Epidemiology is the study of the distribution and determinants of health-related states in populations. It aims to establish causal relationships between exposures and outcomes. However, causal inference is challenging and requires careful consideration of criteria like temporality, biological gradient, plausibility, coherence and experimentation. While randomized controlled trials are ideal, epidemiology relies on observational data and must address validity issues like confounding, bias and error. Ultimately, epidemiologists must make reasoned judgments about causation based on the totality of evidence available.

1. Causal Inference
Giridhara
R
Babu

2. +
What Is Epidemiology?
Studying the distribution & determinants of health-related states in
specific populations
 Usually human, but veterinary too
 Diseases or Health conditions

3. +
Causation
• Rooster crows at the break of dawn
• then the rooster caused the sun to rise?

4. +
Epidemiologic Thinking
 A study showed that 30% of drivers in accidents after 8 p.m.
have been drinking alcohol.
 Is alcohol a risk indicator/risk factor/cause of automobile
accidents?

5. The iceberg phenomenon describe a situation in which a large percentage of a problem is
subclinical, unreported, or otherwise hidden from view. Thus, only the "tip of the iceberg" is
apparent to the epidemiologist.

6. +
Hill’s Criteria
1. Strength of the association
• Large associations are more likely to be causal – Not True
• Small associations can be causal
2. Consistency
• Different investigators using different methodologies in different
populations are all seeing similar results:
• Absence of consistency does not preclude causation
3. Specificity
• A cause should lead to a single effect, and vice versa

7. +
Hill’s Criteria
4. Temporality
• Cause must precede the disease
5. Biological Gradient
• strength increases as exposure level increases
• but could be a “threshold effect”; could be curvilinear
relationship; could be inability to accurately ascertain
exposure level

8. +Hill’s Criteria
6. Plausibility :
• should be existing biologic or social mechanistic model to explain the association
• but could just be beyond our biologic knowledge at this point in time; may require
interdisciplinary research
7. Coherence (consonance with existing knowledge)
• Cause-effect interpretation should not conflict with known facts about the natural
history of the disease (e.g., temporal pattern, histopathology, animal findings)
• Lack of such evidence doesn’t nullify the epidemiologic observations (e.g.,
species)

9. +
Hill’s Criteria
8. Experiment:
 Well designed and well conducted?
 Infeasible and/or unethical
9. Analogy
 Analogies or similarities between the observed association and
other associations
 Depends on depth of knowledge at a given time point

10. +
Epidemiological inferences
“ All scientific work is incomplete - whether it be observational or
experimental. All scientific work is liable to be upset or modified by
advancing knowledge.
This does not confer upon us a freedom to ignore the knowledge we
already have, or to postpone the action that it appears to demand at a
given time. Who knows, asked Robert Browning, but that the world might
end tonight? True, but on available evidence most of us make ready to
commute on the 8.30 next day.”
(A. Bradford Hill, 1965)

11. +
Causal inference
Etiologic inference must face numerous validity problems such
as
 Confounding
 Selection bias, and
 Measurement error / Information bias
 Random error