Purpose: Historically, cost-effectiveness analysis to support health care decision-making has been based on spreadsheet models. Recently, there has been criticism that such models lack transparency and the computational limitations restrict model complexity and realism. Open-source models in efficient statistical software can help increase transparency, flexibility, credibility and reduce errors. In this workshop we demonstrate the advantages of using the R software for economic modelling. Participants will learn the software and its packages to develop multi-state models, build partitioned survival models, conduct value of information analysis, generate graphical result summaries, and create publicly accessible graphical user interfaces.
Description: This workshop will describe the benefits of developing open-source decision models in efficient software, with emphasis on R, for relevant value assessment. These benefits, and the disadvantages of spreadsheet models, will be illustrated with examples from the literature and our own research. A wide range of modelling approaches will be demonstrated, including decision tree, multistate, parametric, spline, and fractional polynomial models. Example models will come from heart disease, oncology, and rheumatology. Efficient approaches to expected value of perfect and sample information will be demonstrated, including meta-modelling and advanced Monte Carlo sampling schemes. Finally, we will demonstrate the development of a cost-effectiveness model by running a script in a novel R package; audience members will be encouraged to suggest modifications to this model in real time. We hope our presentations and live demonstrations will encourage a lively discussion with audience members, who will be encouraged to ask questions and share their experiences.