EPIDEMIOLOGY
Measures of association
Sewnet Getaye (BSC,MPH)
Email:
[email protected] Learning objectives
At the end of this unit students will be able
To understand what is measure of association
To identify which measure of association is
appropriate for a specific study design
To calculate measure of association for cause-
effect relationship arises from each study design
To differentiate the types of measure of
associations (X2 , RR, OR, AR, PAR)
2
Measures of association
Once an analytic study has been designed to
test a hypothesis, and the data has been
gathered, the data have to be analyzed to
determine whether there is an association
between the exposure and the outcome being
studied.
Association is:
• A statistical relationship between two or 3
Exposure and Outcome
• Exposure (E) = the explanatory variable
– Any agent, host, or environmental factor
that may have an effect on health
• Outcome = the response variable
– Any health outcome of the exposure
(Disease, Death, Injury, Disability, etc )
4
Exposure and Outcome
Example: Exposure (E) vs. Disease (D)
We need a way to quantify the effect of E on
D
This is accomplished by comparing rates of
disease in an exposed group (E+) and non-
exposed groups (E-)
5
Measures of association
• Assess the strength of an association between
an exposure and the outcome of interest
• Indicate how more or less likely a group is to
develop disease as compared to another group
The two widely used measures are:
1. Odds ratio (OR) - Case-control, cross -sectional
studies
2. Relative risk (RR) - Cohort , experimental studies
6
Presentation of data by 2X2 table
2x2 tables can arise from each of the designs where
there are two factors of interest (exposure status and
disease) and the data are the numbers of individuals
(frequencies) in each four cells.
Disease
Yes (+) No (+) Total
Exposure Yes (+) a b a+b
No (+) c d c+d
Total a+c b+d a+b+c+d 7
Presentation of data by 2X2 table
We will call the factor levels “exposed” (E)
and “unexposed” (NE) and “diseased” (D)
and “not-diseased” (ND).
1. a= Exposed, and diseased
2. b= Exposed, Not diseased
3. c= Not exposed, diseased
4. d= Not exposed, Not diseased
8
Presentation of data by 2X2 table
9
Odds ratio (OR)
• Odds are ratio of two probabilities i.e. The
probability that the event WILL occur to the
probability that the event will NOT occur
• For example, in 100 births, the probability of
a delivery being a boy is 55% and being a girl
is 45%
• The odds of a delivery being a girl is 45/55 =
0.82 10
Odds ratio (OR)
• In simpler term, an odds of an event can be
calculated as:
- Number of events divided by number of non-
events
Odds refer to single entity
If an event has the probability P, then the
odds of the same event is: Odds = P/1-P
11
Odds ratio (OR)
Odds Of Disease:
odds of having the disease if exposed
Ratio of diseased to non – diseased in
exposed
Odds Of Exposure:
Odds of having the exposure in disease
Ratio of exposed to non-exposed in those
with disease 12
Odds ratio (OR)
• The odds of exposure is the probability that
an individuals experience the exposure
divided by the probability that they do not
• The odds ratio (OR) is the odds of exposure
in the diseased group divided by the odds of
exposure in the non-diseased group
13
Odds ratio (OR)
• The odds of disease is the probability that an
individuals experience the disease divided
by the probability that they do not
• The odds ratio (OR) is the odds of disease in
the exposed group divided by the odds of
disease in the unexposed group
14
Odds ratio (OR)
• Note: in the exposure-disease 2x2 table, the
odds of having a disease in the exposed
group is the same as the odds that an
exposed person develops the disease
• The exposure odds ratio is equal to the
disease odds ratio
15
Odds ratio (OR)
16
Odds ratio (OR)
17
Odds ratio (OR)
When is Odds Ratio(OR) a good estimate of
relative risk?
– The odds ratio is a good approximation of the risk
ratio if the disease is rare
– If the disease is rare, then OR RR
WHY?
– If the disease is rare, the probability of it NOT
happening is close to 1, and the odds is close to
18
the risk
Odds ratio (OR)
19
Interpretation of OR
• OR can assume values anywhere between 0
and infinity
• A value of 1 indicates no association
between the risk factor and disease status
• A value 1 indicates increased odds of having
the disease among subjects in whom the
risk factor is present
20
Relative Risk (RR)
• Relative Risk is a ratio of two risks ,
• Also called Risk Ratio
• It is a ratio of the probability (risk) of
developing the disease among those exposed
to the probability of developing disease
among those unexposed .
• An indicator of the strength of the association
21
between the exposure and the disease
Relative Risk (RR)
•
22
Relative Risk (RR)
23
Interpretation of RR
• The value of RR ranges anywhere between 0
and infinity
• A value of 1 indicates that there is no
association between the status of the risk
factor and the status of the dependent
variables (i.e., the risk of acquiring the disease
is the same for those subjects with the risk
factor and without the risk factor) 24
Interpretation of RR
• A value of RR greater than 1 indicates that
there is an association between the status of
the risk factor and the status of the dependent
variables (i.e., the risk of acquiring the disease
is greater among subjects with the risk factor
than among subjects without the risk factor)
25
Interpretation of RR
• A RR value that is less than 1 indicates less
risk of acquiring the disease among
subjects with the risk factor than among
subjects without the risk factor, suggestive
that the factor is protective
26
Interpretation of RR
27
Protective Fraction
• Exposure is associated with decreased risk
if RR<1 (example, immunization)
• Is the reduction in risk to the exposed group
compared to the control group
• Indicates the proportion of potential cases
prevented by the exposure
• Is called Relative Risk Reduction (RRR)
28
Measure of Public Health Impact
Four closely related measures are used:
1. Attributable Risk
2. Attributable Risk Percent
3. Population Attributable Risk
4. Population Attributable Risk Percent
• Note: All these measures assume that the
association between exposure and disease
29
Attributable Risk (AR)
• The Incidence of disease in the Exposed
population whose disease can be attributed to
the exposure
• Attributable risk is also called the risk
difference (RD)
• The risk difference reflects an excess risk
attributable to exposure
30
– What is the excess risk due to the exposure?
Attributable Risk (AR)
• Attributable Risk(Risk Difference) is the prob.
that exposed person develops outcome
because of added influence of exposure
31
Interpretation of AR
• If AR = 0, there is no association between
exposure and disease
• If AR>0, there is an association between
exposure and disease (it indicates the
number of cases among the exposed that
can be eliminated if the exposure is
removed
32
• If AR<0, the exposure is preventive
Attributable Risk Percent(AR%)
• The proportion of disease among the
exposed that is attributable to the exposure,
or the proportion of the disease that could
be prevented by eliminating the exposure
• Other names:
– Attributable Rate Percent
– Attributable proportion
– Etiologic fraction 33
Attributable Risk Percent(AR%)
34
Population Attributable Risk(PAR)
• The Incidence of disease in the total
population whose disease can be attributed
to the exposure
• The excess rate of disease in the total study
population of exposed and non-exposed
individuals that is attributable to the
exposure
35
Population Attributable Risk Percent
(PAR%)
•
36
