The relationship between oral contraceptives (OC) and breast-cancer risk was analysed using data from a case-control study conducted between June 1991 and February 1994 in 6 Italian centres on 1,991 patients below age 65 with histologically confirmed incident breast cancer and 1,899 controls admitted to hospital for a wide range of acute, non-neoplastic, non-hormone-related diseases. "Ever OC use" was reported by 18% of cases versus 14% of controls, corresponding to a multivariate odds ratio (OR) of 1.1 (95%) confidence interval, Cl 0.9 to 1.4). The ORs were 1.3 for use lasting < 1 year, 1.1 for 1 to 4 years, 0.9 for 5 to 8 years, and 1.2 for over 8 years. With reference to age at first use, there was some indication that the OR was elevated in women who had started use before age 30, but not in those starting at a later age. With reference to time since last OC use, the OR was above unity for women who had stopped for less than 10 years (1.6 for 1 to 4 years; 1.7 for 5 to 9 years), but the OR declined to unity for women who had stopped OC use for 10 years or longer. The OR for women who had stopped OC use for less than 10 years was consistently elevated across strata of selected covariates, and was directly related to the duration of use (OR 1.3 for < 5 years, 1.7, for > or = 5 years). In contrast, the OR was 0.6, for use lasting > or = 5 years in women who had stopped for 10 years or more. The elevated OR for women who had recently stopped OC use, together with the absence of association (or the suggestion of some protection) for those who had stopped for 10 years or more is consistent with the pattern of breast-cancer risk observed after a full-term pregnancy, and provides important reassurance on a public health level on the long-term impact of OCs on breast carcinogenesis.
PIP: During June 1991-February 1994 in Italy, researchers conducted a case control study to examine the association between oral contraceptive (OC) use and breast cancer risk. Cases comprised 1991 women aged less than 65 with confirmed breast cancer. Controls included 1899 same-age patients at the same network of hospitals as the cases with an acute, non-neoplastic, non-hormone-related condition. 18% of cases and 14% of controls had ever used OCs. Overall, there was no association between OC use and breast cancer. Very short term OC use (1 year) had an odds ratio (OR) of 1.3 (borderline significance only). Recall or selection bias may have accounted for this moderate association. However, women who had recently stopped using OCs faced an elevated risk of breast cancer (OR = 1.6 for 1-4 years since last OC and 1.7 for 5-9 years). This elevated risk remained above unity across strata of selected covariates (education, age at menarche, parity, age at 1st birth, no history of breast cancer, and history of benign breast disease). Among women who had last used OCs 10 years ago, the risk of breast cancer was directly related to duration of use (OR = 1.3 for 5 years and 1.7 for =or 5 years; p = 0.02 [for the trend]). The OR for women who had last used an OC longer than 10 years ago reached unity. Among these women, those that used OCs for at least five years had a lower risk of developing breast cancer (OR = 0.66). Women who started using OCs before age 30 had a somewhat elevated risk of developing breast cancer (OR = 1.3 for 25 years and 1.4 for 25-29 years). The increased OR for women who had recently ceased using OCs and the apparent protection against breast cancer among women who ceased using OCs for at least 10 years match the pattern of breast cancer risk noted after a full-term pregnancy. These findings should reassure the public that OCs apparently do not induce breast carcinogenesis.