5. MEASURES OF ASSOCIATION

Question 1

1. Which kind of variables can we relate Exposure and Outcome to?

Answer 1

• Exposure is an Independent Variable
• Outcome is a dependent Variable

EXAMPLE:
- exposure to harmful risk factors will lead to a poor
health outcome

Question 2

2. Define Exposure.

Answer 2

• this is an event, state or behaviour to which a group of
  people is exposed
• it has an effect on the probability of the outcome

Question 3

3. Provide some examples of Exposure.

Answer 3

• exposure to the Sun (UV Radiation)
• smoking
• intaking of fruits and vegetables
• physical activity
• oral contraceptive use
• obesity

Question 4

4. Define Outcome.

Answer 4

• this is an event, state or behaviour
• we hypothesise that is has resulted from an exposure
  to some other variable (event, state, behaviour)

Question 5

5. Provide some examples of Outcomes.

Answer 5

• they are usually a disease
• they can also be a measure related to health and
  disease

EXAMPLES:
- obesity
- colorectal cancer
- type 2 diabetes
- cardiovascular disease
- plasma triglyceride levels
(an increase in triglycerides= an increase in blood sugar)

Question 6

6. Does this diagram make sense?

Answer 6

• yes

Question 7

7. Provide labels for the spaces numbered 1-6.

Answer 7

1. Binary Outcomes
2. Existing Cases
3. New Cases
4. Odds Ratio
5. Risk Ratio
6. Rate Ratio

Question 8

8. Provide labels for the spaces numbered 7-12.

Answer 8

7. Numeric Outcomes
8. Categorical Exposure
9. Numeric Exposure
10. Mean Difference
11. Correlation Coefficient
12. Regression Coefficient

Question 9

9. What is meant by Binary Outcomes?

Answer 9

• you either have the disease or you do not

Question 10

10. What is meant by New Cases?

Answer 10

• this means that the exposure can happen before the
  outcome
• this helps with calculating incidence

Question 11

11. What are the 3 ways of calculating Binary Outcomes?

Answer 11

1. Odds Ratio
2. Risk Ratio
3. Rate Ratio

Question 12

12. What do measures of Association assess?

Answer 12

• they assess the presence between the magnitude of an
  association
• this association happens between the Exposure and
  the Outcome

NB:
- they assess the change in outcome for a given change
in exposure
- this applies to both numeric and categorical exposures
and outcomes

Question 13

13. What does Relative Risk refer to?

Answer 13

• it refers to the comparison between two risks
• risks are also known as incidences

IT IS ALSO KNOWN AS:
- as risk in relation to a reference risk

Question 14

14. What is Relative Risk a measure of?

Answer 14

• it is a measure of association
• it assesses the probability of an event occurring in an
  exposed group of people
• this is then relative to the probability of the same event
  occurring in an unexposed group of people

Question 15

15. What are other terms for the following:
    15.1. the event occurring
    15.2. the exposed group of people
    15.3. the unexposed group of people

Answer 15

15.1. the risk of the outcome

15.2. this is the numerator in the equation

15.3. the reference

Question 16

16. What types of variables are Exposure and Outcome?

Answer 16

• they are both categorical

Question 17

17. What are the two Measures of Association referred to as Relative Risk?

Answer 17

1. Risk Ratio
2. Rate Ratio

Question 18

18. What is Risk Ratio?

Answer 18

• this is the ratio of two risks
• this is the ratio of two cumulative incidences

Question 19

19. What is Rate Ratio?

Answer 19

• this is the ratio of two rates
• this is the ratio of two incidence rates
• the denominator for this is total person time

Question 20

20. What is another name for a Rate Ratio?

Answer 20

• a Hazard Ratio

Question 21

21. In a Relative Risk, which two groups of individuals are compared?

Answer 21

1. the exposed category
2. the unexposed (reference) category

Question 22

22. Let’s say the Exposure is smoking status, what is the exposed group and what is the unexposed group?

Answer 22

1. EXPOSED GROUP:
   
   • smokers
2. UNEXPOSED GROUP:
   
   • non-smokers
   • this is the reference group

Question 23

23. Let’s say the Exposure is Medication type
    (a positive exposure), what is the exposed group and
    what is the unexposed group?

Answer 23

1. EXPOSED GROUP:
   
   • those who are taking active medication
2. UNEXPOSED GROUP:
   
   • the Placebo group

Question 24

24. What happens when there are more than 2 categories in the Categorical Exposure Variable?

Answer 24

• we have to decide which category will be treated as the
  Unexposed (Reference) Category
• all the other categories are then treated as Exposed
  Categories
• they are then compared to the Unexposed Category
  one at a time

Question 25

25. Let’s say our exposure is Alcohol consumption, what are out exposed and unexposed groups?

Answer 25

EXPOSED:
- heavy consumer

EXPOSED:
- moderate consumer

UNEXPOSED:
- light consumer
- this is the reference against which the other categories
are compared

Question 26

26. When we calculate Relative Risk, which category is always compared?

Answer 26

• the Exposed Category is always compared to the
  Unexposed (Reference) Category
• IT IS NEVER THE OTHER WAY AROUND

Question 27

27. In which situations are both Risk and Rate Ratio applicable?

Answer 27

• in situations where there is a follow up
• these are known as Prospective Studies
• in situations where there are new, certain outcome
  cases
• rather than existing ones

Question 28

28. How do we calculate Risk Ratio?

Answer 28

• your numerator is the incidence of the disease in the
  group of the exposed
• your denominator is looking at your total population
  at risk

Question 29

29. How do we calculate Rate Ratio?

Answer 29

THIS DIFFERS FROM THE RISK RATIO:
- only when looking at the denominator