Causal models and causal modelling in obesity: foundations, methods and evidence

Area of Science:

• Epidemiology
• Causal Inference
• Public Health

Background:

• Current obesity research often conflates different meanings of 'cause'.
• This ambiguity leads to confusion and unclear thinking in understanding obesity etiology.
• A foundational discussion on causal concepts is needed for robust scientific inquiry.

Purpose of the Study:

• To define 'cause' in the context of obesity research.
• To explore appropriate causal models and structures for obesity.
• To examine methods for evaluating postulated causes and estimating causal effects.

Main Methods:

• Conceptual analysis of causal definitions.
• Exploration of abstract causal models.
• Discussion of experimental and observational study designs for causal inference.
• Use of analogies from other research areas.

Main Results:

• Different interpretations of causal factors in obesity research are frequently confused.
• A clear framework for defining and evaluating causes is essential.
• Inelegant, rigorously designed studies are most likely to yield informative results.

Conclusions:

• Establishing precise definitions of causality is crucial for advancing obesity research.
• A multi-faceted approach using diverse study designs is necessary to understand obesity's complex causes.
• Clearer causal reasoning will reduce confusion and improve the validity of findings.