CANCELED: Microsimulation Modeling and Bayesian Model Calibration

Public resources are limited, and policymakers are under increased pressure to use these resources as efficiently as possible. Individual-level simulation, or microsimulation, is a tool for estimating long-term effectiveness and costs of interventions to quantify cost-benefit tradeoffs and ultimately support more informed decision making. Microsimulation models describe disease progression and the action of intervention on disease and can incorporate relationships between clinical history and future events. Before application, models are calibrated so that they accurately predict a set of observed targets. Calibration is a key part of model building. Microsimulation models have been applied to a wide range of health policy questions, including optimal cancer screening and treatment guidelines; technology reimbursement and coverage decisions; and hospital operations management. This course will provide an overview of discrete event simulation (DES) models, which stochastically simulate individual trajectories in continuous time, the general structure for their implementation in the R programming language, and Bayesian model calibration methods. Tradeoffs between different calibration methods will be discussed. This course will involve hands-on programming exercises in R using code templates, and is intended for an audience interested in creating individual-level DES models and using them to conduct policy or clinical analysis.