.15
## 12 A C 0.5208 [0.4640; 0.5846] 2.43 0.12
## 13 A C 0.5208 [0.4640; 0.5846] 0.84 0.01
## 14 A C 0.5208 [0.4640; 0.5846] 0.05 0.04
## 15 A D 0.4883 [0.3379; 0.7057] 3.51 0.02
## 16 A B 0.8189 [0.6397; 1.0483] 1.18 0.08
## 17 A C 0.5208 [0.4640; 0.5846] 0.65 0.18
## 18 A C 0.5208 [0.4640; 0.5846] 5.91 0.03
## 19 A C 0.5208 [0.4640; 0.5846] 26.41 0.12
## 20 A C 0.5208 [0.4640; 0.5846] 10.72 0.10
## 21 B C 0.6359 [0.4894; 0.8263] 1.98 0.10
## 22 B D 0.5963 [0.4030; 0.8822] 1.37 0.20
## 23 C D 0.9376 [0.6526; 1.3472] 2.27 0.20
## 24 C D 0.9376 [0.6526; 1.3472] 0.43 0.07
##
## Number of studies: k = 24
## Number of treatments: n = 4
## Number of pairwise comparisons: m = 28
## Number of designs: d = 8
##
## Fixed effect model
##
## Treatment estimate (sm = 'OR'):
## A B C D
## A . 0.8189 0.5208 0.4883
## B 1.2211 . 0.6359 0.5963
## C 1.9202 1.5725 . 0.9376
## D 2.0479 1.6771 1.0665 .
##
## Lower 95%-confidence limit:
## A B C D
## A . 0.6397 0.4640 0.3379
## B 0.9539 . 0.4894 0.4030
## C 1.7107 1.2102 . 0.6526
## D 1.4170 1.1335 0.7423 .
##
## Upper 95%-confidence limit:
## A B C D
## A . 1.0483 0.5846 0.7057
## B 1.5631 . 0.8263 0.8822
## C 2.1554 2.0433 . 1.3472
## D 2.9598 2.4814 1.5323 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.5989; I^2 = 88.6%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 202.62 23 < 0.0001
## Within designs 187.40 16 < 0.0001
## Between designs 15.22 7 0.0333
# Example with incidence rates
data(dietaryfat)
# Transform data from arm-based format to contrast-based format
p4 <- pairwise(list(treat1, treat2, treat3),
list(d1, d2, d3),
time=list(years1, years2, years3),
studlab=ID,
data=dietaryfat)
p4
## TE seTE studlab treat1 treat2 event1
## 1 0.02202212 0.13363595 2 DART 1 2 113
## 2 -1.66363256 1.09544512 10 London Corn /Olive 1 2 1
## 3 -1.23608328 1.15470054 10 London Corn /Olive 1 3 1
## 4 0.42754928 0.73029674 10 London Corn /Olive 2 3 5
## 5 0.13122887 0.30276504 11 London Low Fat 1 2 24
## 6 -0.05863541 0.08803255 14 Minnesota Coronary 1 2 248
## 7 0.15090580 0.26071507 15 MRC Soya 1 2 31
## 8 0.31442233 0.19031014 18 Oslo Diet-Heart 1 2 65
## 9 1.13441029 1.15470054 22 STARS 1 2 3
## 10 -0.40523688 0.24770004 23 Sydney Diet-Heart 1 2 28
## 11 0.04519334 0.10675600 26 Veterans Administration 1 2 177
## 12 0.67701780 1.22474487 27 Veterans Diet & Skin CA 1 2 2
## time1 event2 time2 incr
## 1 1917.0 111 1925.0 0
## 2 43.6 5 41.3 0
## 3 43.6 3 38.0 0
## 4 41.3 3 38.0 0
## 5 393.5 20 373.9 0
## 6 4715.0 269 4823.0 0
## 7 715.0 28 751.0 0
## 8 885.0 48 895.0 0
## 9 87.8 1 91.0 0
## 10 1011.0 39 939.0 0
## 11 1544.0 174 1588.0 0
## 12 125.0 1 123.0 0
# Conduct network meta-analysis using incidence rate ratios (sm="IRR").
# Note, the argument 'sm' is not necessary as this is the default in R
# function metainc called internally
net4 <- netmeta(p4, sm="IRR")
summary(net4)
## Number of studies: k = 10
## Number of treatments: n = 3
## Number of pairwise comparisons: m = 12
## Number of designs: d = 2
##
## Fixed effect model
##
## Treatment estimate (sm = 'IRR'):
## 1 2 3
## 1 . 1.0096 1.1714
## 2 0.9905 . 1.1603
## 3 0.8537 0.8618 .
##
## Lower 95%-confidence limit:
## 1 2 3
## 1 . 0.9084 0.2921
## 2 0.8913 . 0.2902
## 3 0.2129 0.2155 .
##
## Upper 95%-confidence limit:
## 1 2 3
## 1 . 1.1220 4.6969
## 2 1.1008 . 4.6400
## 3 3.4229 3.4464 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.0043; I^2 = 11.1%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 10.12 9 0.3405
## Within designs 7.79 8 0.4546
## Between designs 2.34 1 0.1262
# Example with long data format
data(Woods2010)
# Transform data from long arm-based format to contrast-based format
# Argument 'sm' has to be used for odds ratio as summary measure; by
# default the risk ratio is used in the metabin function called
# internally.
p5 <- pairwise(treatment, event=r, n=N,
studlab=author, data=Woods2010, sm="OR")
p5
## TE seTE studlab treat1 treat2 event1 n1
## 1 0.00881063 1.4173252 Boyd 1997 Placebo Salmeterol 1 227
## 2 -0.60382188 0.6311772 Calverly 2003 Fluticasone Placebo 4 374
## 3 0.28497571 0.7672979 Calverly 2003 Fluticasone Salmeterol 4 374
## 4 0.65457491 0.8692018 Calverly 2003 Fluticasone SFC 4 374
## 5 0.88879759 0.6940644 Calverly 2003 Placebo Salmeterol 7 361
## 6 1.25839679 0.8052894 Calverly 2003 Placebo SFC 7 361
## 7 0.36959919 0.9158888 Calverly 2003 Salmeterol SFC 3 372
## 8 1.41751820 1.2270043 Celli 2003 Placebo Salmeterol 2 270
## event2 n2 incr allstudies
## 1 1 229 0 FALSE
## 2 7 361 0 FALSE
## 3 3 372 0 FALSE
## 4 2 358 0 FALSE
## 5 3 372 0 FALSE
## 6 2 358 0 FALSE
## 7 2 358 0 FALSE
## 8 1 554 0 FALSE
# Conduct network meta-analysis
net5 <- netmeta(p5)
net5
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms
## Boyd 1997 Placebo Salmeterol 0.0088 1.4173 1.4173 2
## Calverly 2003 Fluticasone Placebo -0.6038 0.6312 0.7623 4
## Calverly 2003 Fluticasone Salmeterol 0.2850 0.7673 1.1578 4
## Calverly 2003 Fluticasone SFC 0.6546 0.8692 1.4163 4
## Calverly 2003 Placebo Salmeterol 0.8888 0.6941 0.8791 4
## Calverly 2003 Placebo SFC 1.2584 0.8053 1.0753 4
## Calverly 2003 Salmeterol SFC 0.3696 0.9159 1.6332 4
## Celli 2003 Placebo Salmeterol 1.4175 1.2270 1.2270 2
## multiarm
## Boyd 1997
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Celli 2003
##
## Number of treatment arms (by study):
## narms
## Boyd 1997 2
## Calverly 2003 4
## Celli 2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 OR 95%-CI Q
## Boyd 1997 Placebo Salmeterol 2.3678 [0.7967; 7.0370] 0.36
## Calverly 2003 Fluticasone Placebo 0.5512 [0.1640; 1.8526] 0.00
## Calverly 2003 Fluticasone Salmeterol 1.3051 [0.3243; 5.2517] 0.00
## Calverly 2003 Fluticasone SFC 1.9243 [0.3503; 10.5717] 0.00
## Calverly 2003 Placebo Salmeterol 2.3678 [0.7967; 7.0370] 0.00
## Calverly 2003 Placebo SFC 3.4913 [0.7344; 16.5976] 0.00
## Calverly 2003 Salmeterol SFC 1.4745 [0.2686; 8.0933] 0.00
## Celli 2003 Placebo Salmeterol 2.3678 [0.7967; 7.0370] 0.21
## leverage
## Boyd 1997 0.15
## Calverly 2003 0.66
## Calverly 2003 0.38
## Calverly 2003 0.38
## Calverly 2003 0.40
## Calverly 2003 0.55
## Calverly 2003 0.28
## Celli 2003 0.21
##
## Number of studies: k = 3
## Number of treatments: n = 4
## Number of pairwise comparisons: m = 8
## Number of designs: d = 2
##
## Fixed effect model
##
## Treatment estimate (sm = 'OR'):
## Fluticasone Placebo Salmeterol SFC
## Fluticasone . 0.5512 1.3051 1.9243
## Placebo 1.8143 . 2.3678 3.4913
## Salmeterol 0.7662 0.4223 . 1.4745
## SFC 0.5197 0.2864 0.6782 .
##
## Lower 95%-confidence limit:
## Fluticasone Placebo Salmeterol SFC
## Fluticasone . 0.1640 0.3243 0.3503
## Placebo 0.5398 . 0.7967 0.7344
## Salmeterol 0.1904 0.1421 . 0.2686
## SFC 0.0946 0.0602 0.1236 .
##
## Upper 95%-confidence limit:
## Fluticasone Placebo Salmeterol SFC
## Fluticasone . 1.8526 5.2517 10.5717
## Placebo 6.0982 . 7.0370 16.5976
## Salmeterol 3.0834 1.2552 . 8.0933
## SFC 2.8549 1.3616 3.7225 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0; I^2 = 0%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 0.57 2 0.7525
## Within designs 0.56 1 0.4524
## Between designs 0.00 1 0.9485
data(parkinson)
# Transform data from arm-based format to contrast-based format
p1 <- pairwise(list(Treatment1, Treatment2, Treatment3),
n=list(n1, n2, n3),
mean=list(y1, y2, y3),
sd=list(sd1, sd2, sd3),
data=parkinson, studlab=Study)
p1
## TE seTE studlab treat1 treat2 n1 mean1 sd1 n2 mean2 sd2
## 1 0.31 0.6680897 1 1 3 54 -1.22 3.70 95 -1.53 4.28
## 2 1.70 0.3826406 2 1 2 172 -0.70 3.70 173 -2.40 3.40
## 3 2.30 0.7177460 3 1 2 76 -0.30 4.40 71 -2.60 4.30
## 4 0.90 0.6949881 3 1 4 76 -0.30 4.40 81 -1.20 4.30
## 5 -1.40 0.6990666 3 2 4 71 -2.60 4.30 81 -1.20 4.30
## 6 0.35 0.4419417 4 3 4 128 -0.24 3.00 72 -0.59 3.00
## 7 -0.55 0.5551146 5 3 4 80 -0.73 3.00 46 -0.18 3.00
## 8 0.30 0.2742763 6 4 5 137 -2.20 2.31 131 -2.50 2.18
## 9 0.30 0.3200872 7 4 5 154 -1.80 2.48 143 -2.10 2.99
# Conduct network meta-analysis
net1 <- netmeta(p1)
net1
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## 1 1 3 0.31 0.6681 0.6681 2
