KEY
FPH7240 – Introduction to Epidemiology
Fall 2018
Assignment 2
Due: Thursday, September 13, 2018 at noon
Readings: Aschengrau & Seague, Ch. 3
... [Show More] “Comparisons of Disease Frequency”
A note about terminology: As Aschengrau & Seague mention in the text, what they call risk differences
and rate differences (e.g. incidence rate difference, cumulative incidence difference) some
epidemiologists refer to as the attributable risk. I was trained by those epidemiologists, and have learned
and taught these concepts as the attributable risk (AR), population attributable risk (PAR), attributable
risk percent (AR%) and population attributable risk percent (PAR%). The concepts are the same, and I will
attempt to use their language, but old habits die hard, so at the end of the assignment is a table adding
my formulas and wording to the comparable concepts at presented by A&S.
1. Researchers at Northwestern University followed 4,226 men and 5,624 women for a total of
111,222 person-years to determine whether several risk factors were associated with incident
cardiovascular disease (CVD). Diabetes was one of the risk factors of interest, and 112 men and
71 women in the study were diabetic at baseline. Of these individuals, 68 diabetic men and 37
diabetic women went on to develop CVD. Throughout follow-up, 43.1% of men and 31.9% of
women went on to develop CVD.
a. Based on the information provided, please construct the appropriate 2x2 table of diabetes
and CVD risk for men.
Exposure CVD No CVD Total
Diabetic 68 112 – 68 = 44 112
Not Diabetic 1,821 – 68 = 1,753 4,114 – 1,753 = 2,361 4,226 – 112 = 4,114
0.431*4,226 = 1,821 4,226 – 1,821= 2,405 4,226
b. Please construct the appropriate 2x2 table for diabetes and CVD risk for women.
Exposure CVD No CVD Total
Diabetic 37 71 – 37 = 34 71
Not Diabetic 1,794 – 37 = 1,757 3,830 – 34 = 3,796 5,624 – 71 = 5,553
0.319*5,624 = 1,794 5,624 – 1,794 = 3,830 5,624
c. What is the relative risk of CVD associated with diabetes among men? Among women?
Relative risk or rate ratio = Incidence exposed / Incidence unexposed
Among men, cumulative incidence in the exposed (diabetics) = 68/112 = 60.7% and
cumulative incidence in the unexposed (non-diabetics) is 1,753/4,114 = 42.6%. [Show Less]