ORAL-CONTRACEPTIVES AND BREAST-CANCER RISK AMONG YOUNGER WOMEN

Background: Several studies have suggested a link between oral contrac eptive use and breast cancer in younger women, but it is possible that chance or bias, including selective screening of contraceptive users, contributed to the putative association. Purpose: Given that oral con traceptives were first marketed in the United States in the early 1960 s, we conducted a population-based case-control study to examine the r elationship between use of oral contraceptives and breast cancer among women in a recently assembled cohort, focusing on women younger than 45 years of age who had the opportunity for exposure throughout their entire reproductive years. Methods: Breast cancer patients and healthy control subjects were identified, the latter group by random-digit di aling, in Atlanta, Ga., Seattle/Puget Sound, Wash., and central New Je rsey, In Seattle and New Jersey, the study was confined to women 20 th rough 44 years of age; in Atlanta the age range was extended through 5 4 years, Patients included women with in situ or invasive breast cance r newly diagnosed during the period of May 1, 1990, through December 3 1, 1992. In-person interviews were completed by 2203 (86.4%) of 2551 e ligible patients and 2009 (78.1%) of 2571 eligible control subjects. A nalyses focused on women younger than 45 years of age (1648 patients a cid 1505 control subjects) to maximize opportunities for extended expo sure. Logistic regression analyses were used to obtain maximum likelih ood estimates of relative risks (RRs) and their 95% confidence interva ls (CIs). Results: Among women younger than 45 years, oral contracepti ve use for 6 months or longer was associated with an RR for breast can cer of 1.3 (95% CI = 1.1-1.5). Risks were enhanced for breast cancers occurring prior to age 35 years (RR = 1.7; 95% CI = 1.2-2.6), with the RR rising to 2.2 (95% CI = 1.2-4.1) for users of 10 or more years. Th e RR for breast cancer for those whose oral contraceptive use began ea rly (before age 18 years) and continued long-term (>10 years) was even higher (RR = 3.1; 95% CI = 1.4-6.7). The RRs observed for those who u sed oral contraceptives within 5 years of cancer diagnosis were higher than for those who had not, with the effect most marked for women you nger than age 35 years (RR = 2.0; 95% CI = 1.3-3.1). Oral contraceptiv e associations were also strongest for cancers diagnosed at advanced s tages. Evaluation of screening histories and methods of diagnosis fail ed to support the speculation that associations could be due to select ive screening. Among women 45 years of age and older, no associations of risk with use of oral contraceptives were noted. Conclusions: The r elationship between oral contraceptives and breast cancer in young wom en appears to have a biologic basis rather than to be an artifact or t he result of bias.