3Who Breastfeeds in the United States?

Before 1970

During the twentieth century, infant-feeding practices have undergone dramatic changes that reflect shifts in values and attitudes in the U.S. society as a whole. They have tended to occur first among those women at the forefront of changes in dominant social values and among those with the resources (whether it is time, energy, or money) to permit adoption of new feeding practices.

Examples of alternative feeding practices, such as the use of wet nurses or human milk substitutes, occur throughout recorded history (Fildes, 1986). The early twentieth century was marked, however, by an unprecedented increase in formula feeding, in part because the development of nutrition science coincided with a pervasive increase in the value placed on scientific products and processes (Rosenberg, 1976; Starr, 1982). As women asserted their rights for self-determination in public life, those who were wealthy enough and sufficiently in tune with contemporary values were adopting formula feeding for their infants.

A corresponding and consistent decline in breastfeeding is evident in data from U.S. fertility surveys (Hendershot, 1980, 1981; Hirschman and Hendershot, 1979), specific studies of infant feeding (Meyer, 1958, 1968), and market research surveys conducted by Ross Laboratories, ¹ a manufacturer of infant formula (Martinez and Krieger, 1985; Martinez and Nalezienski, 1979, 1981; Martinez et al., 1981). Seventy-seven percent of the infants born between 1936 and 1940 were breastfed; the incidence declined during the subsequent decades to about 25% by 1970 (Hendershot, 1980, 1981; Hirschman and Hendershot, 1979; Meyer, 1958, 1968). Duration of breastfeeding declined as well, dropping from a mean of 4.2 months in the early 1930s to 2.2 months in the late 1950s.

Because this decline in breastfeeding was not uniform in all segments of the U.S. population, the demographic characteristics of the group of mothers who breastfed changed substantially (Table 3-1). Ethnic differences in rates of decline are especially striking. Rates fell sharply among blacks; there was a less pronounced decline among whites and Hispanics (Hirschman and Hendershot, 1979). Before 1960, black mothers were more likely than white mothers to breastfeed. After 1960, the reverse was true because of the more rapid decline among black mothers.

TABLE 3-1

Percentages of First-Born Infants Breastfed Between 1951 and 1970 in the United States, by Ethnic Group and Education.

The trends are somewhat more complex for mothers with different educational levels. In the 1950s breastfeeding rates were lower for women with a high school education than for those with more or less education, but by 1970, these rates varied directly with education (Hendershot, 1981; Hirschman and Hendershot, 1979). According to some observers (Meyer, 1968; Salber et al., 1958), the differences in trends for women with different educational levels represent the trickling down of values and behaviors from economically and socially advantaged women to less advantaged women. Those observers conclude that women with less education adopted formula feeding as the culmination of a trend initiated by better educated mothers earlier in the century.

Turnaround in the 1970s

The overall downward trend in breastfeeding incidence reached its nadir at 22% in 1972 (Hendershot, 1981) (Figure 3-1). The subsequent increase was not uniform across the population. Breastfeeding incidence rose among white and black mothers, although the increase was greater among whites (Hendershot, 1981). Although data on Hispanic mothers are incomplete, there are indications that their breastfeeding incidence remained stable or even continued to fall (Smith et al., 1982). The incidence rose in all education groups (Hendershot, 1981; Martinez and Nalezienski, 1981; Martinez et al., 1981). There are few data on the association between breastfeeding and income. Those from the National Survey of Family Growth indicate that breastfeeding rates remained unchanged among low-income mothers but increased among those with middle and higher incomes (Hendershot, 1981).

FIGURE 3-1

Percentage of infants breastfed, by survey and year, based on data from the National Survey of Family Growth (Hendershot, 1981; Pratt et al., 1984) and A. Ryan, Ross Laboratories, personal communication, 1990. Both government and market survey data are (more...)

Duration of breastfeeding increased after 1972 as well. Approximately 10% of breastfeeding mothers continued the practice for at least 3 months in 1972, whereas approximately 20% did so in 1975 and 37% did so in 1984 (Hendershot, 1981; Martinez and Krieger, 1985; Martinez and Nalezienski, 1981; Martinez et al., 1981).

Ironically, this return to breastfeeding in the 1970s parallelled its decline in the early twentieth century. Once again, women of higher socioeconomic status were the first to adopt a "new" feeding method (breastfeeding), and their rationale was couched in scientific language, this time, focused on such issues as immunologic factors and maternal-infant bonding. Reflecting broader cultural values, many women adopted natural childbirth and, with it, what they considered natural infant feeding.

