Comparison of algorithms to simulate disease transmission

Abstract

A complex model to study the spread of influenza often requires efficient algorithms to simulate disease transmission. This article studies the internal mechanisms of existing algorithms. We compare existing algorithms to simulate disease transmission in an effort to identify impact factors and put forth rules for efficient algorithm selection. Specifically, an algorithm from the infectiousness perspective is recommended when both the transmission probabilities and the fraction of infectious individuals are small, or when the transmission probabilities are large but the fraction is either sufficiently small or sufficiently large. In contrast, an algorithm from the susceptible perspective should be adopted in the case of small transmission probabilities but a large fraction of infectious individuals, or large transmission probabilities and a moderate fraction. This investigation not only helps to guide a more-efficient simulation study of disease transmission in practice but also serves as a prerequisite for the development of more-advanced simulation models.

Keywords:

• influenza spread
• simulation models
• algorithms to simulate disease transmission

Acknowledgments

This research was supported by the Research Grants Council Collaborative Research Fund (Ref. CityU8/CRF/12G), Theme-Based Research Scheme (Ref: T32-101/15R) and National Natural Science Foundation of China (Ref: 71420107023).