6: Causality Flashcards

1
Q

Define

Cause
Causality
Association

A

Cause
something that brings about an effect or a result

Causality
the realtion between a cause and its effect

Association
any statistical relationship between 2 variables. association does not mean causality

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Deterministic vs probabilistic theory

A

Deterministic cause
A must always be followed by B

Probabilistic cause
A causally increases the risk for B

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Causes can be classified as…

A

multiple factors
in most diseases multiple factors are needed to cause a disease
sufficient
cause always induces disease but there may be other set of causes that can cause disease

necessary
without the cause, the disease never occurs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How to infer causality?

A

First, is there an association?

If yes, that isnt enough:

a)Henle-Koch postulates:
organism must be present in every case of disease, isolated grown in culture, cultured must causes disease in new human, ….

-> limitations

b)multiple factors are needed to cause disease

c) Bradford-Hill criteria

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Bradford-Hill criteria part 1

A

Is there any other way of explaining the facts we study ?

-> 8 criteria

Temporarily
Does the cause precede the effect?

  • first exposure, then effect

Strength of association
- indication for causality if RR/IR/HR far away from 1, otherwise unmeasured factors can be responsible for false effects, if false effects are big they should be obvoius (cofounding)

Dose-response relationship
the higher the exposure, the more probable the event

Replication of findings
repeated observation in different populations, meta analysis

Pitfall: statistical significance does not equal strength of association

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Bradford Hill criteria part 2

A

Biological plausability
postulated association should be based on biological model

Reversibility
Deletion of the cause should lower disease probability

Specificity
specific exposure causes one disease and no other exposure causes this diesease

Analogy
Are there cause-effect relationship for a similar exposure or disease which are already established

How well did you know this?
1
Not at all
2
3
4
5
Perfectly