QUESTION 1
1. What are the main conditions under which it is desirable to use case-control studies?
2. Name some sources to identify cases for a case-control study. Provide examples of each source and potential biases associated with each source.
3. Suppose that a case-control study was conducted to evaluate the relationship between artificial sweeteners and bladder cancer. 18,000 cases and 18,000 controls were enrolled in the study. Among the cases, 7,758 had used artificial sweeteners in the past, while 10,242 had never used artificial sweeteners. Of the controls, 5,130 had used sweeteners and 12,870 had not.
A. Set up the two by two table for these data.
B. What measure of association/comparison should be used to assess the strength of the relationship between artificial sweeteners and bladder cancer?
C. Calculate this measure.
D. Interpret the result in one sentence.
QUESTION 2
1. What are the major advantages of case-control studies?
2. What are the major disadvantages of case-control studies?
3. How do these advantages and disadvantages compare with those associated with cohort studies?
4. Why can’t we estimate incidence rates in most case-control studies? Illustrate this in an example.
5. Suppose that a case-control study was conducted among men aged 40-70 years in order to determine whether exercising for two or more hours per week decreases the likelihood of heart attack. The case group was comprised of 1,000 men who had recently had a heart attack. Among the cases, 236 reported that they had regularly exercised for two or more hours per week prior to their heart attack. The control group was comprised of 1,000 individuals who never had a heart attack. Among the controls, 379 reported that they exercised regularly.
A. Set up the two by two table for these data.
B. Calculate the strength of the association between regular exercise and heart attack.
C. Do these results support the hypothesis that regular exercise increases or decreases the likelihood of heart attack?
QUESTION 3
1. What is the main principle in identifying an appropriate control group? How can you "test" whether your controls meet the criteria of this principle using the "would criterion"?
2. What are some sources of controls? What are the advantages and disadvantages of each?
3. Which types of bias are of concern in case-control studies? How can these biases be minimized/eliminated?