Markov modeling in R: Advanced method using a cost-effectiveness analysis

A Markov model is the kind of state transition model most widely used in Health Economic Evaluation (HEE) to analyse the efficiency of an intervention and support decision-making. However, it is has transition probabilities which remain constant over time, which limits its use for chronic diseases. Thus, to allow transition probabilities, rewards, or both, to vary over time, we use two types of methods: The first is "Implementing Time Dependency into Markov Transition Probabilities", which allows probabilities to vary over time as measured from the start of the simulation. The second is "Relaxing the Markov Assumption", by adding additional health states to the model, called tunnel states. In this tutorial, a case study on breast cancer is used to illustrate how to implement time dependence in a Markov model and how to conduct analyses of cost-effectiveness, probabilistic sensitivity, and the value of perfect information analyses with R.