## 2 1 2 1.70 0.3826 0.3826 2
## 3 1 2 2.30 0.7177 0.8976 3 *
## 3 1 4 0.90 0.6950 0.8404 3 *
## 3 2 4 -1.40 0.6991 0.8497 3 *
## 4 3 4 0.35 0.4419 0.4419 2
## 5 3 4 -0.55 0.5551 0.5551 2
## 6 4 5 0.30 0.2743 0.2743 2
## 7 4 5 0.30 0.3201 0.3201 2
##
## Number of treatment arms (by study):
## narms
## 1 2
## 2 2
## 3 3
## 4 2
## 5 2
## 6 2
## 7 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q leverage
## 1 1 3 0.4781 [-0.4757; 1.4319] 0.06 0.53
## 2 1 2 1.8116 [ 1.1595; 2.4636] 0.08 0.76
## 3 1 2 1.8116 [ 1.1595; 2.4636] 0.30 0.14
## 3 1 4 0.5240 [-0.4141; 1.4621] 0.20 0.32
## 3 2 4 -1.2876 [-2.3110; -0.2641] 0.02 0.38
## 4 3 4 0.0459 [-0.5877; 0.6795] 0.47 0.54
## 5 3 4 0.0459 [-0.5877; 0.6795] 1.15 0.34
## 6 4 5 0.3000 [-0.1082; 0.7082] 0.00 0.58
## 7 4 5 0.3000 [-0.1082; 0.7082] 0.00 0.42
##
## Number of studies: k = 7
## Number of treatments: n = 5
## Number of pairwise comparisons: m = 9
## Number of designs: d = 5
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## 1 2 3 4 5
## 1 . 1.8116 0.4781 0.5240 0.8240
## 2 -1.8116 . -1.3334 -1.2876 -0.9876
## 3 -0.4781 1.3334 . 0.0459 0.3459
## 4 -0.5240 1.2876 -0.0459 . 0.3000
## 5 -0.8240 0.9876 -0.3459 -0.3000 .
##
## Lower 95%-confidence limit:
## 1 2 3 4 5
## 1 . 1.1595 -0.4757 -0.4141 -0.1991
## 2 -2.4636 . -2.3972 -2.3110 -2.0894
## 3 -1.4319 0.2697 . -0.5877 -0.4079
## 4 -1.4621 0.2641 -0.6795 . -0.1082
## 5 -1.8471 -0.1143 -1.0996 -0.7082 .
##
## Upper 95%-confidence limit:
## 1 2 3 4 5
## 1 . 2.4636 1.4319 1.4621 1.8471
## 2 -1.1595 . -0.2697 -0.2641 0.1143
## 3 0.4757 2.3972 . 0.6795 1.0996
## 4 0.4141 2.3110 0.5877 . 0.7082
## 5 0.1991 2.0894 0.4079 0.1082 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0; I^2 = 0%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 2.29 4 0.6830
## Within designs 1.61 2 0.4473
## Between designs 0.68 2 0.7121
# Draw network graphs
netgraph(net1, points=TRUE, cex.points=3, cex=1.5,
thickness="se.fixed")
netgraph(net1, points=TRUE, cex.points=3, cex = 1.5,
plastic=TRUE, thickness="se.fixed",
iterate=TRUE)
netgraph(net1, points=TRUE, cex.points=3, cex = 1.5,
plastic=TRUE, thickness="se.fixed",
iterate=TRUE, start="eigen")
data(Linde2015)
# Define order of treatments
trts <- c("TCA", "SSRI", "SNRI", "NRI",
"Low-dose SARI", "NaSSa", "rMAO-A", "Hypericum",
"Placebo")
# Outcome labels
outcomes <- c("Early response", "Early remission")
# (1) Early response
p1 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(resp1, resp2, resp3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 14 TCA SSRI NA 10 NA 11 0 FALSE
## NA NA 18 TCA Placebo NA 73 NA 73 0 FALSE
## NA NA 21 TCA rMAO-A NA 46 NA 98 0 FALSE
## NA NA 27 SSRI Placebo NA 80 NA 81 0 FALSE
## NA NA 51 TCA rMAO-A NA 71 NA 71 0 FALSE
## NA NA 130 TCA NaSSa NA 35 NA 36 0 FALSE
## NA NA 131 SSRI SNRI NA 697 NA 688 0 FALSE
## NA NA 130 TCA Placebo NA 35 NA 34 0 FALSE
## NA NA 130 NaSSa Placebo NA 36 NA 34 0 FALSE
net1 <- netmeta(p1,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 21 TCA rMAO-A NA NA
## 27 SSRI Placebo NA NA
## 51 TCA rMAO-A NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## 131 SSRI SNRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# (2) Early remission
p2 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(remi1, remi2, remi3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr
## NA NA 1 TCA SNRI NA 75 NA 78 0.0
## NA NA 11 TCA SSRI NA 108 NA 99 0.0
## NA NA 14 TCA SSRI NA 10 NA 11 0.0
## NA NA 18 TCA Placebo NA 73 NA 73 0.0
## NA NA 20 TCA SSRI NA 55 NA 51 0.0
## NA NA 26 SSRI Placebo NA 314 NA 154 0.0
## NA NA 53 SSRI NaSSa NA 122 NA 121 0.0
## NA NA 56 TCA SSRI NA 92 NA 380 0.0
## NA NA 73 Hypericum Placebo NA 55 NA 57 0.0
## NA NA 90 TCA SSRI NA 42 NA 42 0.0
## NA NA 96 TCA SSRI 0 30 0 29 0.5
## NA NA 121 Low-dose SARI NaSSa NA 43 NA 40 0.0
## NA NA 130 TCA NaSSa NA 35 NA 36 0.0
## NA NA 1 TCA Placebo NA 75 NA 76 0.0
## NA NA 11 TCA Placebo NA 108 NA 101 0.0
## NA NA 53 SSRI Placebo NA 122 NA 129 0.0
## NA NA 130 TCA Placebo NA 35 NA 34 0.0
## NA NA 1 SNRI Placebo NA 78 NA 76 0.0
## NA NA 11 SSRI Placebo NA 99 NA 101 0.0
## NA NA 53 NaSSa Placebo NA 121 NA 129 0.0
## NA NA 130 NaSSa Placebo NA 36 NA 34 0.0
## allstudies
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
## FALSE
net2 <- netmeta(p2,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 1 TCA Placebo NA NA
## 1 TCA SNRI NA NA
## 1 SNRI Placebo NA NA
## 11 SSRI Placebo NA NA
## 11 TCA Placebo NA NA
## 11 TCA SSRI NA NA
## 14 TCA SSRI NA NA
## 18 TCA Placebo NA NA
## 20 TCA SSRI NA NA
## 26 SSRI Placebo NA NA
## 53 NaSSa Placebo NA NA
## 53 SSRI NaSSa NA NA
## 53 SSRI Placebo NA NA
## 56 TCA SSRI NA NA
## 73 Hypericum Placebo NA NA
## 90 TCA SSRI NA NA
## 96 TCA SSRI NA NA
## 121 Low-dose SARI NaSSa NA NA
## 130 NaSSa Placebo NA NA
## 130 TCA NaSSa NA NA
## 130 TCA Placebo NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Partial order of treatment rankings
po2 <- netposet(netrank(net1, small.values = "bad"),
netrank(net2, small.values = "bad"),
outcomes = outcomes)
# Scatter plot
plot(po2)
# Same scatter plot as only two outcomes considered in netposet()
plot(po2, "biplot")
## Warning in plot.netposet(po2, "biplot"): Scatter plot instead of biplot
## generated as only two outcomes considered in netposet().
# Consider three outcomes
# Outcome labels
outcomes <- c("Early response", "Early remission", "Lost to follow-up")
# (3) Loss to follow-up
p3 <- pairwise(treat = list(treatment1, treatment2, treatment3),
event = list(loss1, loss2, loss3),
n = list(n1, n2, n3),
studlab = id, data = Linde2015, sm = "OR")
## Warning: Comparisons with missing TE / seTE or zero seTE will not be
## considered in network meta-analysis.
## Comparisons will not be considered in network meta-analysis:
## TE seTE studlab treat1 treat2 event1 n1 event2 n2 incr allstudies
## NA NA 9 TCA SSRI NA 363 NA 173 0 FALSE
## NA NA 14 TCA SSRI NA 10 NA 11 0 FALSE
## NA NA 48 SSRI Placebo 31 191 NA 189 0 FALSE
## NA NA 53 SSRI NaSSa NA 122 NA 121 0 FALSE
## NA NA 90 TCA SSRI NA 42 NA 42 0 FALSE
## NA NA 116 SSRI NaSSa NA 31 NA 31 0 FALSE
## NA NA 120 TCA SSRI NA 218 NA 109 0 FALSE
## NA NA 53 SSRI Placebo NA 122 NA 129 0 FALSE
## NA NA 53 NaSSa Placebo NA 121 NA 129 0 FALSE
net3 <- netmeta(p3,
comb.fixed = FALSE, comb.random = TRUE,
seq = trts, ref = "Placebo")
## Warning: Comparisons with missing TE / seTE or zero seTE not considered in
## network meta-analysis.
## Comparisons not considered in network meta-analysis:
## studlab treat1 treat2 TE seTE
## 9 TCA SSRI NA NA
## 14 TCA SSRI NA NA
## 48 SSRI Placebo NA NA
## 53 NaSSa Placebo NA NA
## 53 SSRI NaSSa NA NA
## 53 SSRI Placebo NA NA
## 90 TCA SSRI NA NA
## 116 SSRI NaSSa NA NA
## 120 TCA SSRI NA NA
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
# Partial order of treatment rankings (with three outcomes)
po3 <- netposet(netrank(net1, small.values = "bad"),
netrank(net2, small.values = "bad"),
netrank(net3, small.values = "good"),
outcomes = outcomes)
# Scatter plot
plot(po3)
# Biplot (reverse limits of y-axis as biplot is upside down)
plot(po3, "bi", xlim = c(-1, 1.7), ylim = c(2.5, -2.5))
data(Senn2013)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
print(decomp.design(net1))
## Q statistics to assess homogeneity / consistency
##
## Q df p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
##
## Design-specific decomposition of within-designs Q statistic
##
## Design Q df p-value
## benf:plac 4.38 1 0.0363
## metf:plac 42.16 2 < 0.0001
## metf:rosi 0.19 1 0.6655
## migl:plac 6.45 2 0.0398
## plac:rosi 21.27 5 0.0007
##
## Between-designs Q statistic after detaching of single designs
##
## Detached design Q df p-value
## acar:plac 22.44 6 0.0010
## acar:sulf 22.52 6 0.0010
## metf:piog 17.13 6 0.0088
## metf:plac 22.07 6 0.0012
## metf:rosi 22.52 6 0.0010
## metf:sulf 7.51 6 0.2760
## piog:plac 17.25 6 0.0084
## piog:rosi 22.48 6 0.0010
## plac:rosi 16.29 6 0.0123
## rosi:sulf 6.77 6 0.3425
## acar:metf:plac 22.38 5 0.0004
##
## Q statistic to assess consistency under the assumption of
## a full design-by-treatment interaction random effects model
##
## Q df p-value tau.within tau2.within
## Between designs 2.19 7 0.9483 0.3797 0.1442
data(Senn2013)
# Fixed effect model (default)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
print(net1, ref="plac", digits=3)
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.141 0.141 2
## Lewin2007 metf plac -0.82 0.099 0.099 2
## Willms1999 acar metf 0.20 0.358 0.388 3 *
## Davidson2007 plac rosi 1.34 0.144 0.144 2
## Wolffenbuttel1999 plac rosi 1.10 0.114 0.114 2
## Kipnes2001 piog plac -1.30 0.127 0.127 2
## Kerenyi2004 plac rosi 0.77 0.108 0.108 2
## Hanefeld2004 metf piog -0.16 0.085 0.085 2
## Derosa2004 piog rosi 0.10 0.183 0.183 2
## Baksi2004 plac rosi 1.30 0.101 0.101 2
## Rosenstock2008 plac rosi 1.09 0.226 0.226 2
## Zhu2003 plac rosi 1.50 0.162 0.162 2
## Yang2003 metf rosi 0.14 0.224 0.224 2
## Vongthavaravat2002 rosi sulf -1.20 0.144 0.144 2
## Oyama2008 acar sulf -0.40 0.155 0.155 2
## Costa1997 acar plac -0.80 0.143 0.143 2
## Hermansen2007 plac sita 0.57 0.129 0.129 2
## Garber2008 plac vild 0.70 0.127 0.127 2
## Alex1998 metf sulf -0.37 0.118 0.118 2
## Johnston1994 migl plac -0.74 0.184 0.184 2
## Johnston1998a migl plac -1.41 0.224 0.224 2
## Kim2007 metf rosi 0.00 0.234 0.234 2
## Johnston1998b migl plac -0.68 0.283 0.283 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.436 0.436 2
## Stucci1996 benf plac -0.23 0.347 0.347 2
## Moulin2006 benf plac -1.01 0.137 0.137 2
## Willms1999 metf plac -1.20 0.376 0.413 3 *
## Willms1999 acar plac -1.00 0.467 0.824 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q leverage
## DeFronzo1995 metf plac -1.114 [-1.231; -0.997] 30.89 0.18
## Lewin2007 metf plac -1.114 [-1.231; -0.997] 8.79 0.36
## Willms1999 acar metf 0.287 [ 0.062; 0.511] 0.05 0.09
## Davidson2007 plac rosi 1.202 [ 1.108; 1.295] 0.93 0.11
## Wolffenbuttel1999 plac rosi 1.202 [ 1.108; 1.295] 0.80 0.17
## Kipnes2001 piog plac -1.066 [-1.215; -0.918] 3.39 0.36
## Kerenyi2004 plac rosi 1.202 [ 1.108; 1.295] 16.05 0.20
## Hanefeld2004 metf piog -0.048 [-0.184; 0.089] 1.75 0.68
## Derosa2004 piog rosi 0.135 [-0.025; 0.296] 0.04 0.20
## Baksi2004 plac rosi 1.202 [ 1.108; 1.295] 0.94 0.22
## Rosenstock2008 plac rosi 1.202 [ 1.108; 1.295] 0.24 0.04
## Zhu2003 plac rosi 1.202 [ 1.108; 1.295] 3.37 0.09
## Yang2003 metf rosi 0.088 [-0.045; 0.220] 0.05 0.09
## Vongthavaravat2002 rosi sulf -0.762 [-0.943; -0.582] 9.29 0.41
## Oyama2008 acar sulf -0.388 [-0.610; -0.166] 0.01 0.53
## Costa1997 acar plac -0.827 [-1.040; -0.615] 0.04 0.57
## Hermansen2007 plac sita 0.570 [ 0.317; 0.823] 0.00 1.00
## Garber2008 plac vild 0.700 [ 0.450; 0.950] 0.00 1.00
## Alex1998 metf sulf -0.675 [-0.848; -0.501] 6.62 0.56
## Johnston1994 migl plac -0.944 [-1.193; -0.695] 1.23 0.48
## Johnston1998a migl plac -0.944 [-1.193; -0.695] 4.35 0.32
## Kim2007 metf rosi 0.088 [-0.045; 0.220] 0.14 0.08
## Johnston1998b migl plac -0.944 [-1.193; -0.695] 0.87 0.20
## Gonzalez-Ortiz2004 metf plac -1.114 [-1.231; -0.997] 2.69 0.02
## Stucci1996 benf plac -0.905 [-1.154; -0.656] 3.79 0.13
## Moulin2006 benf plac -0.905 [-1.154; -0.656] 0.59 0.87
## Willms1999 metf plac -1.114 [-1.231; -0.997] 0.04 0.02
## Willms1999 acar plac -0.827 [-1.040; -0.615] 0.04 0.02
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD', comparison: other treatments vs 'plac'):
