Association And Causation Flashcards
1
Q
What 3 things should be considered before making a causal link?
A
Bias
Confounding
Chance
2
Q
How can chance be measured in an investigation?
A
Work out p value. If less than 0.05 not due to chance
3
Q
What is a 95% confidence interval?
A
The range in which 95% of the time, the true value would lie in
4
Q
What is bias?
A
A systematic error leading to incorrect estimate due to problem in design or execution
5
Q
What are the 2 types of bias?
A
Selection = reason why some people were chosen and others weren’t Measurement = recall bias, inaccuracy in measuring results
6
Q
What is confounding?
A
A factor that is linked to the outcome and the variable being studied
7
Q
Which factors are considered when establishing a causal link ?
A
BRADFORD HILL 1965
- STRENGTH of association, stronger correlation = more likely to be causal
- CONSISTENCY = similar results in different populations
- SPECIFICITY = exposure increases risk of one disease and not others
- TEMPORAL = THIS IS NECESSARY!! Risk factor must come before disease
- DOSE RESPONSE = increasing amount of exposure increases risk of disease
- PLAUSIBILITY = consistent with other knowledge
- EXPERIMENTAL EVIDENCE
8
Q
Which 2 factors are not necessary for a causal link to be established?
A
- COHERENCE = does not conflict what is already known
- ANALOGY
