Microsimulation

In previous chapters, the author used Markov models to estimate the burden of disease and the potential impact of interventions. One of the key limitations to Markov models is that they don’t take into account a person’s unique individual characteristics: Markov models are designed to efficiently simulate the average outcome for an entire population. For many public health and healthcare system problems, however, we need to consider heterogeneity within a population, or differences in risk and differences in benefit from our programs. For that purpose, microsimulation models, which take into account unique characteristics of individuals and the correlations between these characteristics, can be more useful. This chapter details the construction and use of microsimulation models, using examples related to diabetes prevention and treatment.