Three-quarters of American women who give birth are encouraged by a doctor or nurse to breastfeed, and three-fourths of those who receive such encouragement nurse for some period of time. When background characteristics that may influence a woman's likelihood of breastfeeding are taken into account, women who receive encouragement from a health care provider are more than four times as likely to breastfeed as are those who receive no such encouragement; the impact of a provider's encouragement is similar regardless of women's race or other characteristics. These findings echo results of previous research, but whereas earlier work in this area was based primarily on small convenience samples, the study that produced these estimates was based on a large, randomly selected sample of women from throughout the United States.1
The analysis was part of a larger study, conducted between July 1995 and January 1996, that examined the needs and behaviors of parents with children younger than three. Researchers recruited parents to participate in a telephone survey that included questions on whether doctors and nurses encouraged breastfeeding and the length of time that the mother breastfed, as well as questions on respondents' demographic and socioeconomic characteristics. The investigators assessed the importance of provider encouragement in bivariate and regression analyses based on the responses of 1,229 women who completed the survey.
Seventy-three percent of women in the sample were white, 13% were black and 14% were Hispanic. Respondents were predominantly married (81%) and aged 25 or older (68%). Half had more than a secondary education, one-third had only completed high school and the remainder had less schooling. Twenty-eight percent of women had an annual income of less than $20,000, and 43% reported a yearly income of $40,000 or more. One-third lived in the South, and the rest were about evenly divided among the Northeast, the North Central region and the West.
Overall, 34% of participants had not breastfed, 12% had nursed for less than one month and 53% had done so for longer. In the bivariate analyses, the proportions who had ever breastfed were highest (69-87%) among women older than 25; married women; whites and Hispanics; those with the highest levels of education and income; residents of the West; women not receiving government assistance; those who had attended childbirth classes; and those who had received provider encouragement to nurse.
The majority of respondents (73%) said that a physician or nurse had encouraged them to breastfeed; this proportion did not vary significantly among women of different characteristics. Whereas 75% of these women initiated breastfeeding, only 43% of those who had received no encouragement from a health care provider ever nursed, and the difference was statistically significant.
Regression analyses controlling for factors that may influence a woman's decision to nurse revealed that those who had been encouraged to breastfeed were more than four times as likely as others to have done so (relative risk, 4.4). Other factors were also associated with an increased likelihood of breastfeeding, but the effects were generally smaller: For women who had attended childbirth classes and those who were 25-34 years old, married or divorced, white or Hispanic and residents of the West, relative risks ranged from 1.3 to 2.9. Increasing levels of education were associated with increasing likelihoods of breastfeeding, and women who had graduated from college were 6.8 times as likely as those with less than a high school education to have nursed.
To assess the impact of a provider's encouragement to breastfeed on women of different characteristics, the researchers repeated the regression analysis, stratifying the sample according to race or ethnicity, educational level, income, marital status and age-group. Results showed that encouragement by a physician or nurse increased the likelihood of breastfeeding for virtually all groups. Relative risks ranged from 3.2 for women with less than a high school education to 11.0 for single women; for most groups, women were about 4-6 times as likely to have breastfed if they had received encouragement as if they had not. Divorced women appeared to have an increased likelihood of having breastfed if a provider had encouraged them to nurse, but because the number of divorced women in the sample was small, the relative risk (3.8) did not achieve statistical significance.
The researchers also conducted logistic regression analyses to determine whether a provider's encouragement affected the duration of breastfeeding. These calculations revealed no significant associations.
In the researchers' view, "obstetricians, midwives, and obstetric nurses have a special responsibility and capacity to promote breastfeeding." The findings from this analysis, the investigators comment, highlight the importance of ensuring that providers have the training they need to do so. Furthermore, they note, the results suggest that interventions promoting breastfeeding might have a "particularly strong influence" if they target groups of women who are traditionally unlikely to nurse.--D. Hollander
1. Lu MC et al., Provider encouragement of breast-feeding: evidence from a national survey, Obstetrics & Gynecology, 2001, 97(2):290-295.