dat.collins1985b.RdResults from 9 studies examining the effects of diuretics in pregnancy on various outcomes.
dat.collins1985bThe data frame contains the following columns:
| id | numeric | study number |
| author | character | study author(s) |
| year | numeric | publication year |
| pre.nti | numeric | number of women in treatment group followed up for pre-eclampsia outcome |
| pre.nci | numeric | number of women in control/placebo group followed up for pre-eclampsia outcome |
| pre.xti | numeric | number of women in treatment group with any form of pre-eclampsia |
| pre.xci | numeric | number of women in control/placebo group with any form of pre-eclampsia |
| oedema | numeric | dummy variable indicating whether oedema was a diagnostic criterion |
| fup.nti | numeric | number of women in treatment group followed up for mortality outcomes |
| fup.nci | numeric | number of women in control/placebo group followed up for mortality outcomes |
| ped.xti | numeric | number of perinatal deaths in treatment group |
| ped.xci | numeric | number of perinatal deaths in control/placebo group |
| stb.xti | numeric | number of stillbirths in treatment group |
| stb.xci | numeric | number of stillbirths in control/placebo group |
| ned.xti | numeric | number of neonatal deaths in treatment group |
| ned.xci | numeric | number of neonatal deaths in control/placebo group |
The 9 studies in this dataset examined the effects of diuretics in pregnancy on various outcomes, including the presence of any form of pre-eclampsia, perinatal death, stillbirth, and neonatal death.
Collins, R., Yusuf, S., & Peto, R. (1985). Overview of randomised trials of diuretics in pregnancy. British Medical Journal, 290(6461), 17–23. https://doi.org/10.1136/bmj.290.6461.17
medicine, obstetrics, odds ratios, Peto's method
### copy data into 'dat' and examine data
dat <- dat.collins1985b
dat
#> id author year pre.nti pre.nci pre.xti pre.xci oedema fup.nti fup.nci ped.xti
#> 1 1 Weseley & Douglas 1962 131 136 14 14 0 131 136 1
#> 2 2 Flowers et al. 1962 385 134 21 17 0 335 110 6
#> 3 3 Menzies 1964 57 48 14 24 1 57 48 3
#> 4 4 Fallis et al. 1964 38 40 6 18 0 34 40 1
#> 5 5 Cuadros & Tatum 1964 1011 760 12 35 1 1011 760 14
#> 6 6 Landesman et al. 1965 1370 1336 138 175 0 1370 1336 24
#> 7 7 Kraus et al. 1966 506 524 15 20 0 506 524 14
#> 8 8 Tervila & Vartiainen 1971 108 103 6 2 0 108 103 0
#> 9 9 Campbell & MacGillivray 1975 153 102 65 40 0 153 102 0
#> ped.xci stb.xti stb.xci ned.xti ned.xci
#> 1 4 1 2 0 2
#> 2 3 3 2 3 1
#> 3 2 1 1 2 1
#> 4 3 0 1 1 2
#> 5 13 6 5 8 8
#> 6 19 NA NA NA NA
#> 7 16 6 9 8 7
#> 8 0 0 0 0 0
#> 9 0 0 0 0 0
### load metafor package
library(metafor)
### calculate (log) odds ratio and sampling variance
dat <- escalc(measure="OR", n1i=pre.nti, n2i=pre.nci, ai=pre.xti, ci=pre.xci, data=dat)
summary(dat, digits=2, transf=exp)
#>
#> id author year pre.nti pre.nci pre.xti pre.xci oedema fup.nti fup.nci ped.xti
#> 1 1 Weseley & Douglas 1962 131 136 14 14 0 131 136 1
#> 2 2 Flowers et al. 1962 385 134 21 17 0 335 110 6
#> 3 3 Menzies 1964 57 48 14 24 1 57 48 3
#> 4 4 Fallis et al. 1964 38 40 6 18 0 34 40 1
#> 5 5 Cuadros & Tatum 1964 1011 760 12 35 1 1011 760 14