## MD 95%-CI
## acar -0.827 [-1.040; -0.615]
## benf -0.905 [-1.154; -0.656]
## metf -1.114 [-1.231; -0.997]
## migl -0.944 [-1.193; -0.695]
## piog -1.066 [-1.215; -0.918]
## plac . .
## rosi -1.202 [-1.295; -1.108]
## sita -0.570 [-0.823; -0.317]
## sulf -0.439 [-0.619; -0.260]
## vild -0.700 [-0.950; -0.450]
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
summary(net1)
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## acar benf metf migl piog plac rosi sita
## acar . 0.0778 0.2867 0.1166 0.2391 -0.8274 0.3745 -0.2574
## benf -0.0778 . 0.2089 0.0387 0.1612 -0.9052 0.2967 -0.3352
## metf -0.2867 -0.2089 . -0.1702 -0.0477 -1.1141 0.0877 -0.5441
## migl -0.1166 -0.0387 0.1702 . 0.1225 -0.9439 0.2579 -0.3739
## piog -0.2391 -0.1612 0.0477 -0.1225 . -1.0664 0.1354 -0.4964
## plac 0.8274 0.9052 1.1141 0.9439 1.0664 . 1.2018 0.5700
## rosi -0.3745 -0.2967 -0.0877 -0.2579 -0.1354 -1.2018 . -0.6318
## sita 0.2574 0.3352 0.5441 0.3739 0.4964 -0.5700 0.6318 .
## sulf 0.3879 0.4657 0.6746 0.5044 0.6269 -0.4395 0.7623 0.1305
## vild 0.1274 0.2052 0.4141 0.2439 0.3664 -0.7000 0.5018 -0.1300
## sulf vild
## acar -0.3879 -0.1274
## benf -0.4657 -0.2052
## metf -0.6746 -0.4141
## migl -0.5044 -0.2439
## piog -0.6269 -0.3664
## plac 0.4395 0.7000
## rosi -0.7623 -0.5018
## sita -0.1305 0.1300
## sulf . 0.2605
## vild -0.2605 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.2497 0.0622 -0.2107 -0.0094 -1.0401 0.1506 -0.5879
## benf -0.4054 . -0.0662 -0.3133 -0.1288 -1.1543 0.0306 -0.6903
## metf -0.5113 -0.4841 . -0.4450 -0.1845 -1.2309 -0.0449 -0.8228
## migl -0.4438 -0.3908 -0.1046 . -0.1673 -1.1927 -0.0078 -0.7287
## piog -0.4876 -0.4513 -0.0891 -0.4123 . -1.2151 -0.0249 -0.7899
## plac 0.6147 0.6561 0.9973 0.6952 0.9178 . 1.1084 0.3170
## rosi -0.5983 -0.5627 -0.2203 -0.5236 -0.2957 -1.2953 . -0.9016
## sita -0.0732 -0.0199 0.2654 0.0191 0.2030 -0.8230 0.3621 .
## sulf 0.1662 0.1588 0.5011 0.1978 0.4178 -0.6188 0.5820 -0.1796
## vild -0.2005 -0.1474 0.1386 -0.1084 0.0760 -0.9495 0.2354 -0.4854
## sulf vild
## acar -0.6095 -0.4552
## benf -0.7726 -0.5577
## metf -0.8482 -0.6896
## migl -0.8111 -0.5962
## piog -0.8361 -0.6569
## plac 0.2602 0.4505
## rosi -0.9427 -0.7683
## sita -0.4406 -0.2254
## sulf . -0.0467
## vild -0.5677 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.4054 0.5113 0.4438 0.4876 -0.6147 0.5983 0.0732 -0.1662
## benf 0.2497 . 0.4841 0.3908 0.4513 -0.6561 0.5627 0.0199 -0.1588
## metf -0.0622 0.0662 . 0.1046 0.0891 -0.9973 0.2203 -0.2654 -0.5011
## migl 0.2107 0.3133 0.4450 . 0.4123 -0.6952 0.5236 -0.0191 -0.1978
## piog 0.0094 0.1288 0.1845 0.1673 . -0.9178 0.2957 -0.2030 -0.4178
## plac 1.0401 1.1543 1.2309 1.1927 1.2151 . 1.2953 0.8230 0.6188
## rosi -0.1506 -0.0306 0.0449 0.0078 0.0249 -1.1084 . -0.3621 -0.5820
## sita 0.5879 0.6903 0.8228 0.7287 0.7899 -0.3170 0.9016 . 0.1796
## sulf 0.6095 0.7726 0.8482 0.8111 0.8361 -0.2602 0.9427 0.4406 .
## vild 0.4552 0.5577 0.6896 0.5962 0.6569 -0.4505 0.7683 0.2254 0.0467
## vild
## acar 0.2005
## benf 0.1474
## metf -0.1386
## migl 0.1084
## piog -0.0760
## plac 0.9495
## rosi -0.2354
## sita 0.4854
## sulf 0.5677
## vild .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
# Random effects model
net2 <- netmeta(TE, seTE, treat1, treat2, studlab,
data=Senn2013, sm="MD", comb.random=TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
print(net2, ref="plac", digits=3)
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.141 0.141 2
## Lewin2007 metf plac -0.82 0.099 0.099 2
## Willms1999 acar metf 0.20 0.358 0.388 3 *
## Davidson2007 plac rosi 1.34 0.144 0.144 2
## Wolffenbuttel1999 plac rosi 1.10 0.114 0.114 2
## Kipnes2001 piog plac -1.30 0.127 0.127 2
## Kerenyi2004 plac rosi 0.77 0.108 0.108 2
## Hanefeld2004 metf piog -0.16 0.085 0.085 2
## Derosa2004 piog rosi 0.10 0.183 0.183 2
## Baksi2004 plac rosi 1.30 0.101 0.101 2
## Rosenstock2008 plac rosi 1.09 0.226 0.226 2
## Zhu2003 plac rosi 1.50 0.162 0.162 2
## Yang2003 metf rosi 0.14 0.224 0.224 2
## Vongthavaravat2002 rosi sulf -1.20 0.144 0.144 2
## Oyama2008 acar sulf -0.40 0.155 0.155 2
## Costa1997 acar plac -0.80 0.143 0.143 2
## Hermansen2007 plac sita 0.57 0.129 0.129 2
## Garber2008 plac vild 0.70 0.127 0.127 2
## Alex1998 metf sulf -0.37 0.118 0.118 2
## Johnston1994 migl plac -0.74 0.184 0.184 2
## Johnston1998a migl plac -1.41 0.224 0.224 2
## Kim2007 metf rosi 0.00 0.234 0.234 2
## Johnston1998b migl plac -0.68 0.283 0.283 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.436 0.436 2
## Stucci1996 benf plac -0.23 0.347 0.347 2
## Moulin2006 benf plac -1.01 0.137 0.137 2
## Willms1999 metf plac -1.20 0.376 0.413 3 *
## Willms1999 acar plac -1.00 0.467 0.824 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 MD 95%-CI Q leverage
## DeFronzo1995 metf plac -1.114 [-1.231; -0.997] 30.89 0.18
## Lewin2007 metf plac -1.114 [-1.231; -0.997] 8.79 0.36
## Willms1999 acar metf 0.287 [ 0.062; 0.511] 0.05 0.09
## Davidson2007 plac rosi 1.202 [ 1.108; 1.295] 0.93 0.11
## Wolffenbuttel1999 plac rosi 1.202 [ 1.108; 1.295] 0.80 0.17
## Kipnes2001 piog plac -1.066 [-1.215; -0.918] 3.39 0.36
## Kerenyi2004 plac rosi 1.202 [ 1.108; 1.295] 16.05 0.20
## Hanefeld2004 metf piog -0.048 [-0.184; 0.089] 1.75 0.68
## Derosa2004 piog rosi 0.135 [-0.025; 0.296] 0.04 0.20
## Baksi2004 plac rosi 1.202 [ 1.108; 1.295] 0.94 0.22
## Rosenstock2008 plac rosi 1.202 [ 1.108; 1.295] 0.24 0.04
## Zhu2003 plac rosi 1.202 [ 1.108; 1.295] 3.37 0.09
## Yang2003 metf rosi 0.088 [-0.045; 0.220] 0.05 0.09
## Vongthavaravat2002 rosi sulf -0.762 [-0.943; -0.582] 9.29 0.41
## Oyama2008 acar sulf -0.388 [-0.610; -0.166] 0.01 0.53
## Costa1997 acar plac -0.827 [-1.040; -0.615] 0.04 0.57
## Hermansen2007 plac sita 0.570 [ 0.317; 0.823] 0.00 1.00
## Garber2008 plac vild 0.700 [ 0.450; 0.950] 0.00 1.00
## Alex1998 metf sulf -0.675 [-0.848; -0.501] 6.62 0.56
## Johnston1994 migl plac -0.944 [-1.193; -0.695] 1.23 0.48
## Johnston1998a migl plac -0.944 [-1.193; -0.695] 4.35 0.32
## Kim2007 metf rosi 0.088 [-0.045; 0.220] 0.14 0.08
## Johnston1998b migl plac -0.944 [-1.193; -0.695] 0.87 0.20
## Gonzalez-Ortiz2004 metf plac -1.114 [-1.231; -0.997] 2.69 0.02
## Stucci1996 benf plac -0.905 [-1.154; -0.656] 3.79 0.13
## Moulin2006 benf plac -0.905 [-1.154; -0.656] 0.59 0.87
## Willms1999 metf plac -1.114 [-1.231; -0.997] 0.04 0.02
## Willms1999 acar plac -0.827 [-1.040; -0.615] 0.04 0.02
##
## Results (random effects model):
##
## treat1 treat2 MD 95%-CI
## DeFronzo1995 metf plac -1.127 [-1.429; -0.824]
## Lewin2007 metf plac -1.127 [-1.429; -0.824]
## Willms1999 acar metf 0.285 [-0.221; 0.791]
## Davidson2007 plac rosi 1.233 [ 0.983; 1.484]
## Wolffenbuttel1999 plac rosi 1.233 [ 0.983; 1.484]
## Kipnes2001 piog plac -1.129 [-1.560; -0.699]
## Kerenyi2004 plac rosi 1.233 [ 0.983; 1.484]
## Hanefeld2004 metf piog 0.002 [-0.440; 0.444]
## Derosa2004 piog rosi 0.104 [-0.335; 0.543]
## Baksi2004 plac rosi 1.233 [ 0.983; 1.484]
## Rosenstock2008 plac rosi 1.233 [ 0.983; 1.484]
## Zhu2003 plac rosi 1.233 [ 0.983; 1.484]
## Yang2003 metf rosi 0.107 [-0.217; 0.430]
## Vongthavaravat2002 rosi sulf -0.817 [-1.282; -0.352]
## Oyama2008 acar sulf -0.425 [-0.946; 0.095]
## Costa1997 acar plac -0.842 [-1.324; -0.360]
## Hermansen2007 plac sita 0.570 [-0.124; 1.264]
## Garber2008 plac vild 0.700 [ 0.007; 1.393]
## Alex1998 metf sulf -0.710 [-1.171; -0.249]
## Johnston1994 migl plac -0.950 [-1.404; -0.495]
## Johnston1998a migl plac -0.950 [-1.404; -0.495]
## Kim2007 metf rosi 0.107 [-0.217; 0.430]
## Johnston1998b migl plac -0.950 [-1.404; -0.495]
## Gonzalez-Ortiz2004 metf plac -1.127 [-1.429; -0.824]
## Stucci1996 benf plac -0.731 [-1.292; -0.170]
## Moulin2006 benf plac -0.731 [-1.292; -0.170]
## Willms1999 metf plac -1.127 [-1.429; -0.824]
## Willms1999 acar plac -0.842 [-1.324; -0.360]
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD', comparison: other treatments vs 'plac'):
