dat.hartmannboyce2018.RdResults from 133 studies examining the effectiveness of nicotine replacement therapy (NRT) for smoking cessation at 6+ months of follow-up.
dat.hartmannboyce2018The data frame contains the following columns:
| study | numeric | study identifier |
| x.nrt | numeric | number of participants in the NRT group who were abstinent at the follow-up |
| n.nrt | numeric | number of participants in the NRT group |
| x.ctrl | numeric | number of participants in the control group who were abstinent at the follow-up |
| n.ctrl | numeric | number of participants in the control group |
| treatment | character | type of NRT provided in the treatment group |
The dataset includes the results from 133 studies examining the effectiveness of nicotine replacement therapy (NRT) for smoking cessation. The results given in this dataset pertain to abstinence at 6+ months of follow-up. NRT was provided to participants in the treatment groups in various forms as indicated by the treatment variable (e.g., gum, patch, inhalator). Note that the dataset includes 136 rows, since a few studies included multiple treatments.
Hartmann‐Boyce, J., Chepkin, S. C., Ye, W., Bullen, C. & Lancaster, T. (2018). Nicotine replacement therapy versus control for smoking cessation. Cochrane Database of Systematic Reviews, 5, CD000146. https://doi.org//10.1002/14651858.CD000146.pub5
medicine, smoking, risk ratios, Mantel-Haenszel method
### copy data into 'dat' and examine data
dat <- dat.hartmannboyce2018
head(dat, 10)
#> study x.nrt n.nrt x.ctrl n.ctrl treatment
#> 1 Ahluwalia 2006 53 378 42 377 gum
#> 2 Areechon 1988 56 99 37 101 gum
#> 3 Blondal 1989 30 92 22 90 gum
#> 4 Br Thor Society 1983 39 410 111 1208 gum
#> 5 Campbell 1987 13 424 9 412 gum
#> 6 Campbell 1991 21 107 21 105 gum
#> 7 Clavel 1985 24 205 6 222 gum
#> 8 Clavel-Chapelon 1992 47 481 42 515 gum
#> 9 Cooper 2005 17 146 15 147 gum
#> 10 Fagerstrom 1982 30 50 23 50 gum
### load metafor package
library(metafor)
### turn treatment into a factor with the desired ordering
dat$treatment <- factor(dat$treatment, levels=unique(dat$treatment))
### meta-analysis per treatment using the M-H method
lapply(split(dat, dat$treatment), function(x)
rma.mh(measure="RR", ai=x.nrt, n1i=n.nrt,
ci=x.ctrl, n2i=n.ctrl, data=x, digits=2))
#> $gum
#>
#> Equal-Effects Model (k = 56)
#>
#> I^2 (total heterogeneity / total variability): 39.59%
#> H^2 (total variability / sampling variability): 1.66
#>
#> Test for Heterogeneity:
#> Q(df = 55) = 91.05, p-val < .01
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.40 0.03 11.56 <.01 0.33 0.47
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.49 1.40 1.60
#>
#>
#> $patch
#>
#> Equal-Effects Model (k = 51)
#>
#> I^2 (total heterogeneity / total variability): 23.59%
#> H^2 (total variability / sampling variability): 1.31
#>
#> Test for Heterogeneity:
#> Q(df = 50) = 65.44, p-val = 0.07
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.49 0.03 14.23 <.01 0.43 0.56
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.64 1.53 1.75
#>
#>
#> $inhalator
#>
#> Equal-Effects Model (k = 4)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 0.64
#>
#> Test for Heterogeneity:
#> Q(df = 3) = 1.93, p-val = 0.59
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.64 0.17 3.73 <.01 0.30 0.98
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.90 1.36 2.67
#>
#>
#> $`intranasal spray`
#>
#> Equal-Effects Model (k = 4)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 0.54
#>
#> Test for Heterogeneity:
#> Q(df = 3) = 1.63, p-val = 0.65
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.70 0.15 4.53 <.01 0.40 1.00
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 2.02 1.49 2.73
#>
#>
#> $`tablets/lozenges`
#>
#> Equal-Effects Model (k = 8)
#>
#> I^2 (total heterogeneity / total variability): 71.31%
#> H^2 (total variability / sampling variability): 3.48
#>
#> Test for Heterogeneity:
#> Q(df = 7) = 24.39, p-val < .01
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.42 0.07 5.97 <.01 0.28 0.55
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.52 1.32 1.74
#>
#>
#> $`oral spray`
#>
#> Equal-Effects Model (k = 1)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 1.00
#>
#> Test for Heterogeneity:
#> Q(df = 0) = 0.00, p-val = 1.00
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.91 0.35 2.57 0.01 0.21 1.60
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 2.48 1.24 4.94
#>
#>
#> $`choice of product`
#>
#> Equal-Effects Model (k = 7)
#>
#> I^2 (total heterogeneity / total variability): 42.34%
#> H^2 (total variability / sampling variability): 1.73
#>
#> Test for Heterogeneity:
#> Q(df = 6) = 10.41, p-val = 0.11
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.32 0.05 6.34 <.01 0.22 0.42
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.37 1.25 1.52
#>
#>
#> $`patch and inhalator`
#>
#> Equal-Effects Model (k = 1)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 1.00
#>
#> Test for Heterogeneity:
#> Q(df = 0) = 0.00, p-val = 1.00
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.07 0.32 0.21 0.83 -0.55 0.69
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.07 0.57 1.99
#>
#>
#> $`patch and lozenge`
#>
#> Equal-Effects Model (k = 1)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 1.00
#>
#> Test for Heterogeneity:
#> Q(df = 0) = 0.00, p-val = 1.00
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.60 0.30 1.98 0.05 0.01 1.20
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.83 1.01 3.31
#>
#>
#> $`patch and gum`
#>
#> Equal-Effects Model (k = 2)
#>
#> I^2 (total heterogeneity / total variability): 50.04%
#> H^2 (total variability / sampling variability): 2.00
#>
#> Test for Heterogeneity:
#> Q(df = 1) = 2.00, p-val = 0.16
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.14 0.30 0.46 0.64 -0.45 0.72
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.15 0.64 2.06
#>
#>
#> $`patch, gum, and lozenge`
#>
#> Equal-Effects Model (k = 1)
#>
#> I^2 (total heterogeneity / total variability): 0.00%
#> H^2 (total variability / sampling variability): 1.00
#>
#> Test for Heterogeneity:
#> Q(df = 0) = 0.00, p-val = 1.00
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 2.71 1.03 2.63 <.01 0.69 4.72
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 15.00 2.00 112.54
#>
#>
### all combined
rma.mh(measure="RR", ai=x.nrt, n1i=n.nrt,
ci=x.ctrl, n2i=n.ctrl, data=dat, digits=2)
#>
#> Equal-Effects Model (k = 136)
#>
#> I^2 (total heterogeneity / total variability): 38.69%
#> H^2 (total variability / sampling variability): 1.63
#>
#> Test for Heterogeneity:
#> Q(df = 135) = 220.20, p-val < .01
#>
#> Model Results (log scale):
#>
#> estimate se zval pval ci.lb ci.ub
#> 0.44 0.02 21.25 <.01 0.40 0.48
#>
#> Model Results (RR scale):
#>
#> estimate ci.lb ci.ub
#> 1.55 1.49 1.61
#>