Advancing Sexual and Reproductive Health and Rights
Perspectives on Sexual and Reproductive Health
Volume 34, Number 1, January/February 2002

Citing a Preference for Bottle-Feeding, the Vast Majority of Black Women Do Not Breastfeed

Black women are less than half as likely as white women to breastfeed, and the association is significant even when socioeconomic characteristics are controlled for. They also are considerably less likely than white women to cite logistical reasons for not breastfeeding, and are more likely to say that they simply prefer to bottle-feed. Differences in breastfeeding help explain why black babies are more likely than white infants to die before reaching one year of age: When breastfeeding is taken into account, the racial gap in infant mortality disappears. These are among the key findings of an analysis of data from the last two rounds of the National Survey of Family Growth (NSFG).1

Using information from 1995 NSFG respondents who had a child aged 18 months or younger at the time of the interview, researchers explored the association between breastfeeding and a range of maternal and birth factors that other studies have concluded influence the decision to breastfeed. After calculating bivariate odds ratios to determine the effect of each factor, they conducted multivariate analyses to examine the effects of all factors combined.

The analyses included 1,088 women, of whom 833 were white and 255 were black. Overall, 57% of women had ever breastfed their youngest child, but the proportion was much higher among white women (65%) than among black respondents (30%). White women also were more likely than black women to have high levels of income and education, to be married, to live in the West and to have wanted their last birth--all factors consistent with a greater inclination to breastfeed.

Racial differences were evident in a number of other characteristics as well. For example, black women were less likely than white women to be foreign-born and to be Catholic; they were more likely to live in central cities, to be unemployed, to want no more children and to report that their infant had been low-birth-weight. Black and white women were about equally likely to be smokers, to have only one child and to have delivered by cesarean.

At the bivariate level, all characteristics except parity and type of delivery were significantly associated with the decision to breastfeed. Among the findings were reduced odds of having breastfed among black women, women who wanted no more children or had not wanted the previous birth, and those whose infant had been low-birth-weight. The odds were elevated among women of higher socioeconomic status, residents of the West and Northeast, and foreign-born respondents.

However, few associations remained significant in the multivariate analysis. When all factors were controlled for, black women were still less than half as likely as white women to have breastfed (odds ratio, 0.4), and women who reported wanting no more children had reduced odds compared with those who wanted to continue childbearing (0.6). The odds of having breastfed rose as women's income increased (1.1) and were elevated for women who had gone to college (1.9), who were foreign-born (1.6) and who lived in the West (3.7). An additional set of analyses, including only the factors that were statistically significant, yielded essentially the same results.

The data also revealed significant racial differences in why women choose not to breastfeed. In response to a multiple-choice question on what drove this decision, 83% of black respondents, but only 62% of whites, reported that they preferred bottle-feeding. Four percent and 14%, respectively, attributed their decision to job-related or scheduling difficulties; similarly, 8% of blacks and 18% of whites said that physical or medical problems prevented them from breastfeeding.

Finally, because research on racial variations in infant mortality has focused mainly on racial differences in the incidence of low birth weight, the investigators conducted another set of analyses to determine whether black women's relatively low likelihood of breastfeeding helps explain why their infants are less likely than babies born to white women to survive to age one. These analyses, which were based on data from the 1988 and 1995 cycles of the NSFG combined, indicated that black infants were roughly 50% more likely than others to die before one year of age (odds ratio, 1.5). However, when either low birth weight or breastfeeding was controlled for, each of those factors was significantly associated with the odds of infant death, and race no longer played a role. When both controls were taken into account simultaneously, the risk of death was significantly elevated for low-birth-weight infants (odds ratio, 4.3) and significantly reduced for babies who had been breastfed (0.2); race showed no association with the odds of infant death.

Commenting on their findings, the researchers stress the need to "better understand the factors that inhibit breastfeeding among black women and the factors that promote a preference for bottle-feeding." Efforts to persuade black women of the benefits of breastfeeding, they conclude, are "as critical in reducing black infant mortality as is targeting low-birth-weight infants."--D. Hollander


1. Forste R, Weiss J and Lipincott E, The decision to breastfeed in the United States: does race matter? Pediatrics, 2001, 108(1):291-296.