1. Epidemiology Kept Simple
Chapter 8: Association & Impact
4/20/2017
Epidemiology Kept Simple
Chapter 8 Measures of Association & Potential Impact
Epi Kept Simple

2. Chapter 8: Association & Impact
4/20/2017
Important Jargon
Exposure (E)  an explanatory factor; any potential health determinant; the independent variable
Disease (D)  the response; any health-related outcome; the dependent variable
Measure of association (syn. measure of effect)  a statistic that quantifies the relationship between an exposure and a disease
Measure of potential impact  a statistic that quantifies the potential impact of removing a hazardous exposure
In epidemiology it is common to use the term exposure to denote any explanatory variable
i.E we may speak of smoking as an exposure that causes lung cancer or advanced maternal age at pregnancy as an exposure that causes Down Syndrome
(Prof. G. could you please help me understand the inactive lifestyle E+??)){We arbitrary define the risk factor as “exposure” to an inactive lifestyle. Then we compare the mortality rate in the exposed (inactive lifestyle) and nonexposed (active lifestyle) groups. I’ve added a third bullet to this effect.}
Gerstman
Chapter 8
Epi Kept Simple

3. Arithmetic (αριθμός) Comparisons
Chapter 8: Association & Impact
4/20/2017
Arithmetic (αριθμός) Comparisons
Measures of association are mathematical comparisons
Mathematic comparisons can be done in absolute terms or relative terms
Let us start with this ridiculously simple example:
I have $2
You have $1
We compare the weight of a man of 100 kg to the weight of a woman of 50 kg.
-Absolute comparisons are derived by subtraction and using (original units of measure kg)
-{Relative comparisons are derived by division (the division cancels out units, making a unit-free comparison}
"For the things of this world cannot be made known without a knowledge of mathematics."- Roger Bacon
Gerstman
Chapter 8
Epi Kept Simple

4. Chapter 8: Association & Impact
4/20/2017
Absolute Comparison
In absolute terms, I have $2 – $1 = $1 more than you
Note: the absolute comparison was made with subtraction
It is as simple as that…
We compare the weight of a man of 100 kg to the weight of a woman of 50 kg.
-Absolute comparisons are derived by subtraction and using (original units of measure kg)
-{Relative comparisons are derived by division (the division cancels out units, making a unit-free comparison}
Gerstman
Chapter 8
Epi Kept Simple

5. Chapter 8: Association & Impact
4/20/2017
Relative Comparison
Recall that I have $2 and you have $1.
In relative terms, I have $2 ÷ $1 = 2, or
“twice as much as you”
Note: relative comparison was made by division
We compare the weight of a man of 100 kg to the weight of a woman of 50 kg.
-Absolute comparisons are derived by subtraction and using (original units of measure kg)
-{Relative comparisons are derived by division (the division cancels out units, making a unit-free comparison}
Gerstman
Chapter 8
Epi Kept Simple

6. Chapter 8: Association & Impact
4/20/2017
Applied to Risks
Suppose, I am exposed to a risk factor and have a 2% risk of disease.
You are not exposed and you have a 1% risk of the disease.
Of course we are assuming we are the same in every way except for this risk factor.
In absolute terms, I have 2% – 1% = 1% greater risk of the disease
This is the risk difference
We compare the weight of a man of 100 kg to the weight of a woman of 50 kg.
-Absolute comparisons are derived by subtraction and using (original units of measure kg)
-{Relative comparisons are derived by division (the division cancels out units, making a unit-free comparison}
Gerstman
Chapter 8
Epi Kept Simple

7. Chapter 8: Association & Impact
4/20/2017
Applied to Risks
In relative terms I have
2% ÷ 1% = 2, or twice the risk
This is the relative risk associated with the exposure
We compare the weight of a man of 100 kg to the weight of a woman of 50 kg.
-Absolute comparisons are derived by subtraction and using (original units of measure kg)
-{Relative comparisons are derived by division (the division cancels out units, making a unit-free comparison}
Gerstman
Chapter 8
Epi Kept Simple

8. Chapter 8: Association & Impact
4/20/2017
Terminology
For simplicity sake, the terms “risk” and “rate” will be applied to all incidence and prevalence measures.
Let’s apply arithmetic to risks. These are the formulas. {The formulas are simple: RD is a subtraction and RR is a division. The key is to understand how we interpret the RR and the RD. They both quantify the relation between E and D, but they tell you something different about the association. Going through lots of examples in the book will help understand subtleties.}
Gerstman
Chapter 8
Epi Kept Simple

9. Chapter 8: Association & Impact
4/20/2017
Risk Difference
Risk Difference (RD)  absolute effect associated with exposure
where
R1 ≡ risk in the exposed group R0 ≡ risk in the non-exposed group
Let’s apply arithmetic to risks. These are the formulas. {The formulas are simple: RD is a subtraction and RR is a division. The key is to understand how we interpret the RR and the RD. They both quantify the relation between E and D, but they tell you something different about the association. Going through lots of examples in the book will help understand subtleties.}
Interpretation: Excess risk in absolute terms
Gerstman
Chapter 8
Epi Kept Simple

10. Chapter 8: Association & Impact
4/20/2017
Relative Risk
Relative Risk (RR)  relative effect associated with exposure or the “risk ratio”
where
R1 ≡ risk in the exposed group R0 ≡ risk in the non-exposed group
Let’s apply arithmetic to risks. These are the formulas. {The formulas are simple: RD is a subtraction and RR is a division. The key is to understand how we interpret the RR and the RD. They both quantify the relation between E and D, but they tell you something different about the association. Going through lots of examples in the book will help understand subtleties.}
Interpretation: excess risk in relative terms.
Gerstman
Chapter 8
Epi Kept Simple

