Modeling in R: Tutorial for Beginner Using a Cost-Effectiveness Analysis Made With Microsoft Excel
Author(s)
Koffi Jean M. KOUAME, PhD1, Carole SIANI, PhD2, Simon LaRue, Msc3, Soualio Gnanou, PhD Student4, Jason R. Guertin, PhD5;
1Université Laval/CHU-QUEBEC, Postdoctorat, Quebec, QC, Canada, 2Aix-Marseille University, Associate Professor, Marseille, France, 3Centre Recherche CHU QUEBEC, Quebec, QC, Canada, 4Université Laval, Quebec, QC, Canada, 5Université Laval/CHU-QUEBEC, Associate Professor, Quebec, QC, Canada
1Université Laval/CHU-QUEBEC, Postdoctorat, Quebec, QC, Canada, 2Aix-Marseille University, Associate Professor, Marseille, France, 3Centre Recherche CHU QUEBEC, Quebec, QC, Canada, 4Université Laval, Quebec, QC, Canada, 5Université Laval/CHU-QUEBEC, Associate Professor, Quebec, QC, Canada
OBJECTIVES: Economic Evaluation (EE) is increasingly used to inform the decision of various health care systems about which health care interventions to fund from available resources. Generally, cost-effectiveness analysis are performed with Microsoft Excel (ME). Now the trend is using softwares that can improve decision model and that can resolve more complex problems, methods, and data, as well as improved reproducibility and transparency of results for decision making in collectively-funded health care systems. Our objective in this tutorial is to provide a step-by-step guide on how to implement a decision model of EE, namely a Markov model, in R, an open-source mathematical and statistical programming language
METHODS: The method focuses on time-independent transition, where transition probabilities among health states remain constant over time. We used a previously published decision model made with ME, to illustrate calcul of costs and results, conduct a cost-effectiveness analysis, including a probabilistic sensitivity analysis with R.
RESULTS: BT+CT versus CT alone were €1895.65 vs €3055.20 and 2.03 QALYs vs 1.23 QALYs, respectively. Consequently, the ICUR equalled -€1651.5/QALY, which demonstrates that, although the initial costs of BT are higher than those of CT, the reduced follow-up costs associated with the long-term efficacy of BT make it the most effective and economically dominant option at 1, 5 and 10 years. Sensitivity analyses show that 100% of Monte Carlo iterations are below the willingness-to-pay threshold of €30,000/QALY, making BT+CT an efficient strategy that could be adopted and reimbursed.
CONCLUSIONS: We hope that this tutorial can facilitate wider adoption of R to do decision models, providing two different open-access code for this case study, and Health Economics R Packages.
METHODS: The method focuses on time-independent transition, where transition probabilities among health states remain constant over time. We used a previously published decision model made with ME, to illustrate calcul of costs and results, conduct a cost-effectiveness analysis, including a probabilistic sensitivity analysis with R.
RESULTS: BT+CT versus CT alone were €1895.65 vs €3055.20 and 2.03 QALYs vs 1.23 QALYs, respectively. Consequently, the ICUR equalled -€1651.5/QALY, which demonstrates that, although the initial costs of BT are higher than those of CT, the reduced follow-up costs associated with the long-term efficacy of BT make it the most effective and economically dominant option at 1, 5 and 10 years. Sensitivity analyses show that 100% of Monte Carlo iterations are below the willingness-to-pay threshold of €30,000/QALY, making BT+CT an efficient strategy that could be adopted and reimbursed.
CONCLUSIONS: We hope that this tutorial can facilitate wider adoption of R to do decision models, providing two different open-access code for this case study, and Health Economics R Packages.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
EE479
Topic
Economic Evaluation
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Reproductive & Sexual Health