OBJECTIVES: Markov models are a currently popular means of estimating the cost-effectiveness of interventions; however they are associated with certain limitations which may make them ill-suited to inform some health-care decisions.  Patient simulation models offer an alternative methodology which may overcome some of these limitations potentially providing more accurate estimates of the cost-effectiveness. This study aims to explore the strengths and limitations associated with simulation modelling, and the appropriateness of this methodology in ophthalmology. METHODS: An ophthalmology model that was previously developed using a Markov structure was adapted to a patient simulation model using the same cost and quality of life inputs, and clinical inputs adapted to the appropriate format. The deterministic and probabilistic results of each model were then compared to each other, with the costs being broken down by health state to identify key areas of differences. Each model’s results were verified against real-world observational data. RESULTS: Results suggest that cost and quality of life outcomes are similar when unilateral disease is considered, with differences in quality of life and costs being seen when bilateral disease is incorporated. The key area of differences in costs is those relating to blindness, as this is a function of both eyes which is not captured well in a Markov model. One-off costs when a patient becomes blind are also not well captured in a Markov model. CONCLUSIONS: Our results suggest that the ability of simulation models to more accurately represent the real-world patient pathway allows a number of aspects of disease progression to be modelled, which cannot be easily done using a Markov structure. In particular, the benefits of a simulation approach can be demonstrated in the modelling of quality of life as a function of visual acuity in both eyes, and capturing the costs relating to blindness, resulting in differences in estimated cost effectiveness.