#> 6 6 Landesman et al. 1965 1370 1336 138 175 0 1370 1336 24
#> 7 7 Kraus et al. 1966 506 524 15 20 0 506 524 14
#> 8 8 Tervila & Vartiainen 1971 108 103 6 2 0 108 103 0
#> 9 9 Campbell & MacGillivray 1975 153 102 65 40 0 153 102 0
#> ped.xci stb.xti stb.xci ned.xti ned.xci yi ci.lb ci.ub
#> 1 4 1 2 0 2 1.04 0.48 2.28
#> 2 3 3 2 3 1 0.40 0.20 0.78
#> 3 2 1 1 2 1 0.33 0.14 0.74
#> 4 3 0 1 1 2 0.23 0.08 0.67
#> 5 13 6 5 8 8 0.25 0.13 0.48
#> 6 19 NA NA NA NA 0.74 0.59 0.94
#> 7 16 6 9 8 7 0.77 0.39 1.52
#> 8 0 0 0 0 0 2.97 0.59 15.07
#> 9 0 0 0 0 0 1.14 0.69 1.91
#>
### meta-analysis using Peto's method for any form of pre-eclampsia
rma.peto(n1i=pre.nti, n2i=pre.nci, ai=pre.xti, ci=pre.xci, data=dat, digits=2)
#>
#> Equal-Effects Model (k = 9)
#>
#> I^2 (total heterogeneity / total variability): 72.74%
#> H^2 (total variability / sampling variability): 3.67
#>
#> Test for Heterogeneity:
#> Q(df = 8) = 29.34, p-val < .01
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> -0.41 0.09 -4.65 <.01 -0.58 -0.24
#>
#> Model Results (OR scale):
#>
#> estimate ci.lb ci.ub
#> 0.66 0.56 0.79
#>
### meta-analysis including only studies where oedema was not a diagnostic criterion
rma.peto(n1i=pre.nti, n2i=pre.nci, ai=pre.xti, ci=pre.xci, data=dat, digits=2, subset=(oedema==0))
#>
#> Equal-Effects Model (k = 7)
#>
#> I^2 (total heterogeneity / total variability): 60.84%
#> H^2 (total variability / sampling variability): 2.55
#>
#> Test for Heterogeneity:
#> Q(df = 6) = 15.32, p-val = 0.02
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> -0.28 0.09 -2.97 <.01 -0.47 -0.10
#>
#> Model Results (OR scale):
#>
#> estimate ci.lb ci.ub
#> 0.76 0.63 0.91
#>
### meta-analyses of mortality outcomes (perinatal deaths, stillbirths, and neonatal deaths)
rma.peto(n1i=fup.nti, n2i=fup.nci, ai=ped.xti, ci=ped.xci, data=dat, digits=2)
#> Warning: Some yi/vi values are NA.
#>
#> Equal-Effects Model (k = 9)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 0.58
#>
#> Test for Heterogeneity:
#> Q(df = 6) = 3.49, p-val = 0.75
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> -0.09 0.18 -0.50 0.62 -0.45 0.27
#>
#> Model Results (OR scale):
#>
#> estimate ci.lb ci.ub
#> 0.91 0.64 1.31
#>
rma.peto(n1i=fup.nti, n2i=fup.nci, ai=stb.xti, ci=stb.xci, data=dat, digits=2)
#> Warning: 1 study with NAs omitted from model fitting.
#> Warning: Some yi/vi values are NA.
#>
#> Equal-Effects Model (k = 8)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 0.20
#>
#> Test for Heterogeneity:
#> Q(df = 5) = 0.99, p-val = 0.96
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> -0.39 0.34 -1.16 0.25 -1.05 0.27
#>
#> Model Results (OR scale):
#>
#> estimate ci.lb ci.ub
#> 0.68 0.35 1.31
#>
rma.peto(n1i=fup.nti, n2i=fup.nci, ai=ned.xti, ci=ned.xci, data=dat, digits=2)
#> Warning: 1 study with NAs omitted from model fitting.
#> Warning: Some yi/vi values are NA.
#>
#> Equal-Effects Model (k = 8)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 0.51
#>
#> Test for Heterogeneity:
#> Q(df = 5) = 2.54, p-val = 0.77
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> -0.15 0.31 -0.47 0.64 -0.76 0.47
#>
#> Model Results (OR scale):
#>
#> estimate ci.lb ci.ub
#> 0.86 0.47 1.59
#>