Introductory Biostatistics

1983–1985, and 539 control women. Of the 539 controls, 280 were hos-

pitalized women without overt cancer, and 259 were chosen from the

general population by random telephone dialing. Data for co¤ee con-

sumption are summarized in Table E4.16. Calculate the odds ratio and

its 95% confidence interval for:

TABLE E4.16

Co¤ee

Drinkers Cases

Hospital

Controls

Population

Controls

No 11 31 26

Yes 177 249 233

(a) Cases versus hospital controls.

(b)Cases versus population controls.

4.17 Postneonatal mortality due to respiratory illnesses is known to be in-
versely related to maternal age, but the role of young motherhood as a
risk factor for respiratory morbidity in infants has not been explored
thoroughly. A study was conducted in Tucson, Arizona, aimed at the
incidence of lower respiratory tract illnesses during the first year of life.
In this study, over 1200 infants were enrolled at birth between 1980 and

1984. The data shown in Table E4.17 are concerned with wheezing lower
      respiratory tract illnesses (wheezing LRI): no/yes. Using ‘‘>30’’ as the
      baseline, calculate the odds ratio and its 95% confidence interval for each
      other maternal age group.

TABLE E4.17

Boys Girls

Maternal Age (years) No Yes No Yes

< 21 19 8 20 7

21–25 98 40 128 36

26–30 160 45 148 42

> 30 110 20 116 25

4.18 Data were collected from 2197 white ovarian cancer patients and 8893
white controls in 12 di¤erent U.S. case–control studies conducted by
various investigators in the period 1956–1986 (American Journal of Epi-
demiology, 1992). These were used to evaluate the relationship of in-
vasive epithelial ovarian cancer to reproductive and menstrual charac-
teristics, exogenous estrogen use, and prior pelvic surgeries. Part of the
data is shown in Table E4.18.

EXERCISES 179