Founder effects arising from gathering dynamics systematically bias emerging pathogen surveillance

Bradford P Taylor,William P Hanage

DOI: https://doi.org/10.1101/2022.11.15.22282366

2024-10-04

Abstract:Models of infectious disease transmission have shown the importance of heterogeneous contact networks for epidemiology; the most connected individuals are most likely to be infected early. Yet it is cumbersome to parameterize and incorporate such networks into simple models. We introduce an alternative model framework that explicitly includes attendance at and disease transmission within gatherings of different sizes, which disaggregates sequential epidemics moving from the most to least social subpopulations that underly the overall, single-peaked infection curve. This can systematically bias initial estimates of the growth rate for emerging variants and their severity, if vulnerable populations avoid large gatherings. Finally, we show that how often similarly social individuals preferentially interact (i.e., homophily, or assortative mixing) tunes the magnitude and duration of these biases. Together, we provide a simple framework for incorporating socialization and behavior in epidemic models, which can help contextualize surveillance of emerging infectious agents.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to better understand the impact of social behaviors on epidemic surveillance by improving the classical infectious disease transmission model, especially how to quantify the estimation biases of the initial epidemic growth rate and severity caused by these behaviors. Specifically, the paper introduces a new model framework that explicitly includes the disease transmission dynamics in gatherings of different scales and how these dynamics affect the continuous epidemic process from the most social to the least social sub - populations. This helps to explain why there may be systematic biases in the estimation of the initial epidemic growth rate when large gatherings are avoided in high - risk groups.

The core problems of the paper are:
1. **Impact of social behaviors on epidemic transmission**: Traditional infectious disease models usually assume that the population is evenly mixed, but in reality, there is heterogeneity in people's social behaviors. How does this heterogeneity affect the spread of diseases?
2. **Biases in epidemic surveillance**: Due to the heterogeneity of social behaviors, the transmission speed and pattern may be different among different sub - populations, which may lead to biases in epidemic surveillance data. How can these biases be corrected through the model in order to more accurately assess the transmissibility and severity of emerging variants?
3. **Impact of social structure on epidemic transmission**: How does homogeneity (i.e., similar people are more likely to interact) affect the speed and duration of epidemic transmission? How can this impact be quantified through the model?

By solving these problems, the paper aims to provide a more realistic model framework to help public health experts better understand and predict the transmission dynamics of infectious diseases, thereby formulating more effective prevention and control strategies.