The gap between evidence discovery and actual causal relationships

Area of Science:

• Epidemiology
• Philosophy of Biology
• Causal Inference

Background:

• Causation in epidemiology is often discussed separately from general biological causation.
• A comprehensive understanding of causation requires integrating concepts of mechanism and difference-making.

Purpose of the Study:

• To situate epidemiological causation within a broader biological framework.
• To explore the interplay between mechanism and difference-making in causal discovery.
• To reconceptualize existing epidemiological distinctions in light of this integrated framework.

Main Methods:

• Conceptual analysis integrating biological and epidemiological theories of causation.
• Discussion of evidence types (qualitative/quantitative) for mechanism and difference-making.
• Statistical analysis of multiple causation and its epidemiological implications.

Main Results:

• Causal relationships are defined by mechanisms that make a difference.
• Both mechanism and difference-making provide complementary evidence for causation.
• Multiple causation in biology can be statistically analyzed in epidemiology, revealing regular co-variations.
• Existing epidemiological concepts like "causes of incidence" and "causes of cases" may need re-evaluation.

Conclusions:

• A unified concept of causation, integrating mechanism and difference-making, enhances epidemiological understanding.
• Epidemiological analysis can handle multiple and seemingly stochastic causation.
• The framework highlights potential for "epidemiological dark matter" – unobserved causal factors.