Exploring Intended Infant Feeding Decisions Among Low-Income Women

Abstract

Background:

Low-income women have the lowest rates of breastfeeding in the United States. Greater understanding of factors that predict intention to feed artificial breastmilk substitute is needed to inform the design and timing of interventions to promote breastfeeding among vulnerable women. This study aimed to identify demographic and reproductive characteristics and other factors associated with intent to feed artificial breastmilk substitute among low-income women.

Materials and Methods:

Data from 520 low-income women interviewed at 24–41 weeks of gestation during enrollment in a prenatal breastfeeding education intervention study were analyzed. Participant characteristics, reasons for feeding decision, and sources and types of information received were compared among women intending to feed only artificial breastmilk substitute and other women.

Results:

Most participants (95%) had already chosen an infant feeding method at the time of interview. There were no differences in plans to return to work by feeding plan. Women reporting intention to feed only artificial breastmilk substitute were less likely to report receiving information about the benefits of breastfeeding, how to breastfeed, and pumps and were more likely to cite personal preference and convenience as reasons for their decision. Women were more likely to intend to feed artificial breastmilk substitute if they had a previous live birth or had not breastfed a child, including the most recent.

Conclusions:

These findings suggest breastfeeding promotion should target women early and include sensitive, effective ways to promote breastfeeding among women who have not previously successfully breastfed. Breastfeeding history should be elicited, and plans to pump should be supported prenatally.

Introduction

Breastfeeding and the use of human milk are the accepted standards for infant nutrition.¹ Indeed, the use of artificial breastmilk substitute (formula) increases the risk of adverse health outcomes for both infants and mothers. Compared with breastfed infants, those who receive artificial breastmilk substitute have a higher risk of infections and sudden infant death syndrome, and women who do not breastfeed are at increased risk of breast and ovarian cancer.^1–7 Nationwide, only 60% of black women and 67% of low-income women (<100% of the federal poverty level) initiate breastfeeding.^8–10 Maternal intent to breastfeed is associated with both initiation and duration of breastfeeding.¹¹ Thus, interventions to increase intention are needed to improve rates of initiation and duration.

Factors shown to be associated with intent to breastfeed include race, ethnicity, immigration and acculturation, socioeconomic status, marital status, perceived social support, knowledge of, confidence about, and experience with breastfeeding, and employment status.^12–17 Low-income women are more likely than other groups to intend to feed artificial breastmilk substitute, and several of these other aforementioned risk factors persist within this population.^13–17 With the recent emphasis within the Special Supplemental Nutrition Program for Women, Infants and Children (WIC) to promote and support breastfeeding, identifying low-income women most at risk of feeding artificial breastmilk substitute and understanding their sources of information and reasons for feeding plan are necessary to design effective programs to increase their breastfeeding rates.

Women potentially receive information about infant feeding options from a variety of sources. Many women could benefit from information about breastfeeding prior to parturition, but few clinicians fully understand the significance of their advice to mothers regarding breastfeeding.^18,19 Elucidating the reasons women choose whether or not to breastfeed can also help inform educational efforts to promote breastfeeding, especially in those populations of women who are known to be least likely to intend to breastfeed.

Greater understanding of factors that predict intention to feed artificial breastmilk substitute among pregnant women is needed to inform the design and timing of interventions to promote intention to breastfeed and target particularly vulnerable women. The purpose of this study was to examine risk factors associated with intention to feed only artificial breastmilk substitute rather than any breastmilk in a sample of low-income women who may be eligible for the WIC program (gross income falls at or below 185% of the U.S. Poverty Income Guidelines),²⁰ who are known to be at increased risk of feeding artificial breastmilk substitute. In this cohort of low-income women, we identified characteristics associated with intent to feed artificial breastmilk substitute, sources of information on infant feeding, and reasons for infant feeding decision.

Materials and Methods

Overview

This study examined associations between characteristics of WIC-eligible women and their intended method for feeding their newborn infants. Baseline data from the Prenatal Education Video Study, a multisite, randomized, controlled intervention trial conducted at the University of Virginia Health System and the Virginia Commonwealth University Health System, were used. Prenatal Education Video Study participants provided data during prenatal enrollment interviews on demographic characteristics, infant feeding plans, reasons for infant feeding decisions, and sources of information about infant feeding. The study was approved by the Institutional Review Boards of the University of Virginia, the Virginia Commonwealth University, and the Virginia Department of Health.

