Lec 12. Association, casual inference and causality Flashcards
What is the cause definition?
A precursor event, condition, or characteristic required for the occurrence of the disease or outcome
What does association mean again?
Relationships between exposure/treatment and an outcome/disease.
What are the 3 types of association (relationships)?
Artifactual associations, Non-causal associations, and causal associations.
What are artifactual associations?
Flat out wrong. Can arise from Bias and/or confounding
What are non-casual associations?
Most of what we talked about recently.
In what two ways do non-casual associations occur?
- Disease may cause exposure, rather than exposure causing disease. 2. Disease and exposure both are related with a confounder
What is a mediator?
Necessary step to get the outcome
Can Confounders be mediators?
No
What are three types of casual relationships?
- Sufficient cause. 2. Necessary cause. 3. Component cause (risk factor)
Define sufficient cause.
Any precursor by itself will cause disease.
What is an example of sufficient cause?
genetic defect in a person
Define necessary cause.
just because its there doesn’t mean someone is going to get the disease, but it is necessary. By itself not guaranteed.
What is an example of necessary cause?
Mycobacterium tuberculosis, necessary cause for TB to be diagnosed, yet can be present in ppl without clinical symptoms.
Define Component cause (risk factor).
Component that increases likelihood of disease.
What are two interactions in causal research?
Synergism and parallelism
What is an example of component cause?
Smoking leads to lung cancer
Define synergism.
2 variables that make the relationship and association more likely together than any one by themselves.
Define parallelism.
Risk factors that are either or present, can change the association.
What are Hill’s criteria guidelines?
Guidelines to see if the observed association is a verdict of causation.
What are the 5 Hills Criteria?
Strength, Consistency, Temporality, Biological Gradient, and plausibility.
Define strength.
Refers to the size of association (RR/OR/HR). The stronger/bigger the number the more likely to be causal.
True or False? A strong association is neither necessary nor sufficient for causality and weakness of an association is neither necessary nor sufficient for absence of causality.
True
Define consistency
Reproducibility, repeated observations. How consistent have other studies shown this relationship.
Consistency may do what to the truth?
Obscure it.
