Project INSIGHT: Developing Novel Epidemiological Models with Human Behavior, Interaction, and Collaboration

By Folashade Agusto, Carmen Caiseda, Igor Erovenko, Brian L. Levy, and Padmanabhan Seshaiyer 

Project INSIGHT (INclusion of challenges from Social Isolation Governed by Human behavior through Transformative research in epidemiological modeling) is an initiative to develop novel and transformative research that connects the social, behavioral, and economic interactions of humans with mathematics-based epidemiological models. Four diverse institutions of higher education (IHEs) are collaborating on the project and will each work to incorporate social, behavioral, and economic processes within math modeling frameworks to ultimately improve our understanding of the COVID-19 pandemic and its effects on society. Project INSIGHT will also train a new generation of U.S. students for careers at the interface of the mathematical, social, and biobehavioral sciences.

This effort evolved from a common vision based on two successful National Science Foundation (NSF) grants for Rapid Response Research (DMS 2028297 and 2031029), as well as an NSF-funded workshop (DMS 1839608) during which researchers from the mathematical, life, and social sciences defined emerging research challenges and priorities. During the current project, individuals at the University of Kansas, George Mason University, Inter American University of Puerto Rico, and University of North Carolina at Greensboro are attempting to integrate best practices in research and education. These participants have experience with a diverse range of disciplines, including data science, demography, disease ecology, epidemiology, game theory, mathematical modeling, and sociology.

Conceptual Research Questions

People typically value social connection because it deters isolation, which is associated with negative outcomes like substance use and abuse, domestic violence, and reduced mental and physical health [3]. Many of these outcomes are conditioned on the socioeconomic circumstances of individuals or communities, with elevated risks for rural, low-income, and older groups (see Figure 1). Regardless, isolation was a key tactic to curb COVID-19 transmission during the early stages of the pandemic. 

Project INSIGHT asks two sets of questions about behavioral responses to social isolation. First, we will investigate compliance with isolation policies and corresponding COVID-19 outcomes. Do pre-pandemic demographic and socioeconomic contexts of U.S. counties predict their compliance with social distancing policies? Did social distancing effectively curb COVID-19 transmission and deaths, and was its effectiveness moderated by socioeconomic contexts? Second, we will investigate whether social isolation led to unanticipated negative outcomes such as substance use, overdose, or domestic violence. Did these outcomes vary with isolation duration, and was this prospective association moderated by socioeconomic contexts? Does a certain duration of social isolation reduce COVID-19 morbidity while also minimizing harm?

Figure 1. Conceptual model of the interactions between social context, COVID-19 policies, isolation, and salient public health outcomes. Figure courtesy of Brian L. Levy and Jarron M. Saint Onge.

Methodology

We first identified several potential behavior responses: (i) Compliance or non-compliance with isolation policies, (ii) adoption or non-adoption of risky behaviors—related to the lack of access and adherence to treatment regiments—that promote substance use disorder or opioid misuse, and (iii) intimate partner domestic violence. Figure 1 illustrates these responses.

We intend to develop realistic epidemic models that incorporate behavioral responses of compliance and adherence for the following categories:

1. Isolation compliance: We will use classical segmentation to divide the populations into compliant and non-compliant groups, then define a switching function to account for changes in behavior [6]. We will employ games with appropriate payoff functions that inform individuals’ behavioral choices [2].
2. Opioid misuse and treatment adherence: We aim to create a model that investigates the severity of substance use disorder via game-derived utility functions that implement treatment adherence. The model will also incorporate drug-seeking behavior of affected individuals, and well-defined behavioral functions will account for model parameters like recovery and deaths [1, 7].
3. Domestic violence: We plan to focus on intimate partners and use economic-dependent functions to determine choices such as abuse, forgiveness, seeking help, or leaving the domestic violence cycle [4, 5]. 

The definition of a behavioral-dependent switching function, a compliance/adherence payoff function, or partners’ decision functions is not straightforward; modeling efforts depend on available data. Project INSIGHT will therefore utilize a wide range of data sources, including agency databases from multiple levels of government as well as novel data on isolation, mobility patterns, and impacts of the COVID-19 pandemic. The definition of these behavioral-related functions will affect the qualitative and quantitative analysis of our models, including sensitivity analysis. We intend to conduct a rigorous mathematical analysis of these models and use them to address our broad research questions in rural and urban geographical contexts.

