Introductory Biostatistics

tory disease. Calculate the SMRs for each subgroup and the relative risk

for group ‘‘15þ’’ versus group ‘‘1–4.’’

1.34 A long-term follow-up study of diabetes has been conducted among
Pima Indian residents of the Gila River Indian Community of Arizona
since 1965. Subjects of this study at least 5 years old and of at least half
Pima ancestry were examined approximately every two years; examina-
tions included measurements of height and weight and a number of other
factors. Table E1.34 relates diabetes incidence rate (new cases/1000
person-years) to body mass index (a measure of obesity defined as
weight/height^2 ). Display these rates by means of a bar chart.

TABLE E1.34

Body Mass Index Incidence Rate

< 20 0.8

20–25 10.9

25–30 17.3

30–35 32.6

35–40 48.5

b 40 72.2

1.35 In the course of selecting controls for a study to evaluate the e¤ect of
ca¤eine-containing co¤ee on the risk of myocardial infarction among
women 30–49 years of age, a study noted appreciable di¤erences in
co¤ee consumption among hospital patients admitted for illnesses not
known to be related to co¤ee use. Among potential controls, the co¤ee
consumption of patients who had been admitted to hospital by con-
ditions having an acute onset (such as fractures) was compared to that of
patients admitted for chronic disorders (Table E1.35).

TABLE E1.35

Cups of Co¤ee per Day

Admission by: 0 1–4 b 5 Total

Acute conditions 340 457 183 980

Chronic conditions 2440 2527 868 5835

(a) Each of the 6815 subjects above is considered as belonging to one of

the three groups defined by the number of cups of co¤ee consumed

per day (the three columns). Calculate for each of the three groups

the proportion of subjects admitted because of an acute onset. Dis-

play these proportions by means of a bar chart.

EXERCISES 49