HTA Glossary
Markov model
A Markov model is a type of economic model that represents a disease as a set of mutually exclusive health states and moves a group of patients between them in fixed time steps called cycles. The costs and health outcomes attached to each state are added up over the time horizon of the model.
By HTA Course · Last reviewed
In more detail
Markov model explained
Movement between states is governed by transition probabilities. A simple cancer model might have three states: progression-free, progressed disease and dead. Dead is an absorbing state, because patients who enter it cannot leave.
The defining feature is the Markov assumption: the chance of moving to another state depends only on the current state, not on how or when the patient arrived there. Modellers work around this lack of memory with tunnel states and with probabilities that change over time.
Markov models suit chronic conditions in which events can recur and risks continue for many years. Alternatives include decision trees for short-term problems, partitioned survival models, which are common in oncology, and simulations of individual patients.