Results from 4 trials examining the effectiveness of intensive (high dose) versus moderate (standard dose) statin therapy for preventing coronary death or myocardial infarction.

dat.cannon2006

Format

The data frame contains the following columns:

trialcharactertrial name
popcharacterstudy population (post-ACS: post acute coronary syndrome; stable CAD: stable coronary artery disease)
ntnumericnumber of patients in the high dose group
ncnumericnumber of patients in the standard dose group
ep1tnumericnumber of events in the high dose group for end point 1: coronary death or non-fatal myocardial infarction
ep1cnumericnumber of events in the standard dose group for end point 1: coronary death or non-fatal myocardial infarction
ep2tnumericnumber of events in the high dose group for end point 2: coronary death or any cardiovascular event (MI, stroke, hospitalization for unstable angina, or revascularization)
ep2cnumericnumber of events in the standard dose group for end point 2: coronary death or any cardiovascular event (MI, stroke, hospitalization for unstable angina, or revascularization)
ep3tnumericnumber of events in the high dose group for end point 3: cardiovascular death
ep3cnumericnumber of events in the standard dose group for end point 3: cardiovascular death
ep4tnumericnumber of events in the high dose group for end point 4: non-cardiovascular death
ep4cnumericnumber of events in the standard dose group for end point 4: non-cardiovascular death
ep5tnumericnumber of events in the high dose group for end point 5: deaths (all-cause mortality)
ep5cnumericnumber of events in the standard dose group for end point 5: deaths (all-cause mortality)
ep6tnumericnumber of events in the high dose group for end point 6: stroke
ep6cnumericnumber of events in the standard dose group for end point 6: stroke

Details

The data were obtained from Figures 2, 3, 4, and 5 in Cannon et al. (2006). The authors used the Mantel-Haenszel method for combining the results from the 4 trials. This approach is implemented in the rma.mh function.

Source

Cannon, C. P., Steinberg, B. A., Murphy, S. A., Mega, J. L., & Braunwald, E. (2006). Meta-analysis of cardiovascular outcomes trials comparing intensive versus moderate statin therapy. Journal of the American College of Cardiology, 48(3), 438–445. https://doi.org/10.1016/j.jacc.2006.04.070

Concepts

medicine, cardiology, odds ratios, Mantel-Haenszel method

Examples

### copy data into 'dat' and examine data
dat <- dat.cannon2006
dat
#>      trial        pop   nt   nc ep1t ep1c ep2t ep2c ep3t ep3c ep4t ep4c ep5t ep5c ep6t ep6c
#> 1 PROVE IT   post-ACS 2099 2063  147  172  496  554   27   36   17   27   50   69   20   17
#> 2   A-TO-Z   post-ACS 2265 2232  205  235  895  844   86  111   22   21  108  132   28   35
#> 3      TNT stable CAD 4995 5006  334  418 1405 1677  126  155  158  127  284  282  117  155
#> 4    IDEAL stable CAD 4439 4449  411  463 1176 1370  223  218  143  156  366  374  151  174

# \dontrun{

### load metafor package
library(metafor)

### meta-analysis of log odds ratios using the MH method for endpoint 1
res <- rma.mh(measure="OR", ai=ep1t, n1i=nt, ci=ep1c, n2i=nc, data=dat, slab=trial)
print(res, digits=2)
#> 
#> Equal-Effects Model (k = 4)
#> 
#> I^2 (total heterogeneity / total variability):  0.00%
#> H^2 (total variability / sampling variability): 0.38
#> 
#> Test for Heterogeneity: 
#> Q(df = 3) = 1.14, p-val = 0.77
#> 
#> Model Results (log scale):
#> 
#> estimate    se   zval  pval  ci.lb  ci.ub 
#>    -0.18  0.04  -4.18  <.01  -0.26  -0.10 
#> 
#> Model Results (OR scale):
#> 
#> estimate  ci.lb  ci.ub 
#>     0.84   0.77   0.91 
#> 
#> Cochran-Mantel-Haenszel Test:    CMH = 17.33, df = 1, p-val < 0.01
#> Tarone's Test for Heterogeneity: X^2 =  1.14, df = 3, p-val = 0.77
#> 

### forest plot
forest(res, xlim=c(-.8,.8), atransf=exp, at=log(c(2/3, 1, 3/2)),
       header=TRUE, top=2, cex=1.2, xlab="Odds Ratio")
mtext("(high dose better)", side=1, line=par("mgp")[1]-0.5, at=log(2/3), cex=1.2, font=3)
mtext("(standard dose better)", side=1, line=par("mgp")[1]-0.5, at=log(3/2), cex=1.2, font=3)


# }