Decision Modeling in R

Economic evaluations often rely on decision analytical models. These models are becoming increasingly more complex to better represent the underlying clinical conditions and to provide more and better insight to decision makers. In addition, advanced statistical and mathematical methodologies are being introduced within the context of decision analytic modelling (e.g., value of information, calibration, evidence synthesis). Current commercially available software sometimes provides limited flexibility to embed such statistical methods within a decision modelling framework. Finally, model transparency may be compromised, given the “black box” nature of some of the existing software. In contrast, R is a free software in which statistical analyses and decision analytical modelling can be combined within a single framework. That facilitates, among others, more appropriate incorporation of parameter and model uncertainty in decision modelling. Finally, the fact that R is freely available may facilitate model transparency and reproducibility.

Petros Pechlivanoglou, PhD, is a Scientist at The Hospital for Sick Children (SickKids) Research Institute and an Assistant Professor at the Institute of Health Policy Management and Evaluation, University of Toronto. His research focuses on the use of decision analysis in health economics; bridging evidence synthesis, administrative data predictive modelling and decision analysis. Petros obtained his MSc in Econometrics and his PhD in Health Econometrics from the University of Groningen, the Netherlands.