Rothman diagrams: the geometry of causal inference in epidemiology

Area of Science:

• Epidemiology
• Causal Inference
• Biostatistics

Background:

• Understanding causal relationships in observational studies is crucial for public health.
• Traditional methods for assessing confounding and effect modification can be complex.
• Geometric visualization offers a novel perspective for epidemiological analysis.

Conclusions:

• Geometric approaches provide intuitive insights into causal inference concepts.
• Standardization and collapsibility can be understood through geometric principles.
• Teaching causal inference using geometry before regression models is recommended for clarity.