icerui: Calculate Uncertainty Intervals for Incremental Cost-Effectiveness Ratios (ICERs)

This package provides several methods for calculating uncertainty intervals (confidence intervals, credible intervals) for the incremental cost-effectiveness ratio (ICER), which is the ratio of incremental costs to incremental effects. It includes the bootstrap percentile method, Fieller method and bootstrap acceptability method.

⚠️ Early stages of development. This package is in the early stages of development. Please do not assume there will be no breaking changes and check back regularly.

Installation

You can install the latest “stable” version from GitHub with:

# install.packages("devtools")
devtools::install_github("tristansnowsill/icerui")

Example

The package includes a single example dataset ex_smoking constructed from the BCEA package, which we will use here:

library(icerui)

confint(ex_smoking, method = "fieller")
#>                            2.5%      97.5%
#> No treatment           98.26914 4910.70352
#> Self-help              74.72212 -263.56608
#> Individual counselling 37.72638  -82.84373

There are three methods currently implemented:

fieller, Fieller method (parametric)
percentile, Bootstrap percentile (non-parametric)
acceptability, Bootstrap acceptability (non-parametric)

The bootstrap percentile method is implemented as suggested in Glick et al. (2014).

It is strongly recommended that you check the confidence intervals produced are sensible.

The uiplot function can help with this:

uiplot(ex_smoking, method = "percentile", comparison = 1, graph = "base")

Bibliography

Glick HA, Doshi JA, Sonnad SS, Polsky D. (2014) Economic evaluation in clinical trials (2nd Edition). Oxford: Oxford University Press. ISBN 978-0-19-968502-8. doi: 10.1093/med/9780199685028.001.0001