3.3

Chance
Is the
association
causal?
Detectives in the Classroom – Investigation 3-3: Chance
Where are we and where are we going?
1
How is this disease
distributed?
2
Is there
an association?
Hypotheses
What’s
my hypothesis?
3
Is the association
causal?
Did the exposure and
the disease turn up
together?
Why did the exposure
and the disease turn up
together?
4
What should be done
to prevent the disease?
5
Did the prevention
strategy work?
Detectives in the Classroom – Investigation 3-3: Chance
Review
What is a cause?
If an exposure causes an
outcome, how can people
be in cell b of a 2 x 2 table?
If an exposure causes an
outcome, how can people
be in cell c of a 2 x 2 table?
How does this apply to the “… cigarette smoking
causes lung cancer …” label?
Detectives in the Classroom – Investigation 3-3: Chance
Review
1.
Cause
2.
Chance
3.
4.
5.
Detectives in the Classroom – Investigation 3-3: Chance
Review
Population
All the people in a particular
group.
Detectives in the Classroom – Investigation 3-3: Chance
Epi Talk
Sample
A selection of people from a
population.
Detectives in the Classroom – Investigation 3-3: Chance
Review
Inference
Process of predicting from
what is observed in a sample
to what is not observed in a
population.
To generalize back to the
source population.
Detectives in the Classroom – Investigation 3-3: Chance
Review
Process of predicting from what is observed
to what is not observed.
Detectives in the Classroom – Investigation 3-3: Chance
Population
Detectives in the Classroom – Investigation 3-3: Chance
Population
25 cards
25 cards
25 cards
25 cards
Detectives in the Classroom – Investigation 3-3: Chance
Population
=
25 cards
25 cards
25 cards
25 cards
=
Flu
No
Flu
M&M’s
a
b
No
M&M’s
c
d
Detectives in the Classroom – Investigation 3-3: Chance
Population
=
25 cards
25 cards
25 cards
25 cards
=
Flu
No
Flu
Total
M&M’s
25
25
50
No
M&M’s
25
25
50
Detectives in the Classroom – Investigation 3-3: Chance
Review
Risk
A measure of how often an
event occurs in a defined group
of people in a defined period of
time.
The likelihood of developing a
disease.
Detectives in the Classroom – Investigation 3-3: Chance
Population
=
No
Flu
Total
M&M’s
25
25
50
No
M&M’s
25
25
50
=
=
25 cards
25 cards
25 cards
25 cards
Flu
M&M’s
No
M&M’s
Flu
No
Flu
Total
25
25
50
25
25
50
Detectives in the Classroom – Investigation 3-3: Chance
Population
=
=
25 cards
25 cards
25 cards
25 cards
M&M’s
No
M&M’s
Flu
No
Flu
Total
Risk
25
25
50
25 / 50 or 50%
25
25
50
25 / 50 or 50%
Detectives in the Classroom – Investigation 3-3: Chance
Review
Relative Risk
A way of showing the
relationship between two risks.
Tells us the number of times
one risk is larger or smaller
than another.
Calculated by dividing the
risk of an outcome in one
group by the risk of the
outcome in another group.
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Detectives in the Classroom – Investigation 3-3: Chance
1
Population
=
1
=
25 cards
25 cards
25 cards
25 cards
M&M’s
Flu
No
Flu
Total
25
25
50
Risk
Relative Risk
25 / 50 or 50 %
50 % / 50% = 1
No
M&M’s
25
25
50
25 / 50 or 50 %
The risk of getting the flu is no more likely among those who ate M&M’s
than among those who did not eat M&M’s.
Detectives in the Classroom – Investigation 3-3: Chance
Epi Teams
Detectives in the Classroom – Investigation 3-3: Chance
Population
25 cards
25 cards
25 cards
25 cards
Detectives in the Classroom – Investigation 3-3: Chance
Epi Talk
Chance
To occur accidentally.
To occur without design.
A coincidence.
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Detectives in the Classroom – Investigation 3-3: Chance
Sample
25 cards
25 cards
25 cards
25 cards
Detectives in the Classroom – Investigation 3-3: Chance
Sample
25 cards
25 cards
25 cards
25 cards
Flu
No
Flu
Total
M&M’s
5
5
10
No
M&M’s
5
5
10
Detectives in the Classroom – Investigation 3-3: Chance
Sample
25 cards
25 cards
25 cards
25 cards
Flu
No
Flu
Total
M&M’s
5
5
10
5 / 10 or 50 %
No
M&M’s
5
5
10
5 / 10 or 50 %
Risk
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Detectives in the Classroom – Investigation 3-3: Chance
2
Sample
2
25 cards
25 cards
25 cards
25 cards
M&M’s
No
M&M’s
Flu
No
Flu
Total
5
5
10
Risk
Relative Risk
5 / 10 or 50 %
50 % / 50% = 1
5
5
10
5 / 10 or 50 %
The risk of getting the flu is no more likely among those who ate M&Ms
than among those who did not eat M&Ms.
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Flu
M&M’s
No
M&M’s
No
Flu
Total
Risk
Relative Risk
5 / 10 = 50 %
50 % / 50% = 1
5 / 10 = 50 %
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Detectives in the Classroom – Investigation 3-3: Chance
3
Sample
Flu
M&M’s
No
M&M’s
No
Flu
Total
Risk
3
Relative Risk
5 / 10 = 50 %
50 % / 50% = 1
5 / 10 = 50 %
Detectives in the Classroom – Investigation 3-3: Chance
Chance
M&M’s
No
M&M’s
Flu
No
Flu
Total
Risk
5
5
10
5 / 10 or 50 %
Relative Risk
50 % / 50% = 1
5
5
10
5 / 10 or 50 %
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Relative Risks
Greater than 1
Less than 1
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Study Links Eating M&M’s
To Increased Risk of Flu
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Relative Risks
Greater than 1
Less than 1
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Study Links Eating M&M’s
To Decreased Risk of Flu
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Relative Risks
Greater than 1
25 cards
25 cards
25 cards
25 cards
Less than 1
1
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Detectives in the Classroom – Investigation 3-3: Chance
4
Different Sample Sizes
Flu
M&M’s
No
M&M’s
No
Flu
Total
Risk
Relative Risk
5 / 10 = 50 %
50 % / 50% = 1
5 / 10 = 50 %
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Detectives in the Classroom – Investigation 3-3: Chance
5
Different Sample Sizes
Flu
M&M’s
No
M&M’s
No
Flu
Total
Risk
Relative Risk
5 / 10 = 50 %
50 % / 50% = 1
5 / 10 = 50 %
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Detectives in the Classroom – Investigation 3-3: Chance
6
Different Sample Sizes
Flu
M&M’s
No
M&M’s
No
Flu
Total
Risk
Relative Risk
5 / 10 = 50 %
50 % / 50% = 1
5 / 10 = 50 %
Detectives in the Classroom – Investigation 3-3: Chance
Sample
Detectives in the Classroom – Investigation 3-3: Chance
7
Explanations for Finding an Association
1.
2.
Cause
Chance
3.
4.
5.
Study Links Eating M&M’s
To Decreased Risk of Flu
Detectives in the Classroom – Investigation 3-3: Chance
Explanations for Finding an Association
Could
the association
have occurred
by chance?
Detectives in the Classroom – Investigation 3-3: Chance
Chance
Investigation
3-3
has ended.
Detectives in the Classroom – Investigation 3-3: Chance