darthpack is an R package that showcases the Decision Analysis in R for Technologies in Health (DARTH) coding framework to construct model-based cost-effectiveness analysis in R. The main website of darthpack can be found here.

darthpack is part of the following manuscript:

• Alarid-Escudero F, Krijkamp E, Pechlivanoglou P, Jalal H, Kao SY, Yang A, Enns EA. “A need for change! A coding framework for improving transparency in decision modeling”. PharmacoEconomics 2019;37(11):1329–1339. http://dx.doi.org/10.1007/s40273-019-00837-x

Version included in the published paper. The article above describes darthpack release v0.1.1 (package version 0.1.0), archived in Zenodo under DOI 10.5281/zenodo.3445451. Cite that DOI, and use that release, to reproduce the results reported in the paper. Later releases fix defects in the framework and change some of the cost-effectiveness results; see the NEWS file.

Each release has its own Zenodo DOI, and one further DOI covers the deposit as a whole:

The first three are version DOIs, each permanently tied to one release. The last is Zenodo’s concept DOI: it deliberately forwards to whichever release is newest, so it opens a different record number as new releases appear. Cite it for darthpack in general, and a version DOI when you need a specific release. The DOI badge at the top of this page points to the concept DOI.

Version 0.2.0 is a new and improved version of the framework. It corrects a number of defects in the framework functions – including a log-likelihood bug that corrupted Bayesian calibration whenever a parameter set could not be evaluated, discount rates that were applied to the wrong outcome in calculate_ce_out(), a generate_psa_params() that ignored both its n_sim and its seed argument, and a transition-array check that passed invalid matrices – and it adds input validation so that common mistakes, such as a misspelled parameter name, now raise an error instead of silently returning results for the base case. The test suite grew from 27 to 294 assertions and R CMD check now passes cleanly.

The calibration is unchanged: re-running the IMIS calibration reproduces the published posterior distribution bit-for-bit, and the framework functions return the same values as version 0.1.0 at the parameters shipped with the package. All results in data/, output/, figs/ and tables/ have been regenerated by running analysis/_master.R against the fixed code, so every committed figure and table is now what the code in this repository actually produces. The deterministic cost-effectiveness table changed the most, because it had been generated by an older version of calculate_ce_out() and never refreshed; its ICER now agrees with the probabilistic one to within Monte Carlo error, where the two previously differed by 29%. The NEWS file describes every change.

To learn more on Markov/cohort state-transition model in R, we recommend reading our introductory tutorial on time-independent cSTMs in R:

• Alarid-Escudero F, Krijkamp EM, Enns EA, Yang A, Hunink MGM, Pechlivanoglou P, Jalal H. An Introductory Tutorial on Cohort State-Transition Models in R Using a Cost-Effectiveness Analysis Example. Medical Decision Making, 2023;43(1):3-20. https://doi.org/10.1177/0272989X221103163,

our intermediate tutorial on time-dependent cohort state-transition models (cSTMs) in R:

• Alarid-Escudero F, Krijkamp EM, Enns EA, Yang A, Hunink MGM, Pechlivanoglou P, Jalal H. A Tutorial on Time-Dependent Cohort State-Transition Models in R using a Cost-Effectiveness Analysis Example. Medical Decision Making. 2023;43(1):21-41. https://doi.org/10.1177/0272989X221121747,

and understanding the use of multidimensional arrays to represent cSTM dynamics in R described in:

• Krijkamp EM, Alarid-Escudero F, Enns EA, Pechlivanoglou P, Hunink MGM, Yang A, Jalal HJ. A multidimensional array representation of state-transition model dynamics. Medical Decision Making, 2020;40(2):242-248. https://doi.org/10.1177/0272989X19893973

To model individual-level rather than cohort dynamics, we recommend our tutorial on microsimulation, also known as individual-based state-transition models (iSTMs), in R:

