From Association to Causation
Which study design possesses the strongest evidence of causality?
1. Cross sectional
2. Cohort- the best epidemiological study type
3. Case control
4. Randomized clinical trial- we manipulate
And there was the joke that he told in first year...laaaaame
Ethological fallacy- when a correlation doesn't make sense
What is Cause?
• Terminology- etiology, pathogenesis, ect.
• Physician's use of "cause"
○ Prevention- the blood pressure cuff in the pharmacy in order to monitor blood pressure for
○ Diagnosis- going to the doctors to deal with the problem
○ Treatment- taking aspirin to treat hypertension. The aspirin caused the change in the
patient's profile to something favourable
• Physician versus epidemiologist
○ Epis are more careful to use cause, always trying to figure out the cause
○ Physicians know the causes of the diseases that they know
Single and Multiple Causes
• Koch's Postulates of Causation- 1882
1. Every disease has an organism
2. Must be able to isolate and grow the organism
3. Upon inoculation, a susceptible host must get disease
4. Organism must be retrieved and infected host
○ Example: HIV and AIDS
Single cause singledisease
Web of causation (multiplecause, one disease)
Single cause-multiple disease
Non-applicable in today's medicine except..
Proximity of Cause to Effect
*See picture in phone*
Interplay of Multiple Causes
• Interaction- how two di