Socio-Behavioral Influences in Epidemic Modeling: Towards a Unified Framework

The COVID-19 pandemic has underscored the need for behavior-based epidemic models in guiding public health policies. While current works in behavior-based epidemic models offer valuable insights, their broader applicability is often limited by simplistic assumptions about human behavior and their dependence on certain specific types of information and datasets. This paper proposes a generalizable behavioral model that can represent and capture various behaviors and their influences on disease spread, which is crucial for building a unified behavioral-epidemic framework. In addition, by decomposing the behavioral models into sub-models we show that this approach can be efficiently scaled and adapted to different contexts. To evaluate our approach, we modeled COVID-19 outbreaks in two sets of U.S. counties. Our initial findings show that incorporating even a limited set of key behavioral and socio-cultural factors leads to more accurate predictions. These promising initial results indicate the feasibility towards developing a unified socio-behavioral influence driven epidemic modeling framework.