Markov cohort models are commonly used in decision analysis to model the effect of different policies and interventions on population health, but these models are implicitly based on the probability distributions of a single individual transitioning through different health states. I derive theoretical model that explicitly extends the commonly-used Markov model framework to a discrete population with many individuals, including the derivation of a density function and the moments this stochastic process, and show that cohort model represents the average of the derived stochastic process.
Rowan Iskandar is an assistant professor of health service, policy, and practice at Brown University School of Public Health. Dr. Iskandar is interested in developing novel methodologies in, or porting existing methodologies from applied mathematics for application in decision modeling. These topics include the use of stochastic differential equation for decision modeling and the application of approximation theory to computationally expensive simulation models.