Evaluating Risk of Disease

1
Evaluating Risk of Disease

• Bioterrorism Epidemiology
• Module 8
• Missouri Department of Health
• And Senior Services

2
What Is Risk

• A probability (likelihood)
• Comparing risk or likelihood of disease in
  exposure groups or likelihood of exposure in
  disease groups
• Relative risk and odds ratio are the most common
  measures of risk

3
Relative Risk
Incidence in the Exposed Group Incidence in the
Non Exposed Group
4
Relative Risk
Disease
-
a
b

a b
_
Exposure
c
d
c d
a/ab c/cd
Incidence in Exposed
RR
Incidence in Non Exposed
5
Calculation of Person Years
Person 1 2 3 4 5 6 7 8
No. Years Of Observation 3 8 10 9 6 4 1 2 43
Disease Status d nd nd d d nd nd nd
Person Years
6
Calculation of Person Years
Person 1 2 3 4 5 6 7 8
No. Years Of Observation 3 8 10 9 6 4 1 2 43
Person Years
7
Calculation of Person Years
Person 1 2 3 4 5 6 7 8
No. Years Of Observation 3 8 10 9 6 4 1 2 43
Person Years
8
Calculation of Person Years
Person 1 2 3 4 5 6 7 8
No. Years Of Observation 3 8 10 9 6 4 1 2 43
Person Years
9
Calculation of Person Years
Person 1 2 3 4 5 6 7 8
No. Years Of Observation 3 8 10 9 6 4 1 2 43
Person Years
10
Relative Risk Using Incidence Density
Number Diseased Persons
Person Years
a
b
Exposed
Not Exposed
c
d
a/b c/d
RR
11
Relative Risk Using Incidence Density
Number Diseased Persons
Person Years
a
b
Exposed
Not Exposed
c
d
a/b c/d
RR
12
Relative Risk Using Incidence Density
Number Diseased Persons
Person Years
a
b
Exposed
Not Exposed
c
d
a/b c/d
RR
13
Relative Risk Using Incidence Density
Number Diseased Persons
Person Years
a
b
Exposed
Not Exposed
c
d
a/b c/d
RR
14
Relative Risk Using Incidence Density
Number Diseased Persons
Person Years
a
b
Exposed
Not Exposed
c
d
a/b c/d
RR
15
Relative Risk Using Incidence Density
Calculate the Relative Risk for this Data
Diseased Persons
Person Years
10
2000
Exposed
5
6000
Not Exposed
16
Relative Risk Using Incidence Density
Diseased Persons
Person Years
10
2000
Exposed
5
6000
Not Exposed
10 / 2000 5 / 6000
RR

6.00
17
Odds Ratios Can Estimate Relative Risk

• Can be used in case control studies to evaluate
  the likelihood of exposure in two disease groups.
• Can be used in cross-sectional studies to
  evaluate the likelihood of disease in two
  exposure groups when incidence data is not
  available.

18
Odds Ratio (Case Control)
Odds of Exposure in the Cases Odds of Exposure in
the Controls
19
Odds Ratio Case Control Study
Exposed Non Exposed
a
b
Cases
c
d
Controls
a / b c / d
Odds of exposure in diseased
OR
Odds of disease in non diseased
20
Odds Ratio (Cross-Sectional Study)
Odds of Disease in the Exposed Group Odds of
Disease in the Non Exposed Group
21
Odds Ratio Cross-Sectional Study
Diseased Non-Diseased
a
b
Exposed
c
Non-Exposed
d
a / b c / d
Odds of disease in Exposed
OR
Odds of disease in Non Exposed
22
Relative Risk Cohort Study
Calculate the relative risk for this data
The Entire Population
Diseased Non-Diseased
10,000
990,000
1,000,000
Exposed
3,000,000
5,000
2,995,000
Non-Exposed
15,000
3,985,000
4,000,000
23
Relative Risk Cohort Study

• Iexp 10,000 / 1,000,000 X 100,000 1,000 per
  100,000
• Inexp 5,000 / 3,000,000 X 100,000 167 per
  100,000
• RR Iexp / Inexp 1,000 / 167 6.00