## MD 95%-CI
## acar -0.827 [-1.040; -0.615]
## benf -0.905 [-1.154; -0.656]
## metf -1.114 [-1.231; -0.997]
## migl -0.944 [-1.193; -0.695]
## piog -1.066 [-1.215; -0.918]
## plac . .
## rosi -1.202 [-1.295; -1.108]
## sita -0.570 [-0.823; -0.317]
## sulf -0.439 [-0.619; -0.260]
## vild -0.700 [-0.950; -0.450]
##
## Random effects model
##
## Treatment estimate (sm = 'MD', comparison: other treatments vs 'plac'):
## MD 95%-CI
## acar -0.842 [-1.324; -0.360]
## benf -0.731 [-1.292; -0.170]
## metf -1.127 [-1.429; -0.824]
## migl -0.950 [-1.404; -0.495]
## piog -1.129 [-1.560; -0.699]
## plac . .
## rosi -1.233 [-1.484; -0.983]
## sita -0.570 [-1.264; 0.124]
## sulf -0.417 [-0.889; 0.056]
## vild -0.700 [-1.393; -0.007]
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
summary(net2)
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = 'MD'):
## acar benf metf migl piog plac rosi sita
## acar . 0.0778 0.2867 0.1166 0.2391 -0.8274 0.3745 -0.2574
## benf -0.0778 . 0.2089 0.0387 0.1612 -0.9052 0.2967 -0.3352
## metf -0.2867 -0.2089 . -0.1702 -0.0477 -1.1141 0.0877 -0.5441
## migl -0.1166 -0.0387 0.1702 . 0.1225 -0.9439 0.2579 -0.3739
## piog -0.2391 -0.1612 0.0477 -0.1225 . -1.0664 0.1354 -0.4964
## plac 0.8274 0.9052 1.1141 0.9439 1.0664 . 1.2018 0.5700
## rosi -0.3745 -0.2967 -0.0877 -0.2579 -0.1354 -1.2018 . -0.6318
## sita 0.2574 0.3352 0.5441 0.3739 0.4964 -0.5700 0.6318 .
## sulf 0.3879 0.4657 0.6746 0.5044 0.6269 -0.4395 0.7623 0.1305
## vild 0.1274 0.2052 0.4141 0.2439 0.3664 -0.7000 0.5018 -0.1300
## sulf vild
## acar -0.3879 -0.1274
## benf -0.4657 -0.2052
## metf -0.6746 -0.4141
## migl -0.5044 -0.2439
## piog -0.6269 -0.3664
## plac 0.4395 0.7000
## rosi -0.7623 -0.5018
## sita -0.1305 0.1300
## sulf . 0.2605
## vild -0.2605 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.2497 0.0622 -0.2107 -0.0094 -1.0401 0.1506 -0.5879
## benf -0.4054 . -0.0662 -0.3133 -0.1288 -1.1543 0.0306 -0.6903
## metf -0.5113 -0.4841 . -0.4450 -0.1845 -1.2309 -0.0449 -0.8228
## migl -0.4438 -0.3908 -0.1046 . -0.1673 -1.1927 -0.0078 -0.7287
## piog -0.4876 -0.4513 -0.0891 -0.4123 . -1.2151 -0.0249 -0.7899
## plac 0.6147 0.6561 0.9973 0.6952 0.9178 . 1.1084 0.3170
## rosi -0.5983 -0.5627 -0.2203 -0.5236 -0.2957 -1.2953 . -0.9016
## sita -0.0732 -0.0199 0.2654 0.0191 0.2030 -0.8230 0.3621 .
## sulf 0.1662 0.1588 0.5011 0.1978 0.4178 -0.6188 0.5820 -0.1796
## vild -0.2005 -0.1474 0.1386 -0.1084 0.0760 -0.9495 0.2354 -0.4854
## sulf vild
## acar -0.6095 -0.4552
## benf -0.7726 -0.5577
## metf -0.8482 -0.6896
## migl -0.8111 -0.5962
## piog -0.8361 -0.6569
## plac 0.2602 0.4505
## rosi -0.9427 -0.7683
## sita -0.4406 -0.2254
## sulf . -0.0467
## vild -0.5677 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.4054 0.5113 0.4438 0.4876 -0.6147 0.5983 0.0732 -0.1662
## benf 0.2497 . 0.4841 0.3908 0.4513 -0.6561 0.5627 0.0199 -0.1588
## metf -0.0622 0.0662 . 0.1046 0.0891 -0.9973 0.2203 -0.2654 -0.5011
## migl 0.2107 0.3133 0.4450 . 0.4123 -0.6952 0.5236 -0.0191 -0.1978
## piog 0.0094 0.1288 0.1845 0.1673 . -0.9178 0.2957 -0.2030 -0.4178
## plac 1.0401 1.1543 1.2309 1.1927 1.2151 . 1.2953 0.8230 0.6188
## rosi -0.1506 -0.0306 0.0449 0.0078 0.0249 -1.1084 . -0.3621 -0.5820
## sita 0.5879 0.6903 0.8228 0.7287 0.7899 -0.3170 0.9016 . 0.1796
## sulf 0.6095 0.7726 0.8482 0.8111 0.8361 -0.2602 0.9427 0.4406 .
## vild 0.4552 0.5577 0.6896 0.5962 0.6569 -0.4505 0.7683 0.2254 0.0467
## vild
## acar 0.2005
## benf 0.1474
## metf -0.1386
## migl 0.1084
## piog -0.0760
## plac 0.9495
## rosi -0.2354
## sita 0.4854
## sulf 0.5677
## vild .
##
## Random effects model
##
## Treatment estimate (sm = 'MD'):
## acar benf metf migl piog plac rosi sita
## acar . -0.1106 0.2850 0.1079 0.2873 -0.8418 0.3917 -0.2718
## benf 0.1106 . 0.3956 0.2186 0.3979 -0.7311 0.5023 -0.1611
## metf -0.2850 -0.3956 . -0.1770 0.0023 -1.1268 0.1067 -0.5568
## migl -0.1079 -0.2186 0.1770 . 0.1794 -0.9497 0.2837 -0.3797
## piog -0.2873 -0.3979 -0.0023 -0.1794 . -1.1291 0.1044 -0.5591
## plac 0.8418 0.7311 1.1268 0.9497 1.1291 . 1.2335 0.5700
## rosi -0.3917 -0.5023 -0.1067 -0.2837 -0.1044 -1.2335 . -0.6635
## sita 0.2718 0.1611 0.5568 0.3797 0.5591 -0.5700 0.6635 .
## sulf 0.4252 0.3146 0.7102 0.5332 0.7125 -0.4166 0.8169 0.1534
## vild 0.1418 0.0311 0.4268 0.2497 0.4291 -0.7000 0.5335 -0.1300
## sulf vild
## acar -0.4252 -0.1418
## benf -0.3146 -0.0311
## metf -0.7102 -0.4268
## migl -0.5332 -0.2497
## piog -0.7125 -0.4291
## plac 0.4166 0.7000
## rosi -0.8169 -0.5335
## sita -0.1534 0.1300
## sulf . 0.2834
## vild -0.2834 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.8499 -0.2208 -0.5542 -0.3313 -1.3236 -0.1189 -1.1166
## benf -0.6286 . -0.2414 -0.5030 -0.3089 -1.2918 -0.1118 -1.0534
## metf -0.7908 -1.0327 . -0.7227 -0.4398 -1.4291 -0.2170 -1.3138
## migl -0.7701 -0.9402 -0.3686 . -0.4465 -1.4040 -0.2350 -1.2092
## piog -0.9059 -1.1048 -0.4444 -0.8052 . -1.5596 -0.3347 -1.3758
## plac 0.3600 0.1705 0.8244 0.4955 0.6986 . 0.9830 -0.1240
## rosi -0.9023 -1.1164 -0.4304 -0.8025 -0.5435 -1.4839 . -1.4013
## sita -0.5731 -0.7311 -0.2002 -0.4497 -0.2576 -1.2640 -0.0744 .
## sulf -0.0951 -0.4184 0.2491 -0.1220 0.1205 -0.8887 0.3521 -0.6859
## vild -0.7020 -0.8601 -0.3291 -0.5787 -0.3865 -1.3927 -0.2032 -1.1106
## sulf vild
## acar -0.9456 -0.9856
## benf -1.0476 -0.9224
## metf -1.1713 -1.1826
## migl -1.1883 -1.0781
## piog -1.3045 -1.2447
## plac -0.0556 0.0073
## rosi -1.2817 -1.2701
## sita -0.9928 -0.8506
## sulf . -0.5549
## vild -1.1218 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.6286 0.7908 0.7701 0.9059 -0.3600 0.9023 0.5731 0.0951