• Krijkamp EM, Alarid-Escudero F, Enns EA, Jalal HJ, Hunink MGM, Pechlivanoglou P. Microsimulation Modeling for Health Decision Sciences Using R: A Tutorial. Medical Decision Making, 2018;38(3):400-422. https://doi.org/10.1177/0272989X18754513

and, for simulating individual-level state-transition models in continuous time, our tutorial on discrete event simulation (DES) in R:

• Lopez-Mendez M, Goldhaber-Fiebert JD, Alarid-Escudero F. A Tutorial on Discrete Event Simulation Models Using a Cost-Effectiveness Analysis Example in R. Medical Decision Making, 2026;46(5):533-548. https://doi.org/10.1177/0272989X261437826

• Install RStudio
• Install devtools to install darthpack as a package and modify it to generate your own package

# Install release version from CRAN
install.packages("devtools")

# Or install development version from GitHub
# devtools::install_github("r-lib/devtools")

• Install pkgdown to publish darthpack or your own darthpack-based repository or package as a website (optional)

# Install release version from CRAN
install.packages("pkgdown")

# Or install development version from GitHub
# devtools::install_github("r-lib/pkgdown")

darthpack repository could be used in at least three different ways:

1. GitHub coding template for using it to generate a repository of your own model-based decision or cost-effectiveness analysis linked to the original darthpack GitHub repository
2. Regular coding template for using it to generate a repository of your own model-based decision or cost-effectiveness analysis
3. R package for using it as a standalone package to run current functions of darthpack

The main website of the package could be found in: https://darth-git.github.io/darthpack/

1. Sign in to GitHub. You need to sign in to use this repository as a template. If you don’t have a GitHub account, you can create one here.
2. On the darthpack GitHub repository, navigate to the main page of the repository (https://github.com/DARTH-git/darthpack).
3. Above the file list, click Use this template.
4. Use the Owner drop-down menu, and select the account you want to own the repository.
5. Type a name for your repository of your decision model, and an optional description.
6. Choose to make the repository either public or private. Public repositories are visible to the public, while private repositories are only accessible to you, and people you share them with. For more information, see “Setting repository visibility.”
7. Click Create repository from template.
8. Either clone the repository or download it.
   1. Clone, which requires the user to have a GitHub desktop installed, or
   2. Download zip that will ask the user to download the whole repository as a .zip file.
9. Open the RStudio project darthpack.Rproj.
10. Install all the required and suggested packages listed in the DESCRIPTION file in the main folder of the repository
    • dampack is on CRAN, but IMIS, used by the calibration component, has been archived and has to be installed from the CRAN archive:

install.packages("dampack")
devtools::install_version("IMIS", version = "0.1", repos = "https://cloud.r-project.org")

11. In RStudio, load all the functions and data from the repository by typing devtools::load_all(".")
12. Run all the decision modeling modules in the analysis folder.

1. On the darthpack GitHub repository, navigate to the main page of the repository (https://github.com/DARTH-git/darthpack).
2. Above the file list, click Clone or download and select either
   1. Open in desktop, which requires the user to have a GitHub desktop installed, or
   2. Download zip that will ask the user to download the whole repository as a .zip file.
3. Open the RStudio project darthpack.Rproj.
4. Install all the required and suggested packages listed in the DESCRIPTION file in the main folder of the repository
   • dampack is on CRAN, but IMIS, used by the calibration component, has been archived and has to be installed from the CRAN archive:

install.packages("dampack")
devtools::install_version("IMIS", version = "0.1", repos = "https://cloud.r-project.org")

5. In RStudio, load all the functions and data from the repository by typing devtools::load_all(".")
6. Run all the decision modeling modules in the analysis folder.

1. Install the development version of darthpack from GitHub with:

devtools::install_github("DARTH-git/darthpack")

2. Load all the functions and data from the repository by typing

library(darthpack)

Once the framework has been modified and updated to your specific needs, run pkgdown from the package directory each time you release your package:

pkgdown::build_site()

For a more detailed description on how to quickly and easily build a website for your package, please go to https://github.com/r-lib/pkgdown and the pkgdown documentation at https://pkgdown.r-lib.org.