Lecture 18: Measures of association

Question 1

What is the base of analytic epidemiology?

Answer 1

Is the exposure associated with the outcome?

Meaning.. Does the exposure increase or decrease
the occurrence of the outcome?

Question 2

What is the PECOT framework we use for analytic epidemiology?

Answer 2

Population - The group of people in the study

Exposure - What the potential determinant is

Comparison - What the potential determinant is being compared to

Outcome - The health outcome being assessed

Time - How long people are being followed-up

Question 3

What is relative risk?

Answer 3

How many times as likely is the exposed group to develop the outcome than the comparison group?

= Exposed / Comparison

Question 4

How would we report relative risk?

Answer 4

The ‘exposed group’ were ‘value’ as likely to develop ‘outcome’ compared to the ‘comparison group’

Question 5

What are the null value, protective factor and risk factors for relative risk?

Answer 5

The null value means there is Equal occurrence of outcome in both groups. This means there is no association between exposure and outcome (1)

A protective factor means greater occurrence of outcome
in comparison group compared to an exposed group if outcome is bad = (less than 1)

A risk factor means there is a Greater occurrence of outcome in exposed group. If outcome is bad, exposure is potentially a risk factor for the outcome = (more than 1)

Question 6

Answer these risk difference questions:

if incidence exposed = incidence comparison (RD = 0) there is ..

If incidence exposed > incidence comparison (RD > 0) exposure is..

If incidence exposed < incidence comparison (RD < 0) exposure is..

Answer 6

No association or null value

Exposure is a risk factor

Exposure is a Protective factor

Question 7

How do you interpret risk difference for an incidence proportion of 0.25?

Answer 7

Example:

There were 25 extra cases per 100 people over one year
of abusive mail in epidemiologists compared to non-epidemiologists

Question 8

How do you interpret risk difference for an incidence rate of 0.25?

Answer 8

Example:

There were 25 extra cases per 100 person-year of abusive mail in epidemiologists compared to non-epidemiologists

Question 9

Relative risk vs Risk difference:

Answer 9

Relative Risk:
gives Clues to aetiology (causes)
shows the strength of association

Risk Difference:
shows the Impact of exposure
shows the Impact of removing exposure

Both are important