Proposed Contributions 

Isolation is both an individual behavior and a community condition. As demonstrated by the COVID-19 pandemic, it is also a key public policy option for the reduction of disease spread. However, compliance behaviors that pertain to isolation policies vary considerably. These complexities are critical to accurate epidemiological models of pandemics with broader effects on community wellbeing.

Project INSIGHT will support 16 students over three years (2023 to 2025) in two cohorts of eight students each at partner IHEs. Students will engage in a summer bootcamp as well as six online modules that explore computational tools, modeling approaches, and professional development. We will emphasize foundational concepts in the modeling of human behavioral, social, and economic processes, as well as their impacts during pandemic isolation. As students complete master’s theses with mentors in math or social science, the project’s structure will provide them with the support of a collaborative cohort, ongoing guidance and mentorship, and career training — elements that are all critical to retention in the STEM pipeline. Project INSIGHT principal investigators will also organize student presentations at various venues.

The newly formed INSIGHT collaboration is an opportunity for participants to share and learn from diverse worldviews that cross barriers—including those between discipline silos, demographics, and cultural and socioeconomic differences—to gain fresh insight into a problem that has affected everyone in recent years.

Acknowledgments: The authors want to recognize the valuable contributions of Jonathan T. Rowell, Jarron M. Saint Onge, and Andrew Townsend Peterson: INSIGHT researchers who are experts in the fields of sociology, game theory, and ecology.

References
[1] Agusto, F.B., & Kim, S. (2019). Impact of mobility on methicillin-resistant Staphylococcus aureus among injection drug users. Antibiotics (Basel), 8(2), 81. 
[2] Agusto, F.B., Erovenko, I.V., Fulk, A., Abu-Saymeh, Q., Romero-Alvarez, D., Ponce, J., … Peterson, A.T. (2022). To isolate or not to isolate: The impact of changing behavior on COVID-19 transmission. BMC Public Health, 22(1), 138.
[3] Holt-Lunstad, J., Robles, T., & Sbarra, D.A. (2017). Advancing social connection as a public health priority in the United States. Am. Psychol., 72(6), 517-530. 
[4] Ohajunwa, C., Caiseda, C., & Seshaiyer, P. (2022). Computational modeling, analysis and simulation for lockdown dynamics of COVID-19 and domestic violence. Electron. Res. Arch., 30(7), 2446-2464.
[5] Ohajunwa, C., Caiseda, C., & Seshaiyer, P. (2023). Mathematically modeling the dynamics of COVID-19 and domestic violence during lockdown. SIAM News Online. Retrieved from https://sinews.siam.org/Details-Page/mathematically-modeling-the-dynamics-of-covid-19-and-domestic-violence-during-lockdown.
[6] Ohajunwa, C., Kumar, K., & Seshaiyer, P. (2020). Mathematical modeling, analysis, and simulation of the COVID-19 pandemic with explicit and implicit behavioral changes. Comput. Math. Biophys., 8(1), 216-232.
[7] Wagner, R., & Agusto, F.B. (2018). Transmission dynamics for methicillin-resistant Staphylococcus aureus with injection drug user. BMC Infect. Dis., 18, 69. 

	Folashade Agusto is an associate professor at the University of Kansas. She is the principal investigator for Project INSIGHT and a project leader in the Mathematical Sciences Research Institute’s 2023 African Diaspora Joint Mathematics Workshop. Agusto is also a member of the SIAM Diversity Advisory Committee and SIAM Student Paper Prize Selection Committee.
	Carmen Caiseda is a professor and coordinator of the mathematics group at the Inter American University of Puerto Rico (IAUPR) – Bayamon. She is co-principal investigator of the Data Science at IAUPR project, which is building a data science community of practice that impacts faculty, students, and professionals. As an undergraduate research mentor, Caiseda engages students in STEM via mathematical models of real-world challenges with sociocultural contexts.
	Igor Erovenko is an associate professor of mathematics at the University of North Carolina at Greensboro. His research specializes on game-theoretic models in biology, including analysis of social dilemmas in structured populations and behavioral epidemiology.
	Brian L. Levy is an assistant professor of sociology at George Mason University. His research analyzes neighborhood effects on wellbeing in the U.S., including the ways in which everyday mobility patterns between neighborhoods impact neighborhood vitality.
	Padmanabhan Seshaiyer is a professor of mathematical sciences at George Mason University who previously served as chair of the SIAM Diversity Advisory Committee. He works in the broad area of computational mathematics, mathematical biology, data science, biomechanics, design thinking, and STEM education. Seshaiyer is also chair of the U.S. National Academies Commission on Mathematics Instruction and Associate Director for Applied Mathematics at the Math Alliance.