Statistical analysis
Using data from CORONA and GISSI-HF, we analysed competing risks
for fatal and non-fatal MI and stroke, other cardiovascular causes of
death, and death from non-cardiovascular causes for the treatment
group compared with the placebo group. We then analysed competing
risks for fatal and non-fatal MI and stroke within risk subgroups for
the treatment group compared with the placebo group. In this competing
risks analysis, the overall competing risk cumulative incidence
for all events is equal to the sum of the cumulative incidences for each
type of event since only the first event is counted.17 For our analysis,
we determined the first event each participant experienced following
randomization over the course of follow-up: MI, stroke, death from
other cardiovascular causes, death from non-cardiovascular causes, or
none of these events. We then estimated the Kaplan–Meier cumulative
incidence for occurrence of these events. We considered inclusion
of unstable angina in a composite ischaemic endpoint; however, there
were few events of unstable angina and the reliability of this measure
is variable, therefore, unstable angina was not included in our
analysis.