Abstract
Markov model allows medical prognosis to be modeled with health state transitions over time and are particularly useful for decisions regarding diseases where uncertain events and outcomes may occur. To provide sufficient detail for operations researchers to carry out a Markov analysis, we present a detailed example of a Markov model with five health states with monthly transitions with stationary transition probabilities between states to model the cost and effectiveness of two treatments for advanced cervical cancer. A different approach uses survival curves to directly model the fraction of patients in each state at each time period without the Markov property. We use this alternative method to analyze the cervical cancer case and compare the Markov and non-Markov approaches. These models provide useful insights about both the effectiveness of treatments and the associated costs for healthcare decision makers.