3: Measures of Association

Question 1

what does measure of association ask

Answer 1

Is a disease more common in one (exposed) population
compared with another (unexposed) population?

Question 2

Independent variables

Answer 2

Exposed or unexposed population

Question 3

Dependent variables

Answer 3

Outcome of the disease

Question 4

Measure of association

Answer 4

Used to compare the frequency of disease between specified populations
Reflect the increase in frequency of disease in one population in comparison to another

Measures of disease frequency (risks, rates) can be compared by estimating their ratios or their differences

Question 5

Measure of association example

Answer 5

One group of cattle are exposed to aflatoxin (exposure of interest) and the other is not.

Question 6

How can we compare health events between groups?

Answer 6

Ratio measures (relative risk)

Difference Measures (attributable risk)

Question 7

Ratio measures (relative risk)

Answer 7

Ratios of risk, rates or odds

Measures strength of association between exposure of interest and disease

The higher the ratio the greater the association

If no association the ratio will be close to 1

Uses contingency tables

Question 8

Prevalence in exposed group =

Answer 8

[a ÷ (a+b)]

Question 9

Prevalence in unexposed group =

Answer 9

[c ÷ (c+d)]

Question 10

Risk (prevalence) ratio =

Answer 10

[a ÷ (a+b)] ÷ [c ÷ (c+d)]

Question 11

Exposure factor

Answer 11

refers to any characteristic that may explain or predict the presence of a study outcome

Question 12

Outcome of interest

Answer 12

refers to the characteristic that is being predicted

Question 13

A study was conducted to investigate
differences in the prevalence of BVDV in
beef and dairy cattle.
1. What is the exposure factor?
2. What is our outcome of interest?

Answer 13

1. BVDV
2. Prevalence of BVDV in population

Question 14

A random selection of 159 dairy cows
were tested. 33 tested positive to BVDV
A random selection of 506 beef cows
were tested. 72 tested positive to BVDV
1. What is the seroprevalence in dairy
cows?
2. What is the seroprevalence in beef
cows?

Answer 14

Dairy: 20.8%

Beef: 14.2%

Question 15

What is the risk ratio

Answer 15

1.45

Dairy cows are 1.5 times as likely to be seropositive to BVDV as beef cows

Question 15

What is the risk ratio

Answer 15

1.45

Dairy cows are 1.5 times as likely to be seropositive to BVDV as beef cows

Question 16

Difference Measures (attributable risk)

Answer 16

risk in exposed minus risk in unexposed

Measures biological importance of the association

Determines effect of a preventive campaign

Protective factors have negative risk

Cannot be calculated for case-control studies

Question 17

How to calculate difference in prevalence between groups

Answer 17

[a ÷ (a + b)] - [c ÷ (c + d)]

Question 18

Risk difference example

Body condition score was compared between adult quenda captured in urbanised environments and bushland
Of 107 urbanised quendas, 58 had a BCS of 4 or 5
(overweight or obese)
Of 151 bushland quendas, 27 had a BCS of 4 or 5

Calculate the prevalence of BCS 4-5 in urbanised quendas

Answer 18

[a ÷ (a+b)]

58 / 107 = 54.2%

Question 19

Body condition score was compared between adult quenda captured in urbanised environments and bushland
Of 107 urbanised quendas, 58 had a BCS of 4 or 5
(overweight or obese)
Of 151 bushland quendas, 27 had a BCS of 4 or 5

Calculate the prevalence of BCS 4-5 in bushland quendas

Answer 19

[c ÷ (c+d)]

27/151

=17.9%

Question 20

Body condition score was compared between adult quenda captured in urbanised environments and bushland
Of 107 urbanised quendas, 58 had a BCS of 4 or 5
(overweight or obese)
Of 151 bushland quendas, 27 had a BCS of 4 or 5

Calculate the risk difference (write you answer in words and interpret your results)

Answer 20

Ratio

[a ÷ (a+b)] ÷ [c ÷ (c+d)]

58/107 / 27/151 = 3.02

Diff

[a ÷ (a + b)] - [c ÷ (c + d)]

58/107 - 27/151

54.2 - 17.9 = 36.3

Urban quendas are 3 times as likely to be overweight
or obese than bushland quendas

Urbanised quendas have a 36.3% excess risk of overweight or obesity compared to bushland quendas

Question 21

Can you estimate prevalence from case-control studies

Answer 21

No but can calculate odds ratio

Question 22

How to calculate:

ODDS in exposed group

ODDS in unexposed group

ODD RATIO

Answer 22

[a ÷ b]

[c ÷ d]

[a ÷ b] ÷ [c ÷ d] or [a x d] ÷ [b x c]

Question 23

Association between feeding regimen in lactating cows and herd-level incidence of displaced abomasum or clinical ketosis

Calculate the odds ratio (write you answer in words and interpret your results)

Answer 23

[a x d] ÷ [b x c]

23 x 15 / 5 x 17 = 345 / 85

Odds ratio = 4.0

high-incidence herds had 4 times the odds of feeding Total Mixed Ration compared to low-incidence herds

Question 24

Absolute measure of association

(LO8)

Answer 24

such as risk differences and incidence rate differences, like risk difference and incidence rate difference quantify excess risk or rate of disease in the exposed group compared to the unexposed group, and this provides a measure of the actual number of people who might be affected and the potential public health impact of the exposure, providing a focus on two key questions:

• How much impact would prevention have?
• How many people would benefit?

Question 25

Relative measures of association

(LO8)

Answer 25

such as risk ratio and rate ratio quantify the relative risk or rate of disease in an exposed group compared to an unexposed group and tend to be thought of as representing the strength of the association between an exposure and a disease by addressing questions such as:

• How much more likely are exposed persons to develop the outcome than the unexposed?