Methods to madness: The utility of complex systems science in a mad, mad world

Existing methods in social [oral] epidemiology primarily utilize statistical models that assume static characteristics of individuals and environments. While useful, an over reliance on these methods in the social and behavioural sciences can unnecessarily limit perspective and progress as even the most advanced statistical methods cannot capture complex behaviour over time given that systems evolve, environments respond, and behaviours and beliefs crystalize or deteriorate based on a variety of social, environmental and access variables. The recent consensus statement on Future Directions for the Behavioral and Social Sciences in Oral Health acknowledges that dental, oral and craniofacial health emerge from the complex interplay of multiple factors at multiple levels over time and highlights the need for the incorporation of new and underutilized methodologies. Complex Systems Science offers a suite of tools and methodologies that are responsive to the generative mechanisms and processes that underlie population distributions of oral health and disease. Specifically, they assume intricate, dynamic interactions between individuals and groups, they facilitate the study and synthesis of interconnections between people (e.g. patients, healthcare providers and policy makers), how these change over time, any differences across settings, and provide an opportunity to guide future longitudinal data collection and intervention science more effectively. This paper aims to provide an introduction to foundational principles of complex systems, complex systems thinking, and methods found in complex systems science, including social network analysis, system dynamics models and agent-based models, and offers perspectives on the challenges faced and opportunities afforded in the incorporation of these methods into the population oral health sciences.