Abstract
BackgroundDirected acyclic graphs (DAGs) are popular tools for identifying appropriate adjustment strategies for epidemiological analysis. However, a lack of direction on how to build them is problematic. As a solution, we propose using a combination of evidence synthesis strategies and causal inference principles to integrate the DAG-building exercise within the review stages of research projects. We demonstrate this idea by introducing a novel protocol: ‘Evidence Synthesis for Constructing Directed Acyclic Graphs’ (ESC-DAGs)’.MethodsESC-DAGs operates on empirical studies identified by a literature search, ideally a novel systematic review or review of systematic reviews. It involves three key stages: (i) the conclusions of each study are ‘mapped’ into a DAG; (ii) the causal structures in these DAGs are systematically assessed using several causal inference principles and are corrected accordingly; (iii) the resulting DAGs are then synthesised into one or more ‘integrated DAGs’. This demonstration article didactically applies ESC-DAGs to the literature on parental influences on offspring alcohol use during adolescence.ConclusionsESC-DAGs is a practical, systematic and transparent approach for developing DAGs from background knowledge. These DAGs can then direct primary data analysis and DAG-based sensitivity analysis. ESC-DAGs has a modular design to allow researchers who are experienced DAG users to both use and improve upon the approach. It is also accessible to researchers with limited experience of DAGs or evidence synthesis.
Highlights
Directed acyclic graphs (DAGs) are popular tools for identifying appropriate adjustment strategies for epidemiological analysis
We present a novel method—Evidence Synthesis for Constructing Directed Acyclic Graphs (ESC-DAGs)—to answer the call for a systematic approach to building DAGs
Perhaps the most obvious strength of ESC-DAGs is how the resulting integrated DAGs’ (I-DAGs) closely align with the fundamental purpose of DAG-based analysis—to identify appropriate adjustment strategies.[3,4]
Summary
Causal inference methods are popular in observational research, with directed acyclic graphs (DAGs) being notably prominent.[1]. Each directed edge in the IG is assessed for three causal criteria: temporality; face-validity; and recourse to theory They are primarily informed by the classic Bradford Hill viewpoints,[24] and are compatible with the ‘inference to the best explanation’ approach advocated by Krieger and Davey Smith.[1] If a relationship is determined to possess each criterion, a counterfactual thought experiment derived from the POF is used to further explicate the reviewers’ assumptions.[25] The translation process combines ‘classic’ and ‘modern’ causal thinking and understands DAGs as ‘conceptual tools’[1] for exploring causation, rather than substitutes for careful causal thinking. Each directed edge from both studies was entered into a new diagram This DAG was saturated, and all ‘new’ relationships were put through the translation process (for example the relationship between family structure and adolescent sex was rejected whereas age of alcohol initiation was hypothesised to cause other substance use). If they feature different directed edges, there is less support for recombining them
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
