bayesmeta (version 2.6)

Peto1980: Aspirin after myocardial infarction example data

Description

Numbers of cases (patients) and events (deaths) in treatment and control groups of six studies.

Usage

data("Peto1980")

Format

The data frame contains the following columns:

publication	character	publication identifier
treat.cases	numeric	number of cases in treatment group
treat.events	numeric	number of events in treatment group
control.cases	numeric	number of cases in control group
control.events	numeric	number of events in control group

Details

Quoting from Brockwell and Gordon (2001): “The collection consists of six studies, each examining the effect of aspirin after myocardial infarction. In each study the number of patients who died after having been given either aspirin or a control drug is recorded.”

References

R. Peto. Aspirin after myocardial infarction. The Lancet, 315(8179):1172-1173, 1980.

P.C. Elwood, A.L. Cochrane, M.L.Burr, P.M. Sweetnam, G. Williams, E. Welsby, S.J. Hughes, R. Renton. A randomized controlled trial of acetyl salicylic acid in the secondary prevention of mortality from myocardial infarction. British Medical Journal, 1(5905):436-440, 1974.

The Coronary Drug Project Research Group. Aspirin in coronary heart disease. Journal of Chronic Diseases, 29(10):625-642, 1976.

K. Breddin, D. Loew, K. Lechner, K. Ueberla, E. Walter. Secondary prevention of myocardial infarction: a comparison of acetylsalicylic acid, placebo and phenprocoumon. Haemostasis, 9(6):325-344, 1980.

P.C. Elwood, P.M. Sweetnam. Aspirin and secondary mortality after myocardial infarction. The Lancet, 314(8156):1313-1315, 1979.

Aspirin Myocardial Infarction Study Research Group. A randomized, controlled trial of aspirin in persons recovered from myocardial infarction. Journal of the American Medical Association, 243(7):661-669, 1980.

The Persantine-Aspirin Reinfarction Study Research Group. Persantine and aspirin in coronary heart disease. Circulation, 62(3):449-461, 1980.

Examples

# NOT RUN {
data("Peto1980")
# }
# NOT RUN {
# compute effect sizes (log odds ratios) from count data
# (using "metafor" package's "escalc()" function):
require("metafor")
peto.es <- escalc(measure="OR",
                  ai=treat.events,   n1i=treat.cases,
                  ci=control.events, n2i=control.cases,
                  slab=publication, data=Peto1980)
print(peto.es)

# check sensitivity to different prior choices:
peto.ma01 <- bayesmeta(peto.es)
peto.ma02 <- bayesmeta(peto.es, tau.prior=function(t){dhalfnormal(t, scale=1)})
 
par(mfrow=c(2,1))
  plot(peto.ma01, which=4, prior=TRUE, taulim=c(0,1), main="uniform prior")
  plot(peto.ma02, which=4, prior=TRUE, taulim=c(0,1), main="half-normal prior")
par(mfrow=c(1,1))

# compare heterogeneity (tau) estimates:
print(rbind("uniform"    =peto.ma01$summary[,"tau"],
            "half-normal"=peto.ma02$summary[,"tau"]))

# compare effect (mu) estimates:
print(rbind("uniform"    =peto.ma01$summary[,"mu"],
            "half-normal"=peto.ma02$summary[,"mu"]))

summary(peto.ma02)
forestplot(peto.ma02)
plot(peto.ma02)
# }