Constructing Directed Acyclic Graphs (DAGs) to Inform Tobacco Cessation Intervention Research: A Methodological Extension Using Evidence Synthesis.

BACKGROUND: Tobacco use remains a leading preventable cause of morbidity and mortality in the United States, with persistent disparities in cessation outcomes across socioeconomic and racial groups. While numerous interventions exist, their effectiveness is shaped by complex interrelated factors at individual, social, and healthcare system levels. Identifying and modeling these causal pathways is essential to inform equitable intervention design. METHODS: This study applied the Evidence Synthesis for Constructing Directed Acyclic Graphs (ESC-DAG) protocol to integrate empirical findings from 35 quantitative studies examining barriers and facilitators of tobacco cessation intervention uptake in the United States. Using the Andersen and Aday Health Services Research Model as a guiding framework, we extracted, harmonized, and synthesized significant causal relationships into a unified DAG, distinguishing exposures, outcomes, mediators, and confounders. RESULTS: The integrated DAG revealed that structural factors such as socioeconomic disadvantage, digital inequities, rurality, and cultural barriers exerted substantial influence on cessation outcomes. These upstream determinants operated through mediators including motivation, treatment engagement, and access barriers, while healthcare system factors such as provider engagement and proactive outreach emerged as consistent facilitators. Digital access and culturally tailored interventions were identified as underexplored yet potentially high-impact pathways. DISCUSSION: The ESC-DAG methodology provided a structured approach to visualize and synthesize causal mechanisms beyond traditional review synthesis, highlighting points of intervention at both policy and practice levels. The findings underscore the importance of multi-level strategies, including financial support, digital equity initiatives, provider outreach, and culturally tailored cessation services. CONCLUSIONS: By applying ESC-DAG methodology, this study contributes a novel causal framework for understanding disparities in tobacco cessation intervention uptake. The resulting DAG can inform future statistical modeling, simulation studies, and equity-focused program design, supporting more effective public health strategies to reduce smoking prevalence and associated inequities.