In the WHO (1990) study risk was particularly enhanced in relation to use in low-risk countries, and in relation to long-term and recent use. Yet, individuals characterized by such use were not more likely than other users or nonusers to have smaller tumors or more localized disease at diagnosis. A small increase in risk in women who had used oral contraceptives for more than three years before age 25 was also observed in that study and such users did tend to have small tumors, but users of shorter duration, with an equally high relative risk, did not. The small increase in relative risk in relation to use before a woman's first live birth that was observed in the WHO study was found to be confined to women who had had an aborted pregnancy or stillbirth prior to their first live birth. In those cases the tumors tended to be smaller than the tumors in other women, suggesting the possibility of enhanced surveillance for breast cancer in this small group.
On balance, it appears unlikely that the major positive finding from these studies (and by inference from other studies as well) can be explained by enhanced screening in users of oral contraceptives.
Women who go through the menopause late in life are at higher risk for breast cancer than women with an early natural or artificial menopause. If use of oral contraceptives late in a woman's potentially reproductive years were to simulate endocrinologically a late menopause, one would expect such use to increase a woman's risk of breast cancer. This possibility has not been adequately investigated, but results of analyses relevant to this question are summarized in Table B-25. The findings are equivocal. Three studies show increased relative risks in users over age 45 (Vessey et al., 1979; Yuan et al., 1988) or over age 50 (Jick et al., 1980), but this was not observed by Rosenberg and colleagues (1984). In addition, although use of oral contraceptives