Cause of Disease and Causal Inference
1. Model of disease cause Ecological Model (Epidemiologic Triangle)
Wheel Model
Disease-Factor Model Socioeconomic Factor Biologic Factor Environmental Factor Disease
Medical and Biologic Factor
( pathogenesis ) Psycho- and behavioral Factor Health care Factor
Fig 2-3 Disease-Factor Model
Causation Web Model The multi-factorial etiology of disease Chain of causation
web of causation
2. Cause of Disease Factors increasing disease incidence rate of population risk factors sufficient cause necessary cause
3. Conditions of disease development Causal Factors Biologic, physics and chemo- causal factors
Host Genetical, immune condition, age, gender, race, character, psychological and behavioral factors
Environment Natural and social environment
4. Epidemiological methods for etiology studies Based on characteristics: Observational study descriptive study analytic study: case-control & cohort study
Experimental study
Based on time: Retrospective study and prospective study Case report Ecological study Cross-sectional study Case-control study Cohort study Interventional study
5. Deriving inferences: from association to causation The multi-factorial etiology of disease To judge whether an association is causal No association False association Association exist: RR or OR is statistically significant Causal association
6. Criteria of causal inference Time sequence of association: Did the exposure occur before the disease began? Strength of association: Is the rate of disease many times greater among the exposed population, or is the disease excess slight? Dose-response relationship: Does the association show a dose-response effect? Consistency (replication) of association : Has the association been noted consistently across many studies in different people, places, circumstances and times?
Biologic plausibility of association : Does a pathophysiologic model of how the exposure could cause the disease make sense? Is the association supported by generally known facts about the natural history and biology of the disease?
Experimental evidence: Has the exposure caused this disease in an animal model? If there are no other epidemiologic or experimental studies of this exposuredisease relationship, has a causal association