Diseased Non Diseased
10,000
990,000
Exposed
5,000
2,995,000
Non-Exposed
24
Odds Ratio Cross-Sectional Study
Calculate the odds ratio for this data
Diseased Non-Diseased
Exposed
10,000
990,000
5,000
2,995,000
Non-Exposed
25
Odds Ratio Cross-Sectional Study

• ODDS OF DISEASEexp 10,000 / 990,000 0.01010
• ODDS OF DISEASEnexp 5,000 / 2,995,000 0.00167
• OR ODDSexp / ODDSnexp 0.0101 / 0.0017 5.94

Diseased Non Diseased
Exposed
10,000
990,000
Non-Exposed
5,000
2,995,000
26
Relative Risk Cohort Study
Calculate the RR for this data
Diseased Non Diseased
250,000
750,000
1,000,000
Exposed
125,000
2,875,000
3,000,000
Non-Exposed
15,000
3,985,000
4,000,000
27
Relative Risk Cohort Study

• Iexp 250,000 / 1,000,000 X 100,000 25,000 per
  100,000
• Inexp 125,000 / 3,000,000 X 100,000 4,167 per
  100,000
• RR Iexp/Inexp 25,000/4,167 6.00

Diseased Non-Diseased
Exposed
250,000
750,000
Non-Exposed
125,000
2,875,000
28
Relative Risk Cohort Study
Calculate the OR for this data
Diseased NonDiseased
750,000
Exposed
250,000
125,000
2,875,000
Non-Exposed
29
Relative Risk Cohort Study

• ODDS OF DISEASEexp 250,000 / 750,000 0.3333
• ODDS OF DISEASEnexp 125,000 / 2,875,000
  0.0435
• OR ORexp / ORnexp 0.3333/0.0435 7.66

Diseased Non-Diseased
250,000
750,000
Exposed
Non-Exposed
125,000
2,875,000
30
Odds Ratio Case Control Study
Cases Controls
a
c
Exposed
b
d
Non-Exposed
a / b c / d
Odds of Exposure in Cases
OR
Odds of Exposure in Controls
31
Odds Ratio Case Control Study
Calculate the Odds Ratio for this data
Cases Controls
100
100
Exposed
50
350
Non-Exposed
32
Odds Ratios Case Control Study

• ODDS OF EXPOSUREcases 100 / 50 2.0
• ODDS OF EXPOSUREcontrols 100 / 350 0.29
• OR ODDScases / ODDScontrols 2 / 0.29 6.90

Cases Controls
100
100
Exposed
Non-Exposed
50
350
33
Dose Response
Calculate the Odds Ratios using no exposure as
your reference category
Cases Controls
24
No Exposure
4
22
lt One Year
28
20
12
1 - 10 Years
10
2
gt 10 Years
34
Dose Response
Cases Controls OR
100
No Exposure
4
24
---
22
lt One Year
28
7.6
20
12
1 - 10 Years
10.0
10
2
gt 10 Years
30.0
35
Dose Response
(28/4) / (22/24) 7.6 (20/4) / (12/24)
10.0 (10/4) / (2/24) 30.0
Cases Controls OR
No Exposure
100
4
24
---
lt One Year
22
28
7.6
1 - 10 Years
20
12
10.0
2
10
gt 10 Years
30.0
36
Attributable Risk (Risk Difference)
Incidence in the Exposed Incidence in the Non
Exposed
37
Attributable Risk As a Percentage
Incidence in the Exposed Incidence in the Non
Exposed Incidence in the Exposed
100
38
Comparison of Relative Risk and Attributable Risk
Lung Cancer 166 7
Exposure Category Heavy Smokers Nonsmokers
Disease per 100,000 population
Coronary Heart Disease 599 422
Calculate the Relative Risk, the Attributable
Risk, and the Percent Attributable Risk for this
data.
39
Comparison of Relative Risk and Attributable Risk
Coronary Heart Disease 599 422
Lung Cancer 166 7
Exposure Category Heavy Smokers Nonsmokers
Measure of Excess Risk Relative
risk Attributable risk Percent Attributable
risk
599 422
166 7
23.7
1.4
166 7 159
599 422 177
159 166
177 599
30
96