Introducing Systems Approaches in Health Behavioral Research

Health behaviors, such as smoking and unhealthy eating, are among the top leading preventable risk factors of non-communicable diseases; still more than 20% of the global population smokes, around 13% is obese, approximately a quarter of the adult population does not fulfill the guidelines for physical activity, and less than a quarter of the population meets the recommendations of fruit and vegetable consumption. The prevalence of unhealthy behaviors is consistently larger among individuals in lower socioeconomic groups as compared to higher socioeconomic groups.
Determinants of health behaviors cannot be fully understood in isolation of the context in which they are shaped and sustained. It is increasingly recognized that individual and environmental determinants interact with and feedback on each other, creating a complex system. Making the pathways and features of such a system explicit is important, because this may help to better identify the etiology of health behaviors and to optimize ways to promote healthy behaviors and prevent unhealthy behaviors. Systems approaches, such as agent-based modeling, are therefore needed.
This thesis aims to explore and quantify the importance of social networks as a determinant of health behaviors, and to investigate the usefulness of agent-based models (ABMs) as a tool for assessing the impact of interventions to reduce socioeconomic inequalities in health behaviors.