Society for Medical Decision Making

Description and Objectives:

This course will teach participants how to implement discrete-time microsimulation models in R. We will first provide a conceptual overview of a microsimulation model and the general structure for its implementation. This will be followed by a brief review of relevant R commands and concepts, including data structures, creating variables and functions, sampling random numbers, and basic numerical manipulations. We will then engage in hands-on programming exercises to implement a basic microsimulation, followed by a model of incrementally increasing complexity. By the end of the course, participants will have implemented a microsimulation model with baseline patient heterogeneity, probabilities dependent on the time since the start of the simulation as well as probabilities depending on state-residence. We will also cover the necessary steps to implement probabilistic sensitivity analysis (PSA) of a microsimulation model and shortly cover methods for visualizing and analyzing the output of microsimulation models. Throughout the course, we will highlight good programming principles. 

At the end of the course, participants will be able to:

• Construct discrete-time microsimulation models in R with any of the following elements:

• Population heterogeneity

• Simulation time dependent probabilities

• State-residence dependent probabilities, costs, and/or utilities

• Visualize and analyze microsimulation outputs in R

• Understand computational efficiency considerations in implementing a microsimulation

• Understand the microsimulation model structure required to perform a probabilistic sensitivity analysis (PSA)

• Appreciate the advantages and challenges of using R in decision modeling

All the material of this short course will be provided to participants after the course for future use.