Designing a robust and cost-effective screening strategy for Diabetic Retinopathy using simulation coupled with Design of Experiments

This paper demonstrates the use of Design of Experiments (DoE) coupled with simulation to design a robust and cost-effective intervention in a systematic way. DoE provides a simple and intuitive method to find the optimum settings of the input factors for the most cost-effective intervention. Moreover, DoE addresses the critique that health care simulations unrealistically assume ideal conditions in simulating the impact of a particular intervention by utilizing robust parameter design to find settings that make the intervention robust to variations, such as uncertainty in some input factors. Both the determination of the optimum input factor settings for the intervention as well as robust parameter design for interventions would be demonstrated here, using the design of Diabetic Retinopathy (DR) screening strategy as illustration. A Markov simulation with elements of screening strategy as input factors and cost as the response take the place of traditional experiments.