Study population

Low-income, pregnant women at 24–41 weeks of gestation presenting for prenatal care during the study enrollment period (2009–2012) were eligible to participate. Women with multiple-gestation pregnancy, with any known contraindication to breastfeeding such as human immunodeficiency virus infection, drug use, or receipt of chemotherapy, and who were not proficient in English were excluded from the study.

Data collection

Medical records of pregnant women were screened daily for eligibility at participating prenatal clinics. Trained research assistants approached eligible women while they were waiting for routine care, explained the study, verified eligibility, and initiated the consent process. Enrollment questionnaires collecting baseline data were administered prior to randomization to the intervention group.

Study variables

Data regarding factors thought to affect infant feeding plans and practices, including sociodemographic characteristics (age, race, Hispanic ethnicity, marital status, and educational attainment), employment status (works outside of the home) and plans to return to work (if and when), relationship status (currently in a committed relationship), other adults living in the home (partner, parent, grandparent of the participant, or other), parity, and previous infant feeding experience (any breastfeeding), were collected during the baseline interview. Women were also asked about potential influences on feeding plan. Women were asked specifically, “What led you to choose [reported intended feeding method, i.e., breast, formula, or both]?” This question was intended to be asked as an open-ended question, with the research assistant recording the verbatim response and coding it in one of the following categories: health benefits, convenience, work, family, cost, cultural acceptability, personal preference, or other. (In practice, the respondents usually prompted the research assistant to show the list of categories from which they selected their responses.) They were also asked, “Have you received information on how to feed your baby?” If the participant responded affirmatively, she was asked “From whom or what?” Research assistants could prompt women by asking directly about the following sources: prenatal class, written materials, video, health professional, WIC, friend, relative, or other. If a participant responded “other,” she was asked to specify. The intended infant feeding method reported by the mother was classified as intention to feed any breastmilk (partial or full), intention to feed only artificial breastmilk substitute, or undecided feeding plan. This outcome variable was coded based on the participant's response to the question, “What are you going to feed your baby when he or she is born?” The response of the participant was recorded as “breastmilk only,” “formula only,” “both,” or “don't know.” We asked additional questions of women who reported an intention to feed breastmilk about their confidence in their ability to breastfeed (scale provided from which to choose: very unconfident, somewhat unconfident, neither confident nor unconfident, somewhat confident, very confident) and their plans for feeding breastmilk (“breastfeed directly,” “pumped breastmilk in a bottle,” “both direct and expressed milk feedings,” or “don't know”).

Data analysis

Descriptive statistics were computed for the various characteristics, as well as influences on feeding choice, by intention to breastfeed; chi-squared tests and analysis of variance were conducted for categorical and continuous variables, respectively. We also explored whether gestational age at the time of enrollment differed by feeding intention. Based on the unadjusted results, logistic regression models were fit to estimate the association between various characteristics and intention to feed only artificial breastmilk substitute versus intention to feed any breastmilk to determine if differences persisted while controlling for other factors. Because of the small number of women who had not yet chosen an infant feeding method, we excluded these women from the multivariable analysis. In addition, only women for whom we had complete data for the variables included in the final model were included in both the final univariable and multivariable analyses. All analyses were performed in SAS version 9.3 software (SAS Institute, Inc., Cary, NC), controlling for the design variable, study site.

Results

In total, 2,875 women were screened in four clinics at the two study sites. After exclusions, 1,581 women were eligible to participate in the study. We were able to approach 816 of these eligible women and invite them to participate in the study. Of these, 522 (64%) were enrolled in the study. Two women did not respond to the question regarding intended infant feeding method, for a final population of 520 women included in these analyses.

Intention to breastfeed

Over two-thirds of the women reported the intention to feed their infant breastmilk (72% of whom intended to do so exclusively and 28% of whom intended to do so in combination with artificial breastmilk substitute), 27% reported the intention to feed their infants only artificial breastmilk substitute, and 5% had not decided at the time of enrollment how they intended to feed their infant (Table 1) (not all data shown).

Table 1.

