Lecture 13 - Association, Causal Inference & Causality Flashcards
What is a cause?
A condition that is required for the occurrence of an outcome
List the types of association and define each.
Artifactual/False:
- no true association
- can come from bias and confounding
Non-causal:
- disease may cause exposure
- disease and exposure may be associated with a third factor (confounding)
Causal
-exposure cause outcome
List the types of causal relationships and define each.
Sufficient:
-if cause is present, outcome will always occur
Necessary
-cause is required for outcome to occur but does not always lead to disease
Component cause:
-cause that makes disease more likely (risk factor) but is not required for disease to occur
List the types of interaction and define each.
Synergism:
-the combined measure of effect is greater than the sum of the individual effects
Parallelism:
-when measure of effect is great if either component cause is present
Multiple __________ causes can work together to become a __________ cause.
Component; sufficient
What are Hill’s criteria used for and what are the criteria.
Used to help determine if an observed association can be considered causation.
- strength
- consistency
- temporality
- biological gradient
- plausibility
Define strength as it relates to Hill’s criteria.
The further the measure of association is from one, the more likely the association might be causal
Define consistency as it relates to Hill’s Criteria.
Repeated observations of the same association under different circumstances increases its likelihood of being causal
Define temporality as it relates to HIll’s Criteria.
The proposed cause must precede the outcome
Define biological gradient as it relates to Hill’s Criteria.
Increase exposure to suspected cause should demonstrate and increased risk of outcome
Define plausibility as it relates to Hill’s Criteria.
There should be a biologically feasible mechanism through which explains how the cause triggers the outcome