Reflection on modern methods: constructing directed acyclic graphs (DAGs) with domain experts for health services...

Area of Science:

• Causal inference and graphical modeling in health research.

Background:

• Directed acyclic graphs (DAGs) visually represent causal assumptions but are underutilized in applied health research.
• Limited practical guidance exists for constructing DAGs, particularly from domain expert input.

Purpose of the Study:

• To provide practical guidance on developing DAGs using domain expert knowledge.
• To address the gap in creating DAGs when published literature is limited.

Main Methods:

• A stakeholder workshop was conducted involving patients, healthcare professionals, researchers, and industry representatives.
• The workshop focused on constructing DAGs for a complex intervention: online consultation within the NHS.

Main Results:

• Lessons were drawn from the workshop to inform the development of DAGs from expert input.
• Initial practical guidance for engaging domain experts in DAG construction is presented.

Conclusions:

• Engaging domain experts is crucial for developing robust DAGs, especially for complex health interventions.
• This approach can enhance the relevance and impact of health research findings on policy and practice.