Abstract for: What Counts as Evidence for Causal Loop Diagram Validation in Public Health?

Public health research has increasingly embraced systems thinking, particularly through the use of causal loop diagrams (CLDs). This trend has produced a variety of insightful diagrams that capture interactions across multiple levels (social and biological) of health determinants. However, CLDs in public health often prioritize depicting large system structure interpret various sources of evidence according to the framework of evidence-based medicine. In this paper, it is argued that CLDs analyzing public health problems with the aim to assist policymaking can, and should, integrate multiple dimensions of evidence - patterns of systems behavior, system dynamics, and mechanisms - using a variety of methods. System dynamics represent bio-social mechanisms unfolding over time. Rather than ranking evidence hierarchically for each link in the CLD, these mechanisms may be better evaluated through an Evidential Pluralist framework. The integration of patterns and dynamics reflects the methodological roots of CLDs in System Dynamics, where system structure is seen as driving behavior. Integrating these dimensions yields important understanding, testable predictions and entry points for intervention leverage. By combining these dimensions of evidence, this paper aims to show that CLDs can generate insights that no single dimension of evidence can provide in isolation.