Examples of Cohort Study and Case-Control Study in Epidemiology

Calculating Cumulative Incidence for Different Levels of Exercise per Week

The table depicts the results of a cohort study assessing the association between hours of exercise per week with 5-year incidence of myocardial infarction (MI) in a cohort study of 3,000 adults.

Hours of exercise per week MI Cases Non-cases Total at Risk

> 5 hours 300 790 1,090

≥ 3 hours – 5 hours 210 700 910

< 3 hours 450 550 1,000

Total 960 2,040 3,000

a)  Calculate the cumulative incidence (risk) of developing myocardial infarction in those exercising greater than 5 hours per week.  Remember what piece of information is important for risk and relative risk calculations!

b)  Calculate the cumulative incidence (risk) of developing myocardial infarction in those exercising greater than 3 to 5 hours per week.

c)  Calculate the cumulative incidence (risk) of developing myocardial infarction in those exercising less than 3 hours per week.

d)  Calculate and interpret the relative risk of having a myocardial infarction, comparing those exercising greater than 5 hours per week, to those exercising 3 to 5 hours per week. (2 points)

In a two-year cohort study of 3,500 Ohioans over the age of 60, comparing daily omega-3 fatty acid supplement users to non-users and the incidence of myocardial infarction (MI), the hypothesis tested was that people taking daily omega-3 supplements were at a lower risk of having a myocardial infarction than those not taking a supplement.

Omega-3 Use Myocardial Infarction No MI Total at risk

Yes 400 1,300 1,700

No 1,100 700 1,800

Total 1,500 2,000 3,500 

a)  Calculate and interpret the relative risk of having a myocardial infarction, comparing supplement users to non-users.

b)  What other factors in these people do you think may be influencing your answer to part a?  Hint: In other words, what other characteristics about them, their exposures, etc. could be influencing your risk measure? 

Trevathan et al. conducted a case-control study of cigarette smoking and biopsy confirmed carcinoma in situ of the uterine cervix among black women ages 17 to 55 in Atlanta, Georgia in 1980.  Cases were women who attended the Dysplasia Clinic and controls were women who attended the Family Planning Clinic.  The following table was adapted from a published report.  Be sure to interpret all results.

Cases Controls

Socioeconomic Status

Low 85 165

Medium 39 10

High 26 46

Age at First Intercourse

14 or younger 18 49

15-16 27 130

17-19 30 105

20 and older 4 14

Number of Sexual Partners

1-4 90 190

5-9 37 81

10+ 30 55

a)  Calculate and interpret the appropriate measure of association of carcinoma in situ associated with low SES, and calculate the same measure of association with medium SES (use high SES as the referent).  Which socioeconomic group (low or medium) appears to be more associated with carcinoma in situ? 

b)  Calculate the odds of having 10+ sexual partners (using 1-4 as the referent) among cases.

c)  Calculate the odds of having 10+ sexual partners (using 1-4 as the referent) among controls.

d)  Calculate and interpret the odds ratio comparing part b to part c (exposure odds ratio)?

e)  Calculate the odds of being a case (versus a control) among women with 10+ sexual partners? 

f)  Calculate the odds of being a case (versus a control) among women with 1-4 sexual partners?

What types of bias are of most concern in case-control studies?  Provide hypothetical examples of how these types of bias may exist.

Briefly describe a cross-sectional study and indicate its main limitation.

Briefly describe an ecologic study and provide 3 reasons why ecologic studies are useful.  What is the main limitation of an ecologic study?

What is an “ecological fallacy”?  Why is it important to epidemiologists? 

Explain the differences among ‘no association’, ‘negative association’, and positive association’.  Provide a specific example of a measure of association (RR, OR, etc.) for each of these and provide a proper interpretation for each.

This assignment partially fulfills the following NUR 735 Course Objectives for this course:

1.Describe concepts of epidemiology and its impact on population health.

2.Interpret biostatistical methods in epidemiology relevant for the development, implementation, and evaluation of health services.

3.Evaluate epidemiological study designs used for disease tracking.