The upward trend in breastfeeding after 1972 appeared to peak in 1982 at about 61% for initiation and 40% for the percentage of mothers breastfeeding 3 months or longer. The exact zenith of the trend is difficult to pinpoint, because Ross Laboratories changed its weighting procedures to obtain a better reflection of the socioeconomic makeup of the U.S. population (Martinez and Krieger, 1985), and data for the 1980s from the National Survey of Family Growth were not yet available for the subcommittee's use. The evidence from Ross Laboratories suggests a gradual, persistent decline in breastfeeding during the 1980s.

1989 Data

The initiation of breastfeeding by women delivering in 1989 was reported to be 52.2% (Table 3-2) (A. Ryan, Ross Laboratories, personal communication, 1990). By age 5 or 6 months,² only 19.6% of the infants were still breastfed. Of the three ethnic groups compared, white mothers initiated breastfeeding at the highest rates and black mothers did so at the lowest rates. Forty percent of white mothers who initiated breastfeeding were still breastfeeding at 5 or 6 months, compared with approximately 30% of black or Hispanic mothers. A breakdown of these data by parity indicates that, except among Hispanics, primiparous and multiparous women initiate breastfeeding at about the same rates, but the former breastfeed for a shorter duration. The data also show that married mothers are much more likely than unmarried mothers to breastfeed and that they experience a far lower rate of attrition by 5 to 6 months. This difference is especially pronounced among black mothers.

TABLE 3-2

Percentage of Mothers Breastfeeding Newborn Infants in the Hospital and Infants at 5 or 6 Months of Age in the United States in 1989, by Ethnic Background and Selected Demographic variables.

Higher percentages of older mothers initiate breastfeeding, and they continue for a longer duration than younger mothers do. Breastfeeding both in the hospital and at 5 or 6 months is positively associated with maternal education and family income. Mothers with any college education are especially likely to initiate breastfeeding. Black mothers with a high school education or less initiate breastfeeding only about half as often as college educated black mothers do.

A breakdown of breastfeeding by census region shows distinct regional trends. The highest rates for initiating breastfeeding (Figure 3-2 [in the hospital]) and at 5 or 6 months post partum (Figure 3-3) are found in the Mountain and Pacific regions; rates in the East South Central region are the lowest. Ethnic differences within census regions are the same as those seen in national data.

FIGURE 3-2

Breastfeeding initiation rates, by census region and ethnic background (W, white; B, black; H, Hispanic). Data from the 1989 Ross Mothers Survey (A. Ryan, Ross Laboratories, personal communication, 1990).

FIGURE 3-3

Breastfeeding rates at 5 to 6 months post partum, by census region and ethnic background (W, white; B, black; H, Hispanic). Data from the 1989 Ross Mothers Survey (A. Ryan, Ross Laboratories, personal communication, 1990).

The Ross Laboratories data on participants in the Special Supplemental Program for Women, Infants, and Children (WIC) indicate that 34% breastfeed and 9% continue the practice for at least 5 or 6 months (Table 3-3). The demographic patterns observed in this group are the same as those in the general population: higher rates of breastfeeding are reported for white women, older women, and those with some college education. Except for the Ross Laboratories data, there are only sparse data on breastfeeding rates among the national WIC population. A comparison of WIC and non-WIC respondents in the Ross Laboratories surveys shows consistently lower rates of breastfeeding among WIC mothers. However, these differences are likely due to differences in income and other sociodemographic variables. Rush et al. (1988) found no difference between WIC and non-WIC women in the rates of breastfeeding when women with similar levels of socioeconomic status were compared.

TABLE 3-3

Percentage of Mothers Breastfeeding Newborn Infants in the Hospital and at 5 or 6 Months of Age in 1989, by Participation in the Special Supplemental Food Program for Women, Infants, and Children (WIC) and Selected Demographic Variables.

Limitations of the Data

There are no data on the total length of breastfeeding (actual duration) and the ages at which other milk or foods are added to the infant's diet. For example, it is not known how much longer the 20% of women in the Ross Laboratories survey still breastfeeding their 5- to 6-month old infants will continue breastfeeding. Neither is it known how many of the breastfed infants are exclusively breastfed and how many are partially breastfed. The lack of these data makes it difficult to calculate the amounts of nutrients delivered to the infant in human milk as well as the nutritional cost of milk synthesis to the mother. The increased incidence of breastfeeding from 1969 to 1980 (Forman et al., 1985) appears to be accounted for primarily by women who chose exclusive breastfeeding (and not partial breastfeeding) over bottle feeding. Thus, the total amount of nutrients delivered in human milk to the infant would have been different on average for lactating women in 1969 and 1980.

