Modeling Infectious Outbreaks in Non-Homogeneous Populations

Abstract

Emerging diseases, novel strains of reemerging diseases, and bioterrorism threats necessitate the development of computational models that can supply health care providers with tools to facilitate analysis and simulation of the progression of infectious diseases in a population. Most computational models assume homogeneous mixing within populations. However, a more realistic approach to the simulation of infectious disease outbreaks includes the stratification of populations in which the interactions between individuals are affinity-based. To examine the effects of heterogeneous populations on the outbreak dynamics, we developed a hybrid model that includes clustered individuals which represent differentiated populations. This facilitates the study of the effects of distinct behavioral properties on the dynamics of an infectious disease epidemic. Our results indicate that non-uniform interactions and affinity-driven behavior can drastically change the outbreak dynamics in the population.

Modeling immune response and its effect on infectious disease outbreak dynamics

“…For this simulation, we utilized the Global Stochastic Contact Model (GSCM). The GSCM is a computational model that simulates the spread of an infectious disease in a population during an infectious outbreak [ 2 ]. The GSCM simulates the interactions between individuals in the population as the infection progresses.…”

“…The most recent approaches to infectious disease outbreak modeling incorporate non-homogeneous components to the individuals to be modeled. Studies of the effects of non-homogeneous populations on the dynamics of infectious outbreaks have shown the importance of integrating individuals with heterogeneous characteristics [ 2 ]. Although the amount of time a person is capable to transmit the disease varies among individuals, many models set that value to be homogeneous for the population.…”

A method for reducing the severity of epidemics by allocating vaccines according to centrality

“…Several methods for simulating epidemics computationally are based on the SEIR model of epidemics [15,19,16,10]. In the SEIR model, the population is divided into four classes: susceptible individuals, who can become infected; exposed or latent individuals, who have been infected but are not capable of spreading the infection; infectious individuals, who can spread the disease to susceptible individuals; and recovered individuals, who can no longer be infected.…”

“…al. have developed multiple stochastic models of disease spread in a population, which use the standard SEIR(S) model [15,19]. Some of these models, such as their Global Stochastic Cellular Automata model [16], can be adapted to the problem of vaccine allocation.…”

“…Future work will primarily focus on addressing the limitations described above. Methods for simulating non-homogenous populations [19,10] will be incorporated into our model. Alternative centrality measures and more accurate weighting formulas will be investigated.…”