(b)For those admitted because of their chronic conditions, express their
co¤ee consumption by means of a pie chart.
(c)Calculate the generalized odds and give your interpretation. Expo-
sure is defined as having an acute condition.1.36 In a seroepidemiologic survey of health workers representing a spectrum
of exposure to blood and patients with hepatitis B virus (HBV), it was
found that infection increased as a function of contact. Table E1.36
provides data for hospital workers with uniform socioeconomic status at
an urban teaching hospital in Boston, Massachusetts.
TABLE E1.36
Personnel Exposure n HBV Positive
Physicians Frequent 81 17
Infrequent 89 7
Nurses Frequent 104 22
Infrequent 126 11(a)Calculate the proportion of HBV-positive workers in each subgroup.
(b)Calculate the odds ratios associated with frequent contacts (com-
pared to infrequent contacts). Do this separately for physicians and
nurses.
(c)Compare the two ratios obtained in part (b). A large di¤erence
would indicate a three-term interaction or e¤ect modification, where
frequent e¤ects are di¤erent for physicians and nurses.
(d)Assuming that the odds ratios for the two groups, physicians and
nurses, are equal (in other words, type of personnel is not an e¤ect
modifier), calculate the Mantel–Haenszel estimate of this common
odds ratio.1.37 The results of the Third National Cancer Survey have shown substantial
variation in lung cancer incidence rates for white males in Allegheny
County, Pennsylvania, which may be due to di¤erent smoking rates.
TABLE E1.37
Lawrenceville South Hills
Age n % n %
35–44 71 54.9 135 37.0
45–54 79 53.2 193 28.5
55–64 119 43.7 138 21.7
b 65 109 30.3 141 18.4
Total 378 46.8 607 27.150 DESCRIPTIVE METHODS FOR CATEGORICAL DATA