## benf 0.8499 . 1.0327 0.9402 1.1048 -0.1705 1.1164 0.7311 0.4184
## metf 0.2208 0.2414 . 0.3686 0.4444 -0.8244 0.4304 0.2002 -0.2491
## migl 0.5542 0.5030 0.7227 . 0.8052 -0.4955 0.8025 0.4497 0.1220
## piog 0.3313 0.3089 0.4398 0.4465 . -0.6986 0.5435 0.2576 -0.1205
## plac 1.3236 1.2918 1.4291 1.4040 1.5596 . 1.4839 1.2640 0.8887
## rosi 0.1189 0.1118 0.2170 0.2350 0.3347 -0.9830 . 0.0744 -0.3521
## sita 1.1166 1.0534 1.3138 1.2092 1.3758 0.1240 1.4013 . 0.6859
## sulf 0.9456 1.0476 1.1713 1.1883 1.3045 0.0556 1.2817 0.9928 .
## vild 0.9856 0.9224 1.1826 1.0781 1.2447 -0.0073 1.2701 0.8506 0.5549
## vild
## acar 0.7020
## benf 0.8601
## metf 0.3291
## migl 0.5787
## piog 0.3865
## plac 1.3927
## rosi 0.2032
## sita 1.1106
## sulf 1.1218
## vild .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
data(Senn2013)
# Fixed effect model (default)
net1 <- netmeta(TE, seTE, treat1, treat2, studlab, data=Senn2013)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
net1
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.1414 0.1414 2
## Lewin2007 metf plac -0.82 0.0992 0.0992 2
## Willms1999 acar metf 0.20 0.3579 0.3884 3 *
## Davidson2007 plac rosi 1.34 0.1435 0.1435 2
## Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141 2
## Kipnes2001 piog plac -1.30 0.1268 0.1268 2
## Kerenyi2004 plac rosi 0.77 0.1078 0.1078 2
## Hanefeld2004 metf piog -0.16 0.0849 0.0849 2
## Derosa2004 piog rosi 0.10 0.1831 0.1831 2
## Baksi2004 plac rosi 1.30 0.1014 0.1014 2
## Rosenstock2008 plac rosi 1.09 0.2263 0.2263 2
## Zhu2003 plac rosi 1.50 0.1624 0.1624 2
## Yang2003 metf rosi 0.14 0.2239 0.2239 2
## Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436 2
## Oyama2008 acar sulf -0.40 0.1549 0.1549 2
## Costa1997 acar plac -0.80 0.1432 0.1432 2
## Hermansen2007 plac sita 0.57 0.1291 0.1291 2
## Garber2008 plac vild 0.70 0.1273 0.1273 2
## Alex1998 metf sulf -0.37 0.1184 0.1184 2
## Johnston1994 migl plac -0.74 0.1839 0.1839 2
## Johnston1998a migl plac -1.41 0.2235 0.2235 2
## Kim2007 metf rosi 0.00 0.2339 0.2339 2
## Johnston1998b migl plac -0.68 0.2828 0.2828 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356 2
## Stucci1996 benf plac -0.23 0.3467 0.3467 2
## Moulin2006 benf plac -1.01 0.1366 0.1366 2
## Willms1999 metf plac -1.20 0.3758 0.4125 3 *
## Willms1999 acar plac -1.00 0.4669 0.8242 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 95%-CI Q
## DeFronzo1995 metf plac -1.1141 [-1.2309; -0.9973] 30.89
## Lewin2007 metf plac -1.1141 [-1.2309; -0.9973] 8.79
## Willms1999 acar metf 0.2867 [ 0.0622; 0.5113] 0.05
## Davidson2007 plac rosi 1.2018 [ 1.1084; 1.2953] 0.93
## Wolffenbuttel1999 plac rosi 1.2018 [ 1.1084; 1.2953] 0.80
## Kipnes2001 piog plac -1.0664 [-1.2151; -0.9178] 3.39
## Kerenyi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 16.05
## Hanefeld2004 metf piog -0.0477 [-0.1845; 0.0891] 1.75
## Derosa2004 piog rosi 0.1354 [-0.0249; 0.2957] 0.04
## Baksi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 0.94
## Rosenstock2008 plac rosi 1.2018 [ 1.1084; 1.2953] 0.24
## Zhu2003 plac rosi 1.2018 [ 1.1084; 1.2953] 3.37
## Yang2003 metf rosi 0.0877 [-0.0449; 0.2203] 0.05
## Vongthavaravat2002 rosi sulf -0.7623 [-0.9427; -0.5820] 9.29
## Oyama2008 acar sulf -0.3879 [-0.6095; -0.1662] 0.01
## Costa1997 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## Hermansen2007 plac sita 0.5700 [ 0.3170; 0.8230] 0.00
## Garber2008 plac vild 0.7000 [ 0.4505; 0.9495] 0.00
## Alex1998 metf sulf -0.6746 [-0.8482; -0.5011] 6.62
## Johnston1994 migl plac -0.9439 [-1.1927; -0.6952] 1.23
## Johnston1998a migl plac -0.9439 [-1.1927; -0.6952] 4.35
## Kim2007 metf rosi 0.0877 [-0.0449; 0.2203] 0.14
## Johnston1998b migl plac -0.9439 [-1.1927; -0.6952] 0.87
## Gonzalez-Ortiz2004 metf plac -1.1141 [-1.2309; -0.9973] 2.69
## Stucci1996 benf plac -0.9052 [-1.1543; -0.6561] 3.79
## Moulin2006 benf plac -0.9052 [-1.1543; -0.6561] 0.59
## Willms1999 metf plac -1.1141 [-1.2309; -0.9973] 0.04
## Willms1999 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## leverage
## DeFronzo1995 0.18
## Lewin2007 0.36
## Willms1999 0.09
## Davidson2007 0.11
## Wolffenbuttel1999 0.17
## Kipnes2001 0.36
## Kerenyi2004 0.20
## Hanefeld2004 0.68
## Derosa2004 0.20
## Baksi2004 0.22
## Rosenstock2008 0.04
## Zhu2003 0.09
## Yang2003 0.09
## Vongthavaravat2002 0.41
## Oyama2008 0.53
## Costa1997 0.57
## Hermansen2007 1.00
## Garber2008 1.00
## Alex1998 0.56
## Johnston1994 0.48
## Johnston1998a 0.32
## Kim2007 0.08
## Johnston1998b 0.20
## Gonzalez-Ortiz2004 0.02
## Stucci1996 0.13
## Moulin2006 0.87
## Willms1999 0.02
## Willms1999 0.02
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = ''):
## acar benf metf migl piog plac rosi sita
## acar . 0.0778 0.2867 0.1166 0.2391 -0.8274 0.3745 -0.2574
## benf -0.0778 . 0.2089 0.0387 0.1612 -0.9052 0.2967 -0.3352
## metf -0.2867 -0.2089 . -0.1702 -0.0477 -1.1141 0.0877 -0.5441
## migl -0.1166 -0.0387 0.1702 . 0.1225 -0.9439 0.2579 -0.3739
## piog -0.2391 -0.1612 0.0477 -0.1225 . -1.0664 0.1354 -0.4964
## plac 0.8274 0.9052 1.1141 0.9439 1.0664 . 1.2018 0.5700
## rosi -0.3745 -0.2967 -0.0877 -0.2579 -0.1354 -1.2018 . -0.6318
## sita 0.2574 0.3352 0.5441 0.3739 0.4964 -0.5700 0.6318 .
## sulf 0.3879 0.4657 0.6746 0.5044 0.6269 -0.4395 0.7623 0.1305
## vild 0.1274 0.2052 0.4141 0.2439 0.3664 -0.7000 0.5018 -0.1300
## sulf vild
## acar -0.3879 -0.1274
## benf -0.4657 -0.2052
## metf -0.6746 -0.4141
## migl -0.5044 -0.2439
## piog -0.6269 -0.3664
## plac 0.4395 0.7000
## rosi -0.7623 -0.5018
## sita -0.1305 0.1300
## sulf . 0.2605
## vild -0.2605 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.2497 0.0622 -0.2107 -0.0094 -1.0401 0.1506 -0.5879
## benf -0.4054 . -0.0662 -0.3133 -0.1288 -1.1543 0.0306 -0.6903
## metf -0.5113 -0.4841 . -0.4450 -0.1845 -1.2309 -0.0449 -0.8228
## migl -0.4438 -0.3908 -0.1046 . -0.1673 -1.1927 -0.0078 -0.7287
## piog -0.4876 -0.4513 -0.0891 -0.4123 . -1.2151 -0.0249 -0.7899
## plac 0.6147 0.6561 0.9973 0.6952 0.9178 . 1.1084 0.3170
## rosi -0.5983 -0.5627 -0.2203 -0.5236 -0.2957 -1.2953 . -0.9016
## sita -0.0732 -0.0199 0.2654 0.0191 0.2030 -0.8230 0.3621 .
## sulf 0.1662 0.1588 0.5011 0.1978 0.4178 -0.6188 0.5820 -0.1796
## vild -0.2005 -0.1474 0.1386 -0.1084 0.0760 -0.9495 0.2354 -0.4854
## sulf vild
## acar -0.6095 -0.4552
## benf -0.7726 -0.5577
## metf -0.8482 -0.6896
## migl -0.8111 -0.5962
## piog -0.8361 -0.6569
## plac 0.2602 0.4505
## rosi -0.9427 -0.7683
## sita -0.4406 -0.2254
## sulf . -0.0467
## vild -0.5677 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.4054 0.5113 0.4438 0.4876 -0.6147 0.5983 0.0732 -0.1662
