The epidemiology of ovarian cancer in Greece: a case-control study.

One hundred and fifty women with common malignant epithelial tumors of the ovary (cases) and 250 comparison women hospitalized for various orthopedic conditions were interviewed regarding demographic, reproductive, socio-economic and biomedical characteristics, including their use of coffee, tobacco, alcohol, drugs and exogenous estrogens. The data were analyzed with standard X2 procedures and by modelling relative risk (r) through multiple logistic regression. The main results are as follows: women with ovarian cancer had fewer liveborn children (one-tailed, P congruent to 0.13, for women with 4+ children, r = 0.6) and later menarche (P congruent to 0.02, for women with age at menarche 15+, r = 1.9) and menopause (P congruent to 0.07, for women with age at menopause 50+, r = 1.5); they were slightly taller and heavier (P congruent to 0.15 and 0.30 respectively); they belonged to smaller sibships (P congruent to 0.05); they reported more frequently than controls familial occurrence of ovarian, endometrial and breast cancer (P less than 0.05 in each instance); they were regular consumers of alcoholic beverages more frequently than controls, and the excess was both statistically significant (P congruent to 0.02) and dose-related; they were regular users of coffee slightly more frequently than controls (r = 1.2) but the excess was small and it was neither statistically significant nor dose-related (P congruent to 0.27); and they reported less frequently than controls the use of oral contraceptives (relative risk for users, 0.4; 95% confidence interval, 0.1-1.1).