Children, age at first birth, and colorectal cancer risk. Data from the Melbourne Colorectal Cancer Study

During 1980 and 1981, data were obtained on the number of children and age at birth of the first child in 675 cases of colorectal cancer and in 720 age-sex frequency-matched controls as part of a large epidemiologic study of colorectal cancer conducted in Melbourne, Australia. For colorectal cancer, the relative risk (RR) for those with one or more children compared with those with no children was statistically significantly lower than one (RR = 0.61, 95% confidence interval (CI): 0.45-0.81, p less than 0.001), and with simultaneous adjustment for number of children, there was an increasing risk with increasing age at birth of first child (RR = 1.03, with each increasing year of age at birth of first child, 95% CI: 1.00-1.05, p = 0.02). The relation between both the number of children and the age at birth of the first child and colorectal cancer risk was not statistically significantly different for males and females. The risk of colorectal cancer with respect to number of children and age at birth of the first child was unaffected by the potential confounding factors of age, occupation, previous diet, oral contraceptive use in females, and a family history of colorectal cancer. Since the protection against colorectal cancer associated with having children and with earlier age at birth of the first child was found to be similar for both males and females in the Melbourne study, a life-style factor, as yet unidentified, rather than a female hormonal factor, is postulated as the mediator of these effects.