11. Example Fitness & Mortality (Blair et al., 1995)
Chapter 8: Association & Impact
4/20/2017
Example Fitness & Mortality (Blair et al., 1995)
Is improved fitness associated with decreased mortality?
Exposure ≡ improved fitness (1 = yes, 0 = no)
Disease ≡ death (1 = yes, 0 = no)
Mortality rate, group 1: R1 = 67.7 per 100,000 p-yrs
Mortality rate, group 0: R0 = per 100,000 p-yrs
{See p. 159 for details.}
Gerstman
Chapter 8
Epi Kept Simple

12. Example Risk Difference
Chapter 8: Association & Impact
4/20/2017
Example Risk Difference
What is the effect of improved fitness on mortality in absolute terms?
{See p. 159 for details.}
The effect of the exposure (improved fitness) is to decrease mortality by 54.4 per 100,000 person-years
Gerstman
Chapter 8
Epi Kept Simple

13. Chapter 8: Association & Impact
4/20/2017
Example Relative Risk
What is the effect of improved fitness on mortality in relative terms?
{See p. 159 for details.}
The effect of the exposure is to cut the risk almost in half.
Gerstman
Chapter 8
Epi Kept Simple

14. Designation of Exposure
Chapter 8: Association & Impact
4/20/2017
Designation of Exposure
Switching the designmation of “exposure” does not materially affect interpretations
For example, if we had let “exposure” ≡ failure to improve fitness
RR = R1 / R0 = / = (1.8 times the risk in the exposed group (“almost double”)
{See p. 159 for details.}
Gerstman
Chapter 8
Epi Kept Simple

15. 2-by-2 Table Format Disease + Disease − Total Exposure + A1 B1 N1
For person-time data: let N1 ≡ person-time in group 1 and N0 ≡ person-time in group 0, and ignore cells B1 and B0
Gerstman
Chapter 8

16. Fitness Data, table format
Fitness Improved?
Died
Person-years
Yes 25 --
4054 No 32
2937
Rates per 10,000 person-years
Gerstman
Chapter 8

17. Food borne Outbreak Example
Exposure ≡ eating a particular dish
Disease ≡ gastroenteritis
Disease +
Disease −
Total
Exposure + 63 25 88
Exposure – 1 6 7 64 31 95
Gerstman
Chapter 8

18. Food borne Outbreak Data
Disease +
Disease −
Total
Exposure + 63 25 88
Exposure – 1 6 7 64 31 95
Exposed group had 5 times the risk
Gerstman
Chapter 8

19. Chapter 8: Association & Impact
4/20/2017
Comparison of RR and RD
RR  strength of effect
RD  effect in absolute terms
Rates (per ) of Lung CA & CHD assoc. w/smoking
Smoker
Nonsmoke RR RD
LungCA
104 10
10.40 94
CHD
565
413
1.37
152
Smoking causes more heart disease even though the association between smoking a heart disease is weaker than the association between smoking an lung cancer. This is because heart disease is more common in the population.
Smoking  Stronger effect for LungCA
Smoking  Causes more CHD
Gerstman
Chapter 8
Epi Kept Simple

20. What do you do when you have multiple levels of exposure?
Compare rates to least exposed “reference” group
LungCA Rate (per 100,000 person-years) RR
Non-smoker (0) 10
1.0 (ref.)
Light smoker (1) 52
5.2
Mod. smoker (2)
106
10.6
Heavy sm. (3)
224
22.4
Gerstman
Chapter 8

21. The Odds Ratio Similar to a RR, but based on odds rather than risks D+
Total E+ A1 B1 N1 E− A0 B0 N0 M1 M0 N
When the disease is rare, interpret the same way you interpret a RR
e.g. an OR of 1 means the risks are the same in the exposed and nonexposed groups
“Cross-product ratio”
Gerstman
Chapter 8

22. Odds Ratio, Example Milunsky et al, 1989, Table 4 NTD = Neural Tube Defect
Folic Acid+ 10
10,703
Folic Acid− 39
11,905
Exposed group had 0.29 times (about a quarter) the risk of the nonexposed group
Gerstman
Chapter 8

23. Measures of Potential Impact
These measures predicted impact of removing a hazardous exposure from the population
Two types
Attributable fraction in exposed cases
Attributable fraction in the population as a whole
Gerstman
Chapter 8

24. Attributable Fraction Exposed Cases (AFe)
Proportion of exposed cases averted with elimination of the exposure
Gerstman
Chapter 8

25. Example: AFe
RR of lung CA associated with moderate smoking is approx Therefore:
Interpretation: 90.4% of lung cancer in moderate smokers would be averted if they had not smoked.
Gerstman
Chapter 8

26. Attributable Fraction, Population (AFp)
Proportion of all cases averted with elimination of exposure from the population
Gerstman
Chapter 8

27. AFp equivalent formulas
Gerstman
Chapter 8

28. AFp for Cancer Mortality, Selected Exposures
Doll & Peto, 1981
Miller, 1992
Tobacco
30%
29%
Dietary
35%
20%
Occupational 4% 9%
Repro/Sexual 7%
Sun/Radiation 3% 1%
Alcohol 6%
Pollution 2% -
Medication
Infection
10%
Gerstman
Chapter 8