Efficient heterogeneous sampling for stochastic simulation with an illustration in healthcare applications

In modeling disease transmission, contacts are assumed to have different infection rates. A proper simulation must model the heterogeneity in the transmission rates. In this paper, we present a computationally efficient algorithm that can be applied to a population with heterogeneous transmission rates. We conducted a simulation study to show that the algorithm is more efficient than other algorithms for sampling the disease transmission in a subset of the heterogeneous population. We use a valid stochastic model of pandemic influenza to illustrate the algorithm and to estimate the overall infection attack rates of influenza A (H1N1) in a Canadian city. Copyright © 2017 Taylor & Francis Group, LLC.