Characteristics of Women Enrolled in the Prenatal Education Video Study by Intended Method of Infant Feeding

	Any breastmilk (n=352)	Artificial breastmilk substitute only (n=142)	Undecided (n=26)	p value
Age (years)	25 (5.9)	25 (4.9)	23 (4.2)	0.15
BMI (kg/m²)	32 (8.6)	32 (8.8)	33 (9.6)	0.95
Race/ethnicity				0.01
Non-Hispanic, white	145 (41)	59 (42)	10 (38)
Non-Hispanic, black	147 (42)	75 (53)	14 (54)
Non-Hispanic, Asian	9 (3)	1 (1)	1 (4)
Non-Hispanic, other	13 (4)	2 (1)	1 (4)
Hispanic	37 (11)	4 (3)	0 (0)
Not reported	1	1	0
Educational attainment
College graduate	23 (7)	2 (1)	1 (4)	0.02
Some college	99 (28)	30 (21)	3 (12)
Technical training	10 (3)	2 (1)	0
High school diploma or GED	159 (45)	69 (49)	13 (50)
Some high school or less	61 (17)	39 (27)	9 (35)
Marital status
Married	101 (29)	27 (19)	3 (12)	0.07
Separated	12 (3)	10 (7)	1 (4)
Divorced	16 (5)	3 (2)	0
Widowed	1 (0)	0	0
Never married	222 (63)	102 (72)	22 (85)
Not in a committed relationship	57 (21)	30 (25)	4 (17)	0.57
Other adults living in the home
Partner	185 (53)	69 (49)	13 (50)	0.66
Parent	115 (33)	49 (35)	9 (35)	0.94
Grandparent	29 (8)	7 (5)	3 (12)	0.33
Other	76 (22)	25 (18)	8 (31)	0.27
No others living in the home	29 (8)	18 (13)	3 (12)	0.28
Works outside of home	144 (41)	60 (43)	11 (42)	0.95

Data are number (percentage) or mean (SD) values.

BMI, body mass index; GED, general educational development test.

Characteristics associated with intent to feed only artificial breastmilk substitute

Women intending to feed only artificial breastmilk substitute were significantly more likely to have less education and differed significantly by race/ethnicity. Neither age of the women enrolled nor body mass index differed significantly by feeding intention, and no significant differences were observed for being in a committed relationship or living with other adults in the home. Nearly half of women in the study reported working outside of the home; employment status did not differ by feeding method (Table 1). Among mothers who report working, 97% overall reported the intention to return to work following the birth of their infant. The mean number of days after birth that women planned to return to work was 50 (SD=41), 48 (SD=27), and 44 (SD=14) for women intending to feed breastmilk, artificial breastmilk substitute, and undecided feeding plan, respectively (p=0.84). These women planned to work a mean of 31 (SD=10), 32 (SD=9), and 33 (SD=8) hours per week, respectively, upon return to the workplace (p=0.71) (data not shown).

Breastfeeding confidence

Among women intending to feed their infant breastmilk, over three-quarters reported being somewhat or very confident in their ability to breastfeed (Table 2). On average, they planned to breastfeed for 6 months. The majority planned to use a combination of at-breast and expressed breastmilk feedings.

Table 2.

Plans for and Confidence About Breastfeeding Among Women Intending to Feed Breastmilk in the Prenatal Education Video Study

	Frequency (n=352)
Intended duration of breastfeeding (days)	181 (144)
Confidence in ability to breastfeed
Very unconfident	17 (5)
Somewhat unconfident	20 (6)
Neither confident nor unconfident	42 (12)
Somewhat confident	97 (28)
Very confident	168 (48)
Don't know	5 (1)
Not reported	3
Plan for breastmilk feeding
Feed at breast	92 (26)
Feed expressed breastmilk	32 (9)
Feed at breast and expressed breastmilk	210 (60)
Don't know	17 (5)
Not reported	1

Data are number (percentage) or mean (SD) values.

Gestational age and parity

Although the gestational age at which women initiated prenatal care and the gestational age at which they enrolled in the study did not differ by their intended feeding plan, parity did. Significantly more women who intended to feed only artificial breastmilk substitute had a previous live birth. However, among women who had a previous live birth, those who had breastfed a child, including the most recent child, were less likely to intend to feed only artificial breastmilk substitute (Table 3).

Table 3.

Influences on Feeding Plans Among Women Enrolled in the Prenatal Education Video Study by Intended Method of Infant Feeding