## benf 0.2497 . 0.4841 0.3908 0.4513 -0.6561 0.5627 0.0199 -0.1588
## metf -0.0622 0.0662 . 0.1046 0.0891 -0.9973 0.2203 -0.2654 -0.5011
## migl 0.2107 0.3133 0.4450 . 0.4123 -0.6952 0.5236 -0.0191 -0.1978
## piog 0.0094 0.1288 0.1845 0.1673 . -0.9178 0.2957 -0.2030 -0.4178
## plac 1.0401 1.1543 1.2309 1.1927 1.2151 . 1.2953 0.8230 0.6188
## rosi -0.1506 -0.0306 0.0449 0.0078 0.0249 -1.1084 . -0.3621 -0.5820
## sita 0.5879 0.6903 0.8228 0.7287 0.7899 -0.3170 0.9016 . 0.1796
## sulf 0.6095 0.7726 0.8482 0.8111 0.8361 -0.2602 0.9427 0.4406 .
## vild 0.4552 0.5577 0.6896 0.5962 0.6569 -0.4505 0.7683 0.2254 0.0467
## vild
## acar 0.2005
## benf 0.1474
## metf -0.1386
## migl 0.1084
## piog -0.0760
## plac 0.9495
## rosi -0.2354
## sita 0.4854
## sulf 0.5677
## vild .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
net1$Q.decomp
## treat1 treat2 Q df pval.Q
## 1 acar metf 0.00000000 0 1.000000e+00
## 2 acar plac 0.05715942 1 8.110433e-01
## 3 acar sulf 0.00000000 0 1.000000e+00
## 4 benf plac 4.38137713 1 3.633363e-02
## 5 metf piog 0.00000000 0 1.000000e+00
## 6 metf plac 42.17838677 3 3.677205e-09
## 7 metf rosi 0.18695080 1 6.654667e-01
## 8 metf sulf 0.00000000 0 1.000000e+00
## 9 migl plac 6.44927778 2 3.977014e-02
## 10 piog plac 0.00000000 0 1.000000e+00
## 11 piog rosi 0.00000000 0 1.000000e+00
## 12 plac rosi 21.27336195 5 7.191616e-04
## 13 plac sita 0.00000000 0 1.000000e+00
## 14 plac vild 0.00000000 0 1.000000e+00
## 15 rosi sulf 0.00000000 0 1.000000e+00
# Forest plot
forest(net1, ref="plac")
# Comparison with reference group
netmeta(TE, seTE, treat1, treat2, studlab, data=Senn2013,
reference="plac")
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.1414 0.1414 2
## Lewin2007 metf plac -0.82 0.0992 0.0992 2
## Willms1999 acar metf 0.20 0.3579 0.3884 3 *
## Davidson2007 plac rosi 1.34 0.1435 0.1435 2
## Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141 2
## Kipnes2001 piog plac -1.30 0.1268 0.1268 2
## Kerenyi2004 plac rosi 0.77 0.1078 0.1078 2
## Hanefeld2004 metf piog -0.16 0.0849 0.0849 2
## Derosa2004 piog rosi 0.10 0.1831 0.1831 2
## Baksi2004 plac rosi 1.30 0.1014 0.1014 2
## Rosenstock2008 plac rosi 1.09 0.2263 0.2263 2
## Zhu2003 plac rosi 1.50 0.1624 0.1624 2
## Yang2003 metf rosi 0.14 0.2239 0.2239 2
## Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436 2
## Oyama2008 acar sulf -0.40 0.1549 0.1549 2
## Costa1997 acar plac -0.80 0.1432 0.1432 2
## Hermansen2007 plac sita 0.57 0.1291 0.1291 2
## Garber2008 plac vild 0.70 0.1273 0.1273 2
## Alex1998 metf sulf -0.37 0.1184 0.1184 2
## Johnston1994 migl plac -0.74 0.1839 0.1839 2
## Johnston1998a migl plac -1.41 0.2235 0.2235 2
## Kim2007 metf rosi 0.00 0.2339 0.2339 2
## Johnston1998b migl plac -0.68 0.2828 0.2828 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356 2
## Stucci1996 benf plac -0.23 0.3467 0.3467 2
## Moulin2006 benf plac -1.01 0.1366 0.1366 2
## Willms1999 metf plac -1.20 0.3758 0.4125 3 *
## Willms1999 acar plac -1.00 0.4669 0.8242 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 95%-CI Q
## DeFronzo1995 metf plac -1.1141 [-1.2309; -0.9973] 30.89
## Lewin2007 metf plac -1.1141 [-1.2309; -0.9973] 8.79
## Willms1999 acar metf 0.2867 [ 0.0622; 0.5113] 0.05
## Davidson2007 plac rosi 1.2018 [ 1.1084; 1.2953] 0.93
## Wolffenbuttel1999 plac rosi 1.2018 [ 1.1084; 1.2953] 0.80
## Kipnes2001 piog plac -1.0664 [-1.2151; -0.9178] 3.39
## Kerenyi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 16.05
## Hanefeld2004 metf piog -0.0477 [-0.1845; 0.0891] 1.75
## Derosa2004 piog rosi 0.1354 [-0.0249; 0.2957] 0.04
## Baksi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 0.94
## Rosenstock2008 plac rosi 1.2018 [ 1.1084; 1.2953] 0.24
## Zhu2003 plac rosi 1.2018 [ 1.1084; 1.2953] 3.37
## Yang2003 metf rosi 0.0877 [-0.0449; 0.2203] 0.05
## Vongthavaravat2002 rosi sulf -0.7623 [-0.9427; -0.5820] 9.29
## Oyama2008 acar sulf -0.3879 [-0.6095; -0.1662] 0.01
## Costa1997 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## Hermansen2007 plac sita 0.5700 [ 0.3170; 0.8230] 0.00
## Garber2008 plac vild 0.7000 [ 0.4505; 0.9495] 0.00
## Alex1998 metf sulf -0.6746 [-0.8482; -0.5011] 6.62
## Johnston1994 migl plac -0.9439 [-1.1927; -0.6952] 1.23
## Johnston1998a migl plac -0.9439 [-1.1927; -0.6952] 4.35
## Kim2007 metf rosi 0.0877 [-0.0449; 0.2203] 0.14
## Johnston1998b migl plac -0.9439 [-1.1927; -0.6952] 0.87
## Gonzalez-Ortiz2004 metf plac -1.1141 [-1.2309; -0.9973] 2.69
## Stucci1996 benf plac -0.9052 [-1.1543; -0.6561] 3.79
## Moulin2006 benf plac -0.9052 [-1.1543; -0.6561] 0.59
## Willms1999 metf plac -1.1141 [-1.2309; -0.9973] 0.04
## Willms1999 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## leverage
## DeFronzo1995 0.18
## Lewin2007 0.36
## Willms1999 0.09
## Davidson2007 0.11
## Wolffenbuttel1999 0.17
## Kipnes2001 0.36
## Kerenyi2004 0.20
## Hanefeld2004 0.68
## Derosa2004 0.20
## Baksi2004 0.22
## Rosenstock2008 0.04
## Zhu2003 0.09
## Yang2003 0.09
## Vongthavaravat2002 0.41
## Oyama2008 0.53
## Costa1997 0.57
## Hermansen2007 1.00
## Garber2008 1.00
## Alex1998 0.56
## Johnston1994 0.48
## Johnston1998a 0.32
## Kim2007 0.08
## Johnston1998b 0.20
## Gonzalez-Ortiz2004 0.02
## Stucci1996 0.13
## Moulin2006 0.87
## Willms1999 0.02
## Willms1999 0.02
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = '', comparison: other treatments vs 'plac'):
## 95%-CI
## acar -0.8274 [-1.0401; -0.6147]
## benf -0.9052 [-1.1543; -0.6561]
## metf -1.1141 [-1.2309; -0.9973]
## migl -0.9439 [-1.1927; -0.6952]
## piog -1.0664 [-1.2151; -0.9178]
## plac . .
## rosi -1.2018 [-1.2953; -1.1084]
## sita -0.5700 [-0.8230; -0.3170]
## sulf -0.4395 [-0.6188; -0.2602]
## vild -0.7000 [-0.9495; -0.4505]
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
# Random effects model
net2 <- netmeta(TE, seTE, treat1, treat2, studlab, data=Senn2013,
comb.random = TRUE)
## Warning: Note, treatments within a comparison have been re-sorted in
## increasing order.
net2
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## DeFronzo1995 metf plac -1.90 0.1414 0.1414 2
## Lewin2007 metf plac -0.82 0.0992 0.0992 2
## Willms1999 acar metf 0.20 0.3579 0.3884 3 *
## Davidson2007 plac rosi 1.34 0.1435 0.1435 2
## Wolffenbuttel1999 plac rosi 1.10 0.1141 0.1141 2
## Kipnes2001 piog plac -1.30 0.1268 0.1268 2
## Kerenyi2004 plac rosi 0.77 0.1078 0.1078 2
## Hanefeld2004 metf piog -0.16 0.0849 0.0849 2
## Derosa2004 piog rosi 0.10 0.1831 0.1831 2
## Baksi2004 plac rosi 1.30 0.1014 0.1014 2
## Rosenstock2008 plac rosi 1.09 0.2263 0.2263 2
## Zhu2003 plac rosi 1.50 0.1624 0.1624 2
## Yang2003 metf rosi 0.14 0.2239 0.2239 2
## Vongthavaravat2002 rosi sulf -1.20 0.1436 0.1436 2
## Oyama2008 acar sulf -0.40 0.1549 0.1549 2
## Costa1997 acar plac -0.80 0.1432 0.1432 2
## Hermansen2007 plac sita 0.57 0.1291 0.1291 2
## Garber2008 plac vild 0.70 0.1273 0.1273 2
## Alex1998 metf sulf -0.37 0.1184 0.1184 2
## Johnston1994 migl plac -0.74 0.1839 0.1839 2
## Johnston1998a migl plac -1.41 0.2235 0.2235 2
## Kim2007 metf rosi 0.00 0.2339 0.2339 2
## Johnston1998b migl plac -0.68 0.2828 0.2828 2
## Gonzalez-Ortiz2004 metf plac -0.40 0.4356 0.4356 2
## Stucci1996 benf plac -0.23 0.3467 0.3467 2
## Moulin2006 benf plac -1.01 0.1366 0.1366 2
## Willms1999 metf plac -1.20 0.3758 0.4125 3 *
## Willms1999 acar plac -1.00 0.4669 0.8242 3 *
##
## Number of treatment arms (by study):
## narms
## Alex1998 2
## Baksi2004 2
## Costa1997 2
## Davidson2007 2
## DeFronzo1995 2
## Derosa2004 2
## Garber2008 2
## Gonzalez-Ortiz2004 2
## Hanefeld2004 2
## Hermansen2007 2
## Johnston1994 2
## Johnston1998a 2
## Johnston1998b 2
## Kerenyi2004 2
## Kim2007 2
## Kipnes2001 2
## Lewin2007 2
## Moulin2006 2
## Oyama2008 2
## Rosenstock2008 2
## Stucci1996 2
## Vongthavaravat2002 2
## Willms1999 3
## Wolffenbuttel1999 2
## Yang2003 2
## Zhu2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 95%-CI Q
## DeFronzo1995 metf plac -1.1141 [-1.2309; -0.9973] 30.89
## Lewin2007 metf plac -1.1141 [-1.2309; -0.9973] 8.79
## Willms1999 acar metf 0.2867 [ 0.0622; 0.5113] 0.05
## Davidson2007 plac rosi 1.2018 [ 1.1084; 1.2953] 0.93
## Wolffenbuttel1999 plac rosi 1.2018 [ 1.1084; 1.2953] 0.80
## Kipnes2001 piog plac -1.0664 [-1.2151; -0.9178] 3.39
## Kerenyi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 16.05
## Hanefeld2004 metf piog -0.0477 [-0.1845; 0.0891] 1.75
## Derosa2004 piog rosi 0.1354 [-0.0249; 0.2957] 0.04
## Baksi2004 plac rosi 1.2018 [ 1.1084; 1.2953] 0.94
## Rosenstock2008 plac rosi 1.2018 [ 1.1084; 1.2953] 0.24
## Zhu2003 plac rosi 1.2018 [ 1.1084; 1.2953] 3.37
## Yang2003 metf rosi 0.0877 [-0.0449; 0.2203] 0.05
## Vongthavaravat2002 rosi sulf -0.7623 [-0.9427; -0.5820] 9.29
## Oyama2008 acar sulf -0.3879 [-0.6095; -0.1662] 0.01
## Costa1997 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## Hermansen2007 plac sita 0.5700 [ 0.3170; 0.8230] 0.00
