Advancing Cost-Effectiveness Modeling With R: A Flexible Approach for HTA Submissions
Author(s)
Rui Cai, PhD1, Jack Ettinger, MSc2, Jean-Etienne Poirrier, MBA, PhD3.
1Principal, PAREXEL, Gouda, Netherlands, 2PAREXEL International - Access Consulting, London, United Kingdom, 3Parexel Belgium, Wavre, Belgium.
1Principal, PAREXEL, Gouda, Netherlands, 2PAREXEL International - Access Consulting, London, United Kingdom, 3Parexel Belgium, Wavre, Belgium.
OBJECTIVES: Cost-effectiveness models (CEMs) are crucial tools in health technology assessment (HTA) submissions. Traditionally, these models have been developed in Excel, but R offers several advantages in terms of flexibility, reproducibility, and efficiency. We aim to demonstrate the benefits throughout the HTA submission lifecycle of developing a CEM template in R that can replicate Excel-based models.
METHODS: We developed a user-friendly R-based partitioned-survival CEM template including survival analysis. The template was built with adaptability in mind, allowing for application across various therapeutic areas and model structures. Input and output can be entered via Excel files or a web interface. The code is segmented into separate functions making the model easy to adapt by local operating companies or HTA bodies.
RESULTS: The R-based CEM template demonstrated several advantages. The integration of survival analysis improves workflows between statisticians and modelers, speeding up model development. The built-in functions are easily interchanged and amended, allowing more efficient development, scenario exploration, and structural changes. Therefore, the template can be easily adapted to meet specific HTA body requirements, maintaining consistency with Excel-based approaches when necessary. The use of R means that probabilistic sensitivity analyses are faster and reproducible, clear limitations of Excel models. Country adaptations can be integrated into standard workflows by those more comfortable using Excel. The template can also be used to validate Excel models where HTA bodies do not accept R-based models.
CONCLUSIONS: The R-based CEM template offers a flexible and efficient approach to cost-effectiveness modeling and validation throughout the HTA submission lifecycle, while maintaining compatibility with Excel-based methods. As HTA bodies increasingly recognize the benefits of R, this approach positions submissions at the forefront of methodological advancements in health economic modeling. Using R could also facilitate the integration of AI into modelling practices in the future.
METHODS: We developed a user-friendly R-based partitioned-survival CEM template including survival analysis. The template was built with adaptability in mind, allowing for application across various therapeutic areas and model structures. Input and output can be entered via Excel files or a web interface. The code is segmented into separate functions making the model easy to adapt by local operating companies or HTA bodies.
RESULTS: The R-based CEM template demonstrated several advantages. The integration of survival analysis improves workflows between statisticians and modelers, speeding up model development. The built-in functions are easily interchanged and amended, allowing more efficient development, scenario exploration, and structural changes. Therefore, the template can be easily adapted to meet specific HTA body requirements, maintaining consistency with Excel-based approaches when necessary. The use of R means that probabilistic sensitivity analyses are faster and reproducible, clear limitations of Excel models. Country adaptations can be integrated into standard workflows by those more comfortable using Excel. The template can also be used to validate Excel models where HTA bodies do not accept R-based models.
CONCLUSIONS: The R-based CEM template offers a flexible and efficient approach to cost-effectiveness modeling and validation throughout the HTA submission lifecycle, while maintaining compatibility with Excel-based methods. As HTA bodies increasingly recognize the benefits of R, this approach positions submissions at the forefront of methodological advancements in health economic modeling. Using R could also facilitate the integration of AI into modelling practices in the future.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
EE43
Topic
Economic Evaluation, Methodological & Statistical Research
Topic Subcategory
Trial-Based Economic Evaluation
Disease
Oncology