Lecture 18: Measures of Association

Question 1

What is the importance of a comparison group in analytic epidemiology?

Answer 1

To have a baseline of the outcome because we need to see whether the exposure is associated with the outcome (ie. does the exposure increase or decrease the likelihood of the outcome).

Question 2

How can we find associations?

Answer 2

through analytic study designs

Question 3

What are five examples of analytic study designs?

Answer 3

1. cross sectional
2. ecological
3. cohort
4. case-control
5. RCTs

Question 4

Fundamental characteristics and logic of analytic study designs can be described using _______ and _______

Answer 4

PECOT

GATE

Question 5

What does PECOT stand for?

Answer 5

Population
Exposure
Comparison
Outcome
Time

Question 6

What is meant by the “population” of analytic study designs?

Answer 6

the group of people in the study

Question 7

What is meant by the “exposure” of analytic study designs?

Answer 7

what the potential determinant is

Question 8

What is meant by the “comparison” of analytic study designs?

Answer 8

what the potential determinant is being compared to

Question 9

What is meant by the “outcome” of analytic study designs?

Answer 9

the health outcome being assessed

Question 10

What is meant by the “time” of analytic study designs?

Answer 10

how long people are being followed up for

Question 11

In the GATE frame, what is the difference between the source and the sample?

Answer 11

The source is the population which the sample is recruited from whereas the sample is the population included in the study

Question 12

What are two measures of association that we can measure?

Answer 12

relative risk

risk difference

Question 13

How can you calculate relative risk?

Answer 13

incidence of the exposed group divided by the incidence of the comparison group

Question 14

What does the relative risk show?

Answer 14

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

Question 15

What is the null value for relative risk and how does this come about?

Answer 15

1
when the incidence in the exposed and comparison groups are the same so there is no association between the exposure and the outcome

Question 16

What does it mean if the relative risk is greater than the null value?

Answer 16

There is a greater incidence of outcome in the exposed group compared to the comparison group and therefore there is a greater likelihood of the outcome in the exposed group

Question 17

What does it mean if the relative risk is less than the null value?

Answer 17

There is a greater incidence of outcome in the comparison group compared to the outcome group and therefore there is a greater likelihood of the outcome in the comparison group

Question 18

What is the template for interpreting relative risk?

Answer 18

The [exposed group] were [value] as likely to develop [outcome] compared to [comparison]

Question 19

How can you calculate risk difference?

Answer 19

incidence in the exposed group minus the incidence in the control group

Question 20

What does the risk difference show?

Answer 20

How many extra/fewer cases of the outcome in the exposed group are attributable to the exposure

Question 21

What is the null value for risk difference and what does this mean?

Answer 21

0
This means that there is no difference in the incidence of the outcome of interest between the exposed and comparison group

Question 22

If the risk difference is greater than the null value, is the exposure a risk or protective factor?

Answer 22

risk

Question 23

If the risk difference is smaller than the null value, is the exposure a risk or protective factor?

Answer 23

protective

Question 24

What is the template for interpreting risk difference?

What is different about reporting incidence rate and incidence proportion?

Answer 24

There were [value] extra/fewer cases of [outcome] in [exposed group] compared [comparison group].

For the value in incidence proportion, you would say there were ___ extra/fewer cases per 100 people over ____ years whereas for incidence rate, you would say something like there were _____ extra/fewer cases per 100 person years

Question 25

What are the benefits of using relative risk as the measure of association?

Answer 25

it gives clues to the causes and shows the strength of the association

Question 26

What are the benefits of risk difference?

Answer 26

it shows the impact of the exposure and removing the exposure