## Garber2008 plac vild 0.7000 [ 0.4505; 0.9495] 0.00
## Alex1998 metf sulf -0.6746 [-0.8482; -0.5011] 6.62
## Johnston1994 migl plac -0.9439 [-1.1927; -0.6952] 1.23
## Johnston1998a migl plac -0.9439 [-1.1927; -0.6952] 4.35
## Kim2007 metf rosi 0.0877 [-0.0449; 0.2203] 0.14
## Johnston1998b migl plac -0.9439 [-1.1927; -0.6952] 0.87
## Gonzalez-Ortiz2004 metf plac -1.1141 [-1.2309; -0.9973] 2.69
## Stucci1996 benf plac -0.9052 [-1.1543; -0.6561] 3.79
## Moulin2006 benf plac -0.9052 [-1.1543; -0.6561] 0.59
## Willms1999 metf plac -1.1141 [-1.2309; -0.9973] 0.04
## Willms1999 acar plac -0.8274 [-1.0401; -0.6147] 0.04
## leverage
## DeFronzo1995 0.18
## Lewin2007 0.36
## Willms1999 0.09
## Davidson2007 0.11
## Wolffenbuttel1999 0.17
## Kipnes2001 0.36
## Kerenyi2004 0.20
## Hanefeld2004 0.68
## Derosa2004 0.20
## Baksi2004 0.22
## Rosenstock2008 0.04
## Zhu2003 0.09
## Yang2003 0.09
## Vongthavaravat2002 0.41
## Oyama2008 0.53
## Costa1997 0.57
## Hermansen2007 1.00
## Garber2008 1.00
## Alex1998 0.56
## Johnston1994 0.48
## Johnston1998a 0.32
## Kim2007 0.08
## Johnston1998b 0.20
## Gonzalez-Ortiz2004 0.02
## Stucci1996 0.13
## Moulin2006 0.87
## Willms1999 0.02
## Willms1999 0.02
##
## Results (random effects model):
##
## treat1 treat2 95%-CI
## DeFronzo1995 metf plac -1.1268 [-1.4291; -0.8244]
## Lewin2007 metf plac -1.1268 [-1.4291; -0.8244]
## Willms1999 acar metf 0.2850 [-0.2208; 0.7908]
## Davidson2007 plac rosi 1.2335 [ 0.9830; 1.4839]
## Wolffenbuttel1999 plac rosi 1.2335 [ 0.9830; 1.4839]
## Kipnes2001 piog plac -1.1291 [-1.5596; -0.6986]
## Kerenyi2004 plac rosi 1.2335 [ 0.9830; 1.4839]
## Hanefeld2004 metf piog 0.0023 [-0.4398; 0.4444]
## Derosa2004 piog rosi 0.1044 [-0.3347; 0.5435]
## Baksi2004 plac rosi 1.2335 [ 0.9830; 1.4839]
## Rosenstock2008 plac rosi 1.2335 [ 0.9830; 1.4839]
## Zhu2003 plac rosi 1.2335 [ 0.9830; 1.4839]
## Yang2003 metf rosi 0.1067 [-0.2170; 0.4304]
## Vongthavaravat2002 rosi sulf -0.8169 [-1.2817; -0.3521]
## Oyama2008 acar sulf -0.4252 [-0.9456; 0.0951]
## Costa1997 acar plac -0.8418 [-1.3236; -0.3600]
## Hermansen2007 plac sita 0.5700 [-0.1240; 1.2640]
## Garber2008 plac vild 0.7000 [ 0.0073; 1.3927]
## Alex1998 metf sulf -0.7102 [-1.1713; -0.2491]
## Johnston1994 migl plac -0.9497 [-1.4040; -0.4955]
## Johnston1998a migl plac -0.9497 [-1.4040; -0.4955]
## Kim2007 metf rosi 0.1067 [-0.2170; 0.4304]
## Johnston1998b migl plac -0.9497 [-1.4040; -0.4955]
## Gonzalez-Ortiz2004 metf plac -1.1268 [-1.4291; -0.8244]
## Stucci1996 benf plac -0.7311 [-1.2918; -0.1705]
## Moulin2006 benf plac -0.7311 [-1.2918; -0.1705]
## Willms1999 metf plac -1.1268 [-1.4291; -0.8244]
## Willms1999 acar plac -0.8418 [-1.3236; -0.3600]
##
## Number of studies: k = 26
## Number of treatments: n = 10
## Number of pairwise comparisons: m = 28
## Number of designs: d = 15
##
## Fixed effect model
##
## Treatment estimate (sm = ''):
## acar benf metf migl piog plac rosi sita
## acar . 0.0778 0.2867 0.1166 0.2391 -0.8274 0.3745 -0.2574
## benf -0.0778 . 0.2089 0.0387 0.1612 -0.9052 0.2967 -0.3352
## metf -0.2867 -0.2089 . -0.1702 -0.0477 -1.1141 0.0877 -0.5441
## migl -0.1166 -0.0387 0.1702 . 0.1225 -0.9439 0.2579 -0.3739
## piog -0.2391 -0.1612 0.0477 -0.1225 . -1.0664 0.1354 -0.4964
## plac 0.8274 0.9052 1.1141 0.9439 1.0664 . 1.2018 0.5700
## rosi -0.3745 -0.2967 -0.0877 -0.2579 -0.1354 -1.2018 . -0.6318
## sita 0.2574 0.3352 0.5441 0.3739 0.4964 -0.5700 0.6318 .
## sulf 0.3879 0.4657 0.6746 0.5044 0.6269 -0.4395 0.7623 0.1305
## vild 0.1274 0.2052 0.4141 0.2439 0.3664 -0.7000 0.5018 -0.1300
## sulf vild
## acar -0.3879 -0.1274
## benf -0.4657 -0.2052
## metf -0.6746 -0.4141
## migl -0.5044 -0.2439
## piog -0.6269 -0.3664
## plac 0.4395 0.7000
## rosi -0.7623 -0.5018
## sita -0.1305 0.1300
## sulf . 0.2605
## vild -0.2605 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.2497 0.0622 -0.2107 -0.0094 -1.0401 0.1506 -0.5879
## benf -0.4054 . -0.0662 -0.3133 -0.1288 -1.1543 0.0306 -0.6903
## metf -0.5113 -0.4841 . -0.4450 -0.1845 -1.2309 -0.0449 -0.8228
## migl -0.4438 -0.3908 -0.1046 . -0.1673 -1.1927 -0.0078 -0.7287
## piog -0.4876 -0.4513 -0.0891 -0.4123 . -1.2151 -0.0249 -0.7899
## plac 0.6147 0.6561 0.9973 0.6952 0.9178 . 1.1084 0.3170
## rosi -0.5983 -0.5627 -0.2203 -0.5236 -0.2957 -1.2953 . -0.9016
## sita -0.0732 -0.0199 0.2654 0.0191 0.2030 -0.8230 0.3621 .
## sulf 0.1662 0.1588 0.5011 0.1978 0.4178 -0.6188 0.5820 -0.1796
## vild -0.2005 -0.1474 0.1386 -0.1084 0.0760 -0.9495 0.2354 -0.4854
## sulf vild
## acar -0.6095 -0.4552
## benf -0.7726 -0.5577
## metf -0.8482 -0.6896
## migl -0.8111 -0.5962
## piog -0.8361 -0.6569
## plac 0.2602 0.4505
## rosi -0.9427 -0.7683
## sita -0.4406 -0.2254
## sulf . -0.0467
## vild -0.5677 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.4054 0.5113 0.4438 0.4876 -0.6147 0.5983 0.0732 -0.1662
## benf 0.2497 . 0.4841 0.3908 0.4513 -0.6561 0.5627 0.0199 -0.1588
## metf -0.0622 0.0662 . 0.1046 0.0891 -0.9973 0.2203 -0.2654 -0.5011
## migl 0.2107 0.3133 0.4450 . 0.4123 -0.6952 0.5236 -0.0191 -0.1978
## piog 0.0094 0.1288 0.1845 0.1673 . -0.9178 0.2957 -0.2030 -0.4178
## plac 1.0401 1.1543 1.2309 1.1927 1.2151 . 1.2953 0.8230 0.6188
## rosi -0.1506 -0.0306 0.0449 0.0078 0.0249 -1.1084 . -0.3621 -0.5820
## sita 0.5879 0.6903 0.8228 0.7287 0.7899 -0.3170 0.9016 . 0.1796
## sulf 0.6095 0.7726 0.8482 0.8111 0.8361 -0.2602 0.9427 0.4406 .
## vild 0.4552 0.5577 0.6896 0.5962 0.6569 -0.4505 0.7683 0.2254 0.0467
## vild
## acar 0.2005
## benf 0.1474
## metf -0.1386
## migl 0.1084
## piog -0.0760
## plac 0.9495
## rosi -0.2354
## sita 0.4854
## sulf 0.5677
## vild .
##
## Random effects model
##
## Treatment estimate (sm = ''):
## acar benf metf migl piog plac rosi sita
## acar . -0.1106 0.2850 0.1079 0.2873 -0.8418 0.3917 -0.2718
## benf 0.1106 . 0.3956 0.2186 0.3979 -0.7311 0.5023 -0.1611
## metf -0.2850 -0.3956 . -0.1770 0.0023 -1.1268 0.1067 -0.5568
## migl -0.1079 -0.2186 0.1770 . 0.1794 -0.9497 0.2837 -0.3797
## piog -0.2873 -0.3979 -0.0023 -0.1794 . -1.1291 0.1044 -0.5591
## plac 0.8418 0.7311 1.1268 0.9497 1.1291 . 1.2335 0.5700
## rosi -0.3917 -0.5023 -0.1067 -0.2837 -0.1044 -1.2335 . -0.6635
## sita 0.2718 0.1611 0.5568 0.3797 0.5591 -0.5700 0.6635 .
## sulf 0.4252 0.3146 0.7102 0.5332 0.7125 -0.4166 0.8169 0.1534
## vild 0.1418 0.0311 0.4268 0.2497 0.4291 -0.7000 0.5335 -0.1300
## sulf vild
## acar -0.4252 -0.1418
## benf -0.3146 -0.0311
## metf -0.7102 -0.4268
## migl -0.5332 -0.2497
## piog -0.7125 -0.4291
## plac 0.4166 0.7000
## rosi -0.8169 -0.5335
## sita -0.1534 0.1300
## sulf . 0.2834
## vild -0.2834 .
##
## Lower 95%-confidence limit:
## acar benf metf migl piog plac rosi sita
## acar . -0.8499 -0.2208 -0.5542 -0.3313 -1.3236 -0.1189 -1.1166
## benf -0.6286 . -0.2414 -0.5030 -0.3089 -1.2918 -0.1118 -1.0534
## metf -0.7908 -1.0327 . -0.7227 -0.4398 -1.4291 -0.2170 -1.3138
## migl -0.7701 -0.9402 -0.3686 . -0.4465 -1.4040 -0.2350 -1.2092
## piog -0.9059 -1.1048 -0.4444 -0.8052 . -1.5596 -0.3347 -1.3758
## plac 0.3600 0.1705 0.8244 0.4955 0.6986 . 0.9830 -0.1240
## rosi -0.9023 -1.1164 -0.4304 -0.8025 -0.5435 -1.4839 . -1.4013
## sita -0.5731 -0.7311 -0.2002 -0.4497 -0.2576 -1.2640 -0.0744 .
## sulf -0.0951 -0.4184 0.2491 -0.1220 0.1205 -0.8887 0.3521 -0.6859
## vild -0.7020 -0.8601 -0.3291 -0.5787 -0.3865 -1.3927 -0.2032 -1.1106
## sulf vild
## acar -0.9456 -0.9856
## benf -1.0476 -0.9224
## metf -1.1713 -1.1826
## migl -1.1883 -1.0781
## piog -1.3045 -1.2447
## plac -0.0556 0.0073
## rosi -1.2817 -1.2701
## sita -0.9928 -0.8506
## sulf . -0.5549
## vild -1.1218 .
##
## Upper 95%-confidence limit:
## acar benf metf migl piog plac rosi sita sulf
## acar . 0.6286 0.7908 0.7701 0.9059 -0.3600 0.9023 0.5731 0.0951
## benf 0.8499 . 1.0327 0.9402 1.1048 -0.1705 1.1164 0.7311 0.4184
## metf 0.2208 0.2414 . 0.3686 0.4444 -0.8244 0.4304 0.2002 -0.2491
## migl 0.5542 0.5030 0.7227 . 0.8052 -0.4955 0.8025 0.4497 0.1220
## piog 0.3313 0.3089 0.4398 0.4465 . -0.6986 0.5435 0.2576 -0.1205
## plac 1.3236 1.2918 1.4291 1.4040 1.5596 . 1.4839 1.2640 0.8887
## rosi 0.1189 0.1118 0.2170 0.2350 0.3347 -0.9830 . 0.0744 -0.3521
## sita 1.1166 1.0534 1.3138 1.2092 1.3758 0.1240 1.4013 . 0.6859
## sulf 0.9456 1.0476 1.1713 1.1883 1.3045 0.0556 1.2817 0.9928 .
## vild 0.9856 0.9224 1.1826 1.0781 1.2447 -0.0073 1.2701 0.8506 0.5549
## vild
## acar 0.7020
## benf 0.8601
## metf 0.3291
## migl 0.5787
## piog 0.3865
## plac 1.3927
## rosi 0.2032
## sita 1.1106
## sulf 1.1218
## vild .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.1087; I^2 = 81.4%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 96.99 18 < 0.0001
## Within designs 74.46 11 < 0.0001
## Between designs 22.53 7 0.0021
forest(net2, ref="plac")
data(smokingcessation)
