(c) From the result of part (b), fill in the missing death rates (per
100,000 population) in the table.1.18 The survey described in Example 1.1, continued in Section 1.1.1, pro-
vided percentages of boys from various ethnic groups who use tobacco
at least weekly. Display these proportions in a bar chart similar to the
one in Figure 1.2.
1.19 A case–control study was conducted relating to the epidemiology of
breast cancer and the possible involvement of dietary fats, vitamins, and
other nutrients. It included 2024 breast cancer cases admitted to Roswell
Park Memorial Institute, Erie County, New York, from 1958 to 1965. A
control group of 1463 was chosen from patients having no neoplasms
and no pathology of gastrointestinal or reproductive systems. The pri-
mary factors being investigated were vitamins A and E (measured in
international units per month). Data for 1500 women over 54 years of
age are given in Table E1.19. Calculate the odds ratio associated with a
decrease (exposure is low consumption) in ingestion of foods containing
vitamin A.
TABLE E1.19
Vitamin A (IU/month) Cases Controls
a150,500 893 392
>150,500 132 83
Total 1025 4751.20 Refer to the data set in Table 1.1 (see Example 1.2).
(a) Calculate the odds ratio associated with employment in shipyards
for nonsmokers.
(b)Calculate the same odds ratio for smokers.
(c) Compare the results of parts (a) and (b). When the di¤erence is con-
firmed properly, we have a three-term interaction or e¤ect modifica-
tion, where smoking alters the e¤ect of employment in shipyards as a
risk for lung cancer.
(d) Assuming that the odds ratios for the two groups, nonsmokers and
smokers, are equal (in other words, smoking is not an e¤ect modi-
fier), calculate the Mantel–Haenszel estimate of this common odds
ratio.1.21 Although cervical cancer is not a major cause of death among American
women, it has been suggested that virtually all such deaths are prevent-
able. In an e¤ort to find out who is being screened for the disease, data
EXERCISES 41