At present, data are inadequate for documenting breastfeeding rates among several subgroups of the U.S. population of particular interest to health care providers. These include Hispanics, Southeast Asians, and working mothers.

Employed Mothers

There are no statistics that adequately compare the incidence and duration of breastfeeding by mothers who work outside the home and those who do not. Ryan and Martinez (1989) attempted to make this comparison by using the Ross Laboratories data for 1987, but their analysis provides only limited information about the infant-feeding practices of women in the labor force (see Table 3-4). In that study, mothers were classified as "employed full time" or "not working" based on their response to a question on employment status when their infants were 6 months of age. Mothers working part time were excluded. The investigators determined the rates of breastfeeding for these two groups both when their infants were newborns (in the hospital) and when they were 6 months old. From these data, they calculated a breastfeeding continuance rate, that is, the percentage of women who initiated breastfeeding who were still breastfeeding when their infants reached 6 months of age.

TABLE 3-4

Newborn Infants Breastfed in the Hospital and at 6 Months of Age, and Breastfeeding Continuance Rate, Among Mothers Employed Full Time and Mothers Not Employed^,.

Ryan and Martinez (1989) reported that identical proportions (54.5%) of women in both employment categories initiated breastfeeding. By 6 months, however, 10% of the employed mothers were still breastfeeding compared with 24.3% of those who were not working. When these data are broken down by sociodemographic variables, trends generally follow those seen in the population as a whole. As described above, women in the Mountain and Pacific regions were more likely to initiate breastfeeding than were those living in other regions, regardless of their employment status.

One intriguing finding is that at 6 months post partum, more black mothers in the "employed full time" category than in the "not working" category had initiated breastfeeding (32.7% compared with 21.2%, respectively). This contrasts with whites, who initiated breastfeeding at similar rates (58.9 and 62.3%, respectively) whether or not they would be employed outside the home at 6 months post partum.

At 6 months after delivery, sociodemographic differences were not associated with breastfeeding continuance rates among employed mothers. At this stage, employed black mothers breastfed at the same rate as white mothers; this was higher than expected for black mothers, considering their generally shorter duration of breastfeeding.

There are no data relating breastfeeding to prenatal employment or prenatal intentions to breastfeed. Ryan and Martinez (1989) did not include the 15% of mothers employed part time in their sample; many of these women may use part-time employment as a strategy to accommodate breastfeeding.

Kurinij and colleagues (1989) examined the effect of intention to return to work outside the home on breastfeeding among mothers in Washington, D.C. Among blacks, but not whites, more mothers who intended to return to work part time initiated breastfeeding than those who intended to return full time. In examining duration of breastfeeding, they found that professional women returning to work breastfed longer than women working in other occupations did, and those employed part time breastfed longer than those in full-time work did. Among white women only, the later the return to work, the longer the duration of breastfeeding. In general, they found breastfeeding duration to be shorter among blacks than whites. This finding may differ from that of Ryan and Martinez (1989) because of the high socioeconomic status of the whites studied by Kurinij and associates (1989).

Ethnic Groups

Employed Mothers

Few studies have been conducted to examine the effects of working outside the home on breastfeeding. On the whole, the demographic characteristics of breastfeeding mothers employed outside the home are the same as those of breastfeeding mothers in the general population. However, the overall rate of breastfeeding is lower among employed mothers and the duration is shorter.

Mothers employed outside the home face special problems when breastfeeding their infants. A survey of employed breastfeeding mothers by Auerbach and Guss (1984) shows that excessive fatigue, the logistics of pumping and storing milk, the excessive time spent traveling to and from the baby during the workday, and concern about having an adequate milk supply were common problems in this group. These problems were compounded by lack of time to complete work duties and to eat properly.

