Women's risk of breast cancer is modestly increased following their first full-term pregnancy, but it then falls gradually and is not affected by subsequent pregnancies. Moreover, the risk related to a first birth is especially elevated among older women: With every five-year increase in the age at which a woman has her first full-term pregnancy, her odds of developing breast cancer rise by 7%. These findings, which are consistent with results of earlier work, are derived from analyses of a large data set that permitted the researchers to control for a more comprehensive set of breast cancer risk factors than had been possible previously.1
To address the sometimes conflicting results of studies based on small samples and limited numbers of risk factors, analysts pooled data from two large, population-based case-control studies conducted in four states (Maine, Massachusetts, New Hampshire and Wisconsin). For each study, women with breast cancer were identified through statewide cancer registries, and women of the same age with no history of the disease were randomly selected to serve as controls. All women participated in telephone interviews, in which they provided extensive information on their background characteristics and reproductive history. The analyses are based on data for 9,891 women aged 20-79 with breast cancer and 12,271 controls interviewed between 1988 and 1996.
Although the studies involved different age-groups (one was based on women aged 20-74 and the other on women 50-79), background characteristics for women in both, regardless of their disease status, were similar. About nine in 10 women had had at least one full-term pregnancy (defined as a pregnancy lasting six or more months and ending in a live birth or a stillbirth): Roughly one in 10 had had one, nearly half had had 2-3 and about a quarter had had 4-9. Fewer than one in five women in all subgroups had been younger than 20 when they first gave birth, about two-thirds had been in their 20s and roughly one in 10 had been 30 or older. Similarly, the distributions according to women's age at their last full-term pregnancy were similar for all subgroups: About one in 10 had been younger than 25, one-quarter each had been in their late 20s and early 30s, and the remainder had been 35 or older.
After combining the data from the two studies, the analysts used logistic regression techniques to examine factors in women's breast cancer risk, controlling for the study and site, as well as the woman's age, education, age at menarche, menopausal status and age at menopause, parity, lifetime duration of lactation, alcohol consumption, body mass index and family history of breast cancer. According to the results for the overall sample, each five-year increase in a woman's age at first full-term pregnancy raised her breast cancer risk by 7% (odds ratio, 1.07), but age at subsequent pregnancies was not a significant factor. The analyses also suggest that breast cancer risk declined as a woman's parity increased. For example, the odds ratio went from 0.83 among women who had had 3-4 full-term pregnancies to 0.78 among those who had had five and 0.71 among those who had had six.
Analyses that included only women who had had at least two full-term pregnancies and that controlled for the same set of variables revealed that the likelihood of developing breast cancer rose by 13% for every five-year increase in a woman's age at her first full-term pregnancy and by 7% for every five years that her age at her last full-term pregnancy increased (odds ratios, 1.13 and 1.07, respectively). However, when women's age at second and later full-term pregnancies was taken into account, the effect of age at first full-term pregnancy was reduced (odds ratio, 1.08), and the impact of age at last full-term pregnancy was no longer statistically significant.
Similar analyses based on women who had had only two full-term pregnancies underscore the importance of age at the first: When all variables were taken into account, a woman's risk of breast cancer rose by 15% for every five-year increase in her age at her first full-term pregnancy (odds ratio, 1.15) but was not affected by her age at her last full-term pregnancy. In addition to exploring the effects on breast cancer risk of a woman's age at her first full-term pregnancy, the analysts examined the influence of birthspacing. They found that in general, for women who had had five or fewer full-term pregnancies, the interval between births had no effect on the risk of breast cancer, regardless of the woman's age at her first full-term pregnancy. At higher parities, however, closely spaced births had a protective effect for young women. For example, women who were 20 at the time of their first full-term pregnancy and who had had seven or more births were 10% less likely than nulliparous women to develop breast cancer if their births were spaced three years apart, but their risk was reduced by almost 40% if their births were separated by only one year.
In their final set of calculations, the analysts found that the risk of breast cancer is elevated after a woman has one birth, but the increase in risk diminishes over time. Thus, shortly after delivering, women who give birth only once may be as much as 50% more likely than nulliparous women to develop breast cancer (depending on their age); however, after about 30 years, their risk is indistinguishable from that of their nulliparous counterparts. Unfortunately, as the analysts point out, this finding implies that women who have a first full-term pregnancy while in their 30s have an elevated risk of breast cancer during the high-risk ages.
According to the analysts, the size of their study and the broad array of risk factors examined add weight to their findings that other research has lacked. In discussing the results, they note that both hormonal and tissue changes that occur with an early first full-term pregnancy may contribute to lowering a woman's risk of breast cancer. Furthermore, they observe that the "negligible effect" of a woman's age at subsequent births suggests that these changes provide long-term protection.--D. Hollander
REFERENCE
1. Chie W-C et al., Age at any full-term pregnancy and breast cancer risk, American Journal of Epidemiology, 2000, 151(7):715-722.