# Transform data from arm-based format to contrast-based format
# Argument 'sm' has to be used for odds ratio as summary measure; by
# default the risk ratio is used in the metabin function called
# internally.
p1 <- pairwise(list(treat1, treat2, treat3),
event=list(event1, event2, event3),
n=list(n1, n2, n3),
data=smokingcessation,
sm="OR")
p1
## TE seTE studlab treat1 treat2 event1 n1 event2 n2
## 1 -1.051293027 0.4132432 1 A C 9 140 23 140
## 2 -0.128527575 0.4759803 1 A D 9 140 10 138
## 3 0.922765452 0.3997972 1 C D 23 140 10 138
## 4 -0.001244555 0.4504070 2 B C 11 78 12 85
## 5 -0.225333286 0.3839393 2 B D 11 78 29 170
## 6 -0.224088731 0.3722995 2 C D 12 85 29 170
## 7 -2.202289286 0.1430439 3 A C 75 731 363 714
## 8 -0.870353637 0.7910933 4 A C 2 106 9 205
## 9 -0.415648522 0.1557329 5 A C 58 549 237 1561
## 10 -2.779683746 1.4698402 6 A C 0 33 9 48
## 11 -2.705393275 0.6251608 7 A C 3 100 31 98
## 12 -2.425187415 1.0422512 8 A C 1 31 26 95
## 13 -0.443616874 0.5219769 9 A C 6 39 17 77
## 14 0.015964936 0.1699150 10 A B 79 702 77 694
## 15 -0.393504544 0.3265754 11 A B 18 671 21 535
## 16 -0.390412268 0.1680177 12 A C 64 642 107 761
## 17 -0.106335650 0.5955997 13 A C 5 62 8 90
## 18 -0.583398287 0.2983467 14 A C 20 234 34 237
## 19 -3.522516830 1.4969970 15 A D 0 20 9 20
## 20 -0.679594214 0.4411158 16 A B 8 116 19 149
## 21 -0.539679846 0.1401199 17 A C 95 1107 143 1031
## 22 0.125505082 0.3199924 18 A C 15 187 36 504
## 23 0.239970162 0.1736564 19 A C 78 584 73 675
## 24 -0.038956007 0.1873842 20 A C 69 1177 54 888
## 25 0.151684587 0.4289753 21 B C 20 49 16 43
## 26 -1.043486306 0.4489795 22 B D 7 66 32 127
## 27 -0.680724661 0.4092394 23 C D 12 76 20 74
## 28 0.405465108 0.7139060 24 C D 9 55 3 26
## incr allstudies
## 1 0.0 FALSE
## 2 0.0 FALSE
## 3 0.0 FALSE
## 4 0.0 FALSE
## 5 0.0 FALSE
## 6 0.0 FALSE
## 7 0.0 FALSE
## 8 0.0 FALSE
## 9 0.0 FALSE
## 10 0.5 FALSE
## 11 0.0 FALSE
## 12 0.0 FALSE
## 13 0.0 FALSE
## 14 0.0 FALSE
## 15 0.0 FALSE
## 16 0.0 FALSE
## 17 0.0 FALSE
## 18 0.0 FALSE
## 19 0.5 FALSE
## 20 0.0 FALSE
## 21 0.0 FALSE
## 22 0.0 FALSE
## 23 0.0 FALSE
## 24 0.0 FALSE
## 25 0.0 FALSE
## 26 0.0 FALSE
## 27 0.0 FALSE
## 28 0.0 FALSE
# Conduct network meta-analysis
net1 <- netmeta(p1)
net1
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms multiarm
## 1 A C -1.0513 0.4132 0.4776 3 *
## 1 A D -0.1285 0.4760 0.6875 3 *
## 1 C D 0.9228 0.3998 0.4551 3 *
## 2 B C -0.0012 0.4504 0.6707 3 *
## 2 B D -0.2253 0.3839 0.4390 3 *
## 2 C D -0.2241 0.3723 0.4204 3 *
## 3 A C -2.2023 0.1430 0.1430 2
## 4 A C -0.8704 0.7911 0.7911 2
## 5 A C -0.4156 0.1557 0.1557 2
## 6 A C -2.7797 1.4698 1.4698 2
## 7 A C -2.7054 0.6252 0.6252 2
## 8 A C -2.4252 1.0423 1.0423 2
## 9 A C -0.4436 0.5220 0.5220 2
## 10 A B 0.0160 0.1699 0.1699 2
## 11 A B -0.3935 0.3266 0.3266 2
## 12 A C -0.3904 0.1680 0.1680 2
## 13 A C -0.1063 0.5956 0.5956 2
## 14 A C -0.5834 0.2983 0.2983 2
## 15 A D -3.5225 1.4970 1.4970 2
## 16 A B -0.6796 0.4411 0.4411 2
## 17 A C -0.5397 0.1401 0.1401 2
## 18 A C 0.1255 0.3200 0.3200 2
## 19 A C 0.2400 0.1737 0.1737 2
## 20 A C -0.0390 0.1874 0.1874 2
## 21 B C 0.1517 0.4290 0.4290 2
## 22 B D -1.0435 0.4490 0.4490 2
## 23 C D -0.6807 0.4092 0.4092 2
## 24 C D 0.4055 0.7139 0.7139 2
##
## Number of treatment arms (by study):
## narms
## 1 3
## 10 2
## 11 2
## 12 2
## 13 2
## 14 2
## 15 2
## 16 2
## 17 2
## 18 2
## 19 2
## 2 3
## 20 2
## 21 2
## 22 2
## 23 2
## 24 2
## 3 2
## 4 2
## 5 2
## 6 2
## 7 2
## 8 2
## 9 2
##
## Results (fixed effect model):
##
## treat1 treat2 OR 95%-CI Q leverage
## 1 A C 0.5208 [0.4640; 0.5846] 0.70 0.02
## 1 A D 0.4883 [0.3379; 0.7057] 0.73 0.07
## 1 C D 0.9376 [0.6526; 1.3472] 4.71 0.17
## 2 B C 0.6359 [0.4894; 0.8263] 0.45 0.04
## 2 B D 0.5963 [0.4030; 0.8822] 0.44 0.21
## 2 C D 0.9376 [0.6526; 1.3472] 0.14 0.19
## 3 A C 0.5208 [0.4640; 0.5846] 117.39 0.17
## 4 A C 0.5208 [0.4640; 0.5846] 0.08 0.01
## 5 A C 0.5208 [0.4640; 0.5846] 2.31 0.14
## 6 A C 0.5208 [0.4640; 0.5846] 2.09 0.00
## 7 A C 0.5208 [0.4640; 0.5846] 10.78 0.01
## 8 A C 0.5208 [0.4640; 0.5846] 2.89 0.00
## 9 A C 0.5208 [0.4640; 0.5846] 0.16 0.01
## 10 A B 0.8189 [0.6397; 1.0483] 1.61 0.55
## 11 A B 0.8189 [0.6397; 1.0483] 0.35 0.15
## 12 A C 0.5208 [0.4640; 0.5846] 2.43 0.12
## 13 A C 0.5208 [0.4640; 0.5846] 0.84 0.01
## 14 A C 0.5208 [0.4640; 0.5846] 0.05 0.04
## 15 A D 0.4883 [0.3379; 0.7057] 3.51 0.02
## 16 A B 0.8189 [0.6397; 1.0483] 1.18 0.08
## 17 A C 0.5208 [0.4640; 0.5846] 0.65 0.18
## 18 A C 0.5208 [0.4640; 0.5846] 5.91 0.03
## 19 A C 0.5208 [0.4640; 0.5846] 26.41 0.12
## 20 A C 0.5208 [0.4640; 0.5846] 10.72 0.10
## 21 B C 0.6359 [0.4894; 0.8263] 1.98 0.10
## 22 B D 0.5963 [0.4030; 0.8822] 1.37 0.20
## 23 C D 0.9376 [0.6526; 1.3472] 2.27 0.20
## 24 C D 0.9376 [0.6526; 1.3472] 0.43 0.07
##
## Number of studies: k = 24
## Number of treatments: n = 4
## Number of pairwise comparisons: m = 28
## Number of designs: d = 8
##
## Fixed effect model
##
## Treatment estimate (sm = 'OR'):
## A B C D
## A . 0.8189 0.5208 0.4883
## B 1.2211 . 0.6359 0.5963
## C 1.9202 1.5725 . 0.9376
## D 2.0479 1.6771 1.0665 .
##
## Lower 95%-confidence limit:
## A B C D
## A . 0.6397 0.4640 0.3379
## B 0.9539 . 0.4894 0.4030
## C 1.7107 1.2102 . 0.6526
## D 1.4170 1.1335 0.7423 .
##
## Upper 95%-confidence limit:
## A B C D
## A . 1.0483 0.5846 0.7057
## B 1.5631 . 0.8263 0.8822
## C 2.1554 2.0433 . 1.3472
## D 2.9598 2.4814 1.5323 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0.5989; I^2 = 88.6%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 202.62 23 < 0.0001
## Within designs 187.40 16 < 0.0001
## Between designs 15.22 7 0.0333
# Draw network graph
netgraph(net1, points=TRUE, cex.points=3, cex=1.25)
tname <- c("No intervention","Self-help","Individual counselling","Group
counselling")
netgraph(net1, points=TRUE, cex.points=3, cex=1.25, labels=tname)
data(Woods2010)
# Transform data from long arm-based format to contrast-based format
# Argument 'sm' has to be used for odds ratio as summary measure; by
# default the risk ratio is used in the metabin function called
# internally.
p1 <- pairwise(treatment, event = r, n = N,
studlab = author, data = Woods2010, sm = "OR")
p1
## TE seTE studlab treat1 treat2 event1 n1
## 1 0.00881063 1.4173252 Boyd 1997 Placebo Salmeterol 1 227
## 2 -0.60382188 0.6311772 Calverly 2003 Fluticasone Placebo 4 374
## 3 0.28497571 0.7672979 Calverly 2003 Fluticasone Salmeterol 4 374
## 4 0.65457491 0.8692018 Calverly 2003 Fluticasone SFC 4 374
## 5 0.88879759 0.6940644 Calverly 2003 Placebo Salmeterol 7 361
## 6 1.25839679 0.8052894 Calverly 2003 Placebo SFC 7 361
## 7 0.36959919 0.9158888 Calverly 2003 Salmeterol SFC 3 372
## 8 1.41751820 1.2270043 Celli 2003 Placebo Salmeterol 2 270
## event2 n2 incr allstudies
## 1 1 229 0 FALSE
## 2 7 361 0 FALSE
## 3 3 372 0 FALSE
## 4 2 358 0 FALSE
## 5 3 372 0 FALSE
## 6 2 358 0 FALSE
## 7 2 358 0 FALSE
## 8 1 554 0 FALSE
# Conduct network meta-analysis
net1 <- netmeta(p1)
net1
## Original data (with adjusted standard errors for multi-arm studies):
##
## treat1 treat2 TE seTE seTE.adj narms
## Boyd 1997 Placebo Salmeterol 0.0088 1.4173 1.4173 2
## Calverly 2003 Fluticasone Placebo -0.6038 0.6312 0.7623 4
## Calverly 2003 Fluticasone Salmeterol 0.2850 0.7673 1.1578 4
## Calverly 2003 Fluticasone SFC 0.6546 0.8692 1.4163 4
## Calverly 2003 Placebo Salmeterol 0.8888 0.6941 0.8791 4
## Calverly 2003 Placebo SFC 1.2584 0.8053 1.0753 4
## Calverly 2003 Salmeterol SFC 0.3696 0.9159 1.6332 4
## Celli 2003 Placebo Salmeterol 1.4175 1.2270 1.2270 2
## multiarm
## Boyd 1997
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Calverly 2003 *
## Celli 2003
##
## Number of treatment arms (by study):
## narms
## Boyd 1997 2
## Calverly 2003 4
## Celli 2003 2
##
## Results (fixed effect model):
##
## treat1 treat2 OR 95%-CI Q
## Boyd 1997 Placebo Salmeterol 2.3678 [0.7967; 7.0370] 0.36
## Calverly 2003 Fluticasone Placebo 0.5512 [0.1640; 1.8526] 0.00
## Calverly 2003 Fluticasone Salmeterol 1.3051 [0.3243; 5.2517] 0.00
## Calverly 2003 Fluticasone SFC 1.9243 [0.3503; 10.5717] 0.00
## Calverly 2003 Placebo Salmeterol 2.3678 [0.7967; 7.0370] 0.00
## Calverly 2003 Placebo SFC 3.4913 [0.7344; 16.5976] 0.00
## Calverly 2003 Salmeterol SFC 1.4745 [0.2686; 8.0933] 0.00
## Celli 2003 Placebo Salmeterol 2.3678 [0.7967; 7.0370] 0.21
## leverage
## Boyd 1997 0.15
## Calverly 2003 0.66
## Calverly 2003 0.38
## Calverly 2003 0.38
## Calverly 2003 0.40
## Calverly 2003 0.55
## Calverly 2003 0.28
## Celli 2003 0.21
##
## Number of studies: k = 3
## Number of treatments: n = 4
## Number of pairwise comparisons: m = 8
## Number of designs: d = 2
##
## Fixed effect model
##
## Treatment estimate (sm = 'OR'):
## Fluticasone Placebo Salmeterol SFC
## Fluticasone . 0.5512 1.3051 1.9243
## Placebo 1.8143 . 2.3678 3.4913
## Salmeterol 0.7662 0.4223 . 1.4745
## SFC 0.5197 0.2864 0.6782 .
##
## Lower 95%-confidence limit:
## Fluticasone Placebo Salmeterol SFC
## Fluticasone . 0.1640 0.3243 0.3503
## Placebo 0.5398 . 0.7967 0.7344
## Salmeterol 0.1904 0.1421 . 0.2686
## SFC 0.0946 0.0602 0.1236 .
##
## Upper 95%-confidence limit:
## Fluticasone Placebo Salmeterol SFC
## Fluticasone . 1.8526 5.2517 10.5717
## Placebo 6.0982 . 7.0370 16.5976
## Salmeterol 3.0834 1.2552 . 8.0933
## SFC 2.8549 1.3616 3.7225 .
##
## Quantifying heterogeneity / inconsistency:
## tau^2 = 0; I^2 = 0%
##
## Tests of heterogeneity (within designs) and inconsistency (between designs):
## Q d.f. p-value
## Total 0.57 2 0.7525
## Within designs 0.56 1 0.4524
## Between designs 0.00 1 0.9485
## Not run:
# Show forest plot
forest(net1, ref = "Placebo", drop = TRUE,
leftlabs = "Contrast to Placebo")
