Modeling of Infectious Diseases: A Core Research Topic for the Next Hundred Years

Abstract

Incidence of infectious diseases is an under-researched topic in regional science. This situation is unfortunate because the occurrence of these types of diseases frequently has far-reaching welfare impacts at household, regional, national, and even international levels. Given its welfare impacts and soaring incidence, inter alia, because of climate change, increasing population density, higher mobility, and increasing immunity to several common medicines, the occurrence and spread of infectious diseases should become a regular research topic in regional science. There are also methodological reasons why regional scientists should pay (more) attention to the incidence of infectious diseases. Although both regional science and epidemiology deal with the spatial distributions of their research topics and apply spatial analytical techniques, important methodological differences between them open possibilities for cross-fertilization. This study presents an overview of the main models and estimators of infectious disease incidence. We first discuss maximum likelihood (ML), which is the most common estimator. It is unbiased but imprecise and unreliable for small regions. Next we discuss several methods that have been proposed to improve ML estimation by smoothing (i.e., Bayesian smoothing techniques and nonparametric estimators). From the review, we conclude that none of the models used so far adequately considers the most basic characteristic of infectious diseases, namely, spatial spillover. We argue that the development and application of infectious disease models that allow for spatial spillover is a core research topic for the years to come. We conclude the chapter with suggestions for future regional science research themes in the area of infectious diseases.