These findings are not surprising in light of a survey of policies and practices to support breastfeeding mothers in the workplace (Moore and Jansa, 1987). The survey sought responses from 100 of the most profitable Fortune 500 companies and an additional 12 companies that were known to have breastfeeding support programs. All the Fortune 500 companies responding had maternal leave policies and guaranteed a return to an equivalent job. However, only about 33% offered flex-time or part-time work, only 14% allowed breastfeeding at work, and fewer than 5% allowed job sharing or provided day care. Most of the 12 companies known to support breastfeeding allowed flex time, part-time schedules, and job sharing. However, fewer than half provided a place for mothers to breastfeed, and none provided day care or permitted mothers breaks for breastfeeding.

In another study, Barber-Madden and colleagues (1987) reviewed published research as well as policy statements of professional groups such as the American Academy of Pediatrics (AAP, 1982) and identified six types of barriers to breastfeeding for employed mothers. They include (1) lack of child care at or near the workplace, (2) work environments that do not provide a place for pumping and storing milk, (3) restrictive employer policies that fail to provide adequate maternity leave and job security, (4) social attitudes of employers and coworkers toward breastfeeding that result in disapproval and harassment, (5) inadequate maternal knowledge about breastfeeding, and (6) lack of knowledge concerning breastfeeding on the part of health professionals, especially those in occupational health. These authors suggest both short- and long-term strategies for dealing with such barriers that would require economic commitments by employers or governments as well as the development of educational materials on breastfeeding targeted specifically at the problems of employed mothers.

Adolescent Mothers

The few small studies conducted on pregnant adolescents or adolescent mothers indicate that these adolescents may not differ substantially from older mothers in their selection of infant-feeding methods. In a survey of pregnant inner-city adolescents aged 12 to 19 years, investigators found that those intending to breastfeed were likely to perceive more benefits of breastfeeding, desired more information about it, had supportive environments, had been breastfed themselves, and perceived relatively fewer barriers to breastfeeding than those who did not intend to breastfeed (Joffee and Radius, 1987; Radius and Joffee, 1988). Overall, perceived differences in benefits differentiated the potential breastfeeders from nonbreastfeeders better than did perceived barriers.

Baisch and associates (1989) found that 13- to 20-year-old low-income adolescents intending to breastfeed had more positive attitudes about breastfeeding than those who did not intend to breastfeed, especially if they had been breastfed themselves. Prenatal intentions were highly predictive of actual postnatal feeding practices.

Neifert and associates (1988b) found that among 129 breastfeeding mothers less than 18 years of age in Denver, 35% nursed for less than 1 month, 22% nursed for 1 to 2 months, and 43% nursed for 2 months or longer. The most common reason given for weaning in the first month was the ''inability of the infant to grasp the maternal nipple and nurse effectively" (Neifert et al., 1988b, p. 472). These infants were said to have had this problem in the hospital and continued to have it at home. Although 75% of these teenage mothers persevered by expressing milk and feeding it to the infant in a bottle, all weaned their infants by 1 month. None of the social and demographic variables studied (mother's age, ethnic group, intended duration of breastfeeding, social support) predicted the duration of breastfeeding. In a companion study, Neifert et al. (1988a) reported that gift packs containing formula presented upon hospital discharge had no observed effect on breastfeeding duration.

These results suggest that decisions on the timing of weaning are probably highly individual; the reasons for the decisions cannot accurately be determined from surveys.

Determinants of Breastfeeding Common to Different Subgroups

Some of the determinants of breastfeeding may be specific to selected segments of the U.S. population, whereas others are common to all. For example, maternal education is a common correlate of breastfeeding initiation and duration among blacks (Kurinij et al., 1988), whites (Rassin et al., 1984), and Hispanics (John and Martorell, 1989) and among mothers employed outside the home (Kurinij et al., 1989). Maternal education is also strongly related to attitudes toward breastfeeding (e.g., Dusdieker et al., 1985). However, the reasons for these associations are not clear. Formal education may be a measure of enculturation, since it provides exposure to, and promotes adoption of, major social values.

The studies reviewed above suggest that two categories of determinants influence the selection of the infant-feeding method: infant-feeding ideology and external constraints. Infant-feeding ideology includes beliefs about the efficacy of different feeding methods and values placed on breastfeeding as compared with bottle feeding. Such ideology has idiosyncratic components based on personal experience, but it also has more general components derived from affiliation with a particular ethnic group or membership in other subcultures. External constraints include factors such as separation from the infant resulting from commitments outside the home and the demands of household and family responsibilities. Particular combinations of these constraints may be common to groups that cut across ethnic boundaries, for example, mothers working outside the home, adolescent mothers, and economically disadvantaged mothers.