4 Association and Causation

Question 1

Q: What does association mean?

Answer 1

A: Association refers to the statistical dependence between two variables

A link, relationship or correlation

Question 2

Q: What 4 factors should be considered when evaluating statistical association? Is there a specific order?

Answer 2

A:  Chance
 Bias
 Confounding
 Cause

MUST consider first three before you look at causal relationship

Question 3

Q: Why does the role of chance need to be assessed when investigating association? How?

Answer 3

A: Most studies based on an estimate from samples rather than whole populations

Performing appropriate statistical significance tests by calculating confidence intervals (p value- the probability that a result could simply be due to chance, threshold is usually <0.05-> i.e. if p<0.05 we can be sure that the result of the study is not due to chance)

Question 4

Q: What are confidence intervals?

Answer 4

A: the range within which the ‘true’ value (e.g. the strength of an association) is expected to lie with a given degree of certainty (e.g. 95% or 99%)

If independent samples are taken repeatedly from the same population, and a confidence interval calculated for each sample, then a certain percentage (e.g. 95%) of the intervals will include the true underlying population parameter

Question 5

Q: What is bias in terms of disease and exposure? Consequence of? Controlled by analysis? Eliminated by increasing sample size?

Answer 5

A: Bias is a systematic error leading to an incorrect estimate of the effect of an exposure on the development of a disease or outcome of interest

A consequence of defects in design or execution of an epidemiological study.

Cannot be controlled in analysis of study.

Cannot be eliminated by increasing sample size.

Question 6

Q: Name and describe the 2 broad types of bias.

Answer 6

A: Selection bias – occurs when there is a systematic difference between the characteristics of the people selected for a study and the characteristics of those who were not (non response bias)

Measurement bias – occurs when measurements or classifications of disease or exposure are inaccurate (recall bias)

Question 7

Q: Describe a potential confounder in relation to investigating disease. Mixing effects between?

Answer 7

A: any factor which is believed to have a real effect on the risk of the disease under investigation and is also related to the risk factor under investigation

Mixing of effects between exposure, the disease and a third factor

Question 8

Q: Give 2 examples of confounding factors.

Answer 8

A: – factors that have a direct causal link with the disease (e.g. smoking and lung cancer)
– factors that are good proxy measures of more direct unknown causes (e.g. age and social class)

Question 9

Q: How is confounding accounted for? (4)

Answer 9

A: Account for confounding using matching, randomisation, stratification and multivariate analysis

Question 10

Q: What are 4 common confounders?

Answer 10

A:  Age
 Sex
 Socio-economic – Poorer people have higher rate of almost all disease – Higher risk of early death in poor people
 Geography – Disease prevalence varies greatly by place

Question 11

Q: What is the judgement of a cause-effect relationship based on?

Answer 11

A: chain of logic that addresses two main areas:

1. Observed association between an exposure and a disease is valid
2. Totality of evidence taken from a number of sources supports a judgement of causality

Question 12

Q: What criteria is used for causation?

Answer 12

A: BRADFORD-HILL CRITERIA FOR CAUSATION includes FACTORS TO CONSIDER

Question 13

Q: What are the 9 suggested criteria for causation? Which is absolutely necessary?

Answer 13

A: Strength
Consistency
Specificity
Temporal relationship - NECESSARY
Dose-response relationship
Plausibility
Experimental evidence
Coherance
Analogy

Question 14

Q: Describe strength as a criteria for causation. How is it measured? Weak or strong more likely? (3)

Answer 14

A: – Strength of association measured by magnitude of relative risk.
– Strong association more likely causal than weak association(likely result of confounding or bias)
– BUT weak does not mean non-causal i.e. lung cancer and smoking

Question 15

Q: Describe consistency as a criteria for causation. Lack of consistency means? (2)

Answer 15

A: – More likely to be causal if similar results in different populations using different study designs- unlikely studies subject to same type of errors.
– A lack of consistency does not exclude a causal association since different exposure levels and other conditions may reduce the impact of the causal factor.

Question 16

Q: Describe specificity as a criteria for causation. Lack of means? (4)

Answer 16

A: – If a particular exposure increases the risk of a certain disease but not the risk of other diseases then this is strong evidence in favour of a cause-effect relationship e.g. Mesothelioma an aspestos.
– BUT one-to-one relationships between exposure and disease are rare and lack of specificity should not be used to refute a causal relationship e.g. cigarette smoking causes many diseases.

Question 17

Q: Describe a temporal relationship as a criteria for causation. Establishing in cohort studies//cross sectional//case control? (4)

Answer 17

A: – Essential criterion.
– For a putative risk factor to be the cause of a disease it has to precede the disease.
– Generally easier to establish from cohort studies but rather difficult to establish from cross-sectional or case-control studies when measurements of the possible cause and the effect are made at the same time.
– HOWEVER, it does not follow that a reverse time order is evidence against the hypothesis.

Question 18

Q: Describe the dose-response relationship as a criteria for causation. (3)

Answer 18

A: - Describes the change in effect on an organism caused by differing levels of exposure (or doses) to a stressor (usually a chemical) after a certain exposure time, or to a food
– Further evidence of a causal relationship is provided if increasing levels of exposure lead to increasing risks of disease.
– Some causal associations, however, show a single jump (threshold) rather than a monotonic trend.

Question 19

Q: Describe plausibility as a criteria for causation. Taken seriously? (2)

Answer 19

A: – Association is more likely to be causal if consistent with other knowledge (e.g. -animal experiments, biological mechanisms, etc.).
– However, this criterion should not be taken too seriously because lack of plausibility may simply reflect lack of scientific knowledge.

Question 20

Q: Describe coherence as a criteria for causation. (2)

Answer 20

A: – Implies that a cause and effect interpretation does not conflict with what is known of the natural history.
– BUT absence of coherent information as distinguished from the presence of conflicting information, should not be taken as evidence against an association being causal.

Question 21

Q: Describe analogy as a criteria for causation.

Answer 21

A: – At best analogy provides a source of more elaborate hypotheses about the association in question.
– Absence of such analogies only reflects lack of imagination or experience, not falsity of the hypothesis

Question 22

Q: Which 2 criteria are vague and not really important in assessing causation?

Answer 22

A: coherence, analogy

Question 23

Q: What is weak association likely the result of?

Answer 23

A: confounding or bias