Paired vs Multiple Comparisons

Introduction

The intention of this vignette is to show how to plot the CEAC and EIB plots depending on whether we consider all interventions simultaneously or pair-wise against a reference.

Multiple interventions

This situation is when there are more than two interventions to consider. Incremental values can be obtained either always against a fixed reference intervention, such as status-quo, or for all comparisons simultaneously. We will call these a paired comparison or a multiple comparison.

Against a fixed reference intervention

R code

This is the default plot for ceac.plot() so we simply follow the same steps as above with the new data set.

data("Smoking")
he <- bcea(eff, cost, ref = 4, Kmax = 500)
par(mfrow = c(2,1))
ceac.plot(he)
abline(h = 0.5, lty = 2)
abline(v = c(160, 225), lty = 3)
eib.plot(he, plot.cri = FALSE)

Pair-wise comparisons

R code

In BCEA we first we must determine all combinations of paired interventions using the multi.ce() function.

he.multi <- multi.ce(he)
par(mfrow = c(2, 1))
ceac.plot(he.multi)
abline(h = 0.5, lty = 2)
abline(v = c(160, 225), lty = 3)
eib.plot(he, plot.cri = FALSE)

mirror server hosted at Truenetwork, Russian Federation.