	Any breastmilk (n=352)	Artificial breastmilk substitute only (n=142)	Undecided (n=26)	p value
Gestational age (weeks) at
Initiation of PNC	14.2 (6.8)	14.7 (7.0)	16.2 (8.5)	0.36
Study enrollment	30.4 (4.7)	30.7 (4.5)	28.7 (3.4)	0.14
Parity
Previous live birth	176 (51)	105 (74)	13 (50)	< 0.01
Previous feeding history^a
Ever breastfed a child	136 (77)	26 (25)	4 (30)	< 0.01
Ever breastfed most recent child	130 (74)	19 (18)	4 (30)	< 0.01
Reasons for infant feeding plan
Health benefits	255 (73)	6 (4)	—	< 0.01
Convenience	42 (12)	33 (23)	—	< 0.01
Work	9 (3)	8 (6)	—	0.24
Family	25 (7)	6 (4)	—	0.52
Cost	54 (15)	3 (2)	—	< 0.01
Personal preference	61 (17)	73 (52)	—	< 0.01
Sources of information on infant feeding
Prenatal class	28 (8)	3 (2)	0	0.02
Written material	59 (17)	16 (11)	2 (8)	0.18
Video	21 (6)	5 (4)	0	0.26
Health professional	81 (23)	30 (21)	4 (15)	0.63
WIC	141 (40)	66 (46)	16 (62)	0.07
Friend	16 (5)	3 (2)	0	0.26
Relative	22 (6)	12 (8)	1 (4)	0.57
Other	31 (9)	17 (12)	1 (4)	0.35
Type of information
Benefits of breastfeeding	213 (61)	61 (43)	16 (62)	< 0.01
How to breastfeed	188 (53)	49 (35)	13 (50)	< 0.01
Timing of feedings	100 (28)	40 (28)	5 (19)	0.45
Best type of formula to use	68 (19)	62 (44)	4 (15)	< 0.01
Type of bottles, nipples, etc. to use	64 (18)	40 (28)	5 (19)	0.02
Breast pump and assorted accessories	104 (30)	27 (19)	5 (19)	0.03
Spitting up, reflux	57 (16)	28 (20)	3 (12)	0.37

Data are number (percentage) or mean (SD) values.

Among the 176, 105, and 13 women who had a previous live birth who intended to feed any breastmilk, artificial breastmilk substitute, and unknown, respectively.

PNC, prenatal care; WIC, Special Supplemental Nutrition Program for Women, Infants and Children.

Reasons for feeding plan

Among the women who reported an intended feeding method, significantly different reasons for choosing that method were selected from the categories provided. Women intending to feed only artificial breastmilk substitute were more likely to choose personal preference and convenience. Women intending to feed any breastmilk were more likely to choose health benefits and cost (Table 3).

Sources and types of infant feeding information

Most reported sources of information about infant feeding did not differ. The most commonly cited sources were WIC and a health professional. Very few women reported prenatal classes as a source of information regarding infant feeding. Sources of information did not differ significantly by the gestational age at which women were enrolled except prenatal class (p=0.0001) and written material (p=0.018). For both, in general, a greater proportion of women enrolling later in their pregnancy reported these sources (data not shown).

Most of the types of information women reported receiving differed significantly, however. Women intending to feed only artificial breastmilk substitute were more likely to report receiving information regarding the best types of formula and bottles to use. Women intending to feed any breastmilk and with unknown plans were more likely to report receiving information about the benefits of breastfeeding and how to breastfeed. In addition, women intending to breastfeed were also significantly more likely to report receiving information about breast pumps (Table 3). With the exception of the best formula to use, the types of information that women received did not differ significantly by the gestational age at which they entered the study (data not shown).

Factors predicting breastfeeding intention

The only demographic differences that persisted after controlling for other factors were education and race/ethnicity. Hispanic women were 81% less likely to intend to feed artificial breastmilk substitute than non-Hispanic white women, and women with less than a high school diploma or general educational development test were 84% more likely to intend to feed artificial breastmilk substitute than breastmilk (Table 4).

Table 4.

Risk Factors for and Protective Factors Against Intention to Feed Only Artificial Breastmilk Substitute Among Women Enrolled in the Prenatal Education Video Study

	Ratio of odds of intention to feed only artificial breastmilk substitute to odds of intention to feed any breastmilk
	Unadjusted ^a	Adjusted ^b
Race/ethnicity
Non-Hispanic, white	1.00	1.00
Non-Hispanic, black	1.32 (0.85, 2.05)	1.27 (0.80, 2.00)
Non-Hispanic, other	0.36 (0.10, 1.27)	0.39 (0.11, 1.38)
Hispanic	0.31 (0.11, 0.93)^c	0.29 (0.10, 0.87)^c
≤high school diploma or GED	1.86 (1.18, 2.94)^c	1.84 (1.14, 2.97)^c
Not married	1.60 (0.98, 2.60)	1.34 (0.79, 2.26)
Not in committed relationship	1.38 (0.82, 2.34)	1.15 (0.66, 2.01)
Works outside of home	1.03 (0.68, 1.56)	1.09 (0.71, 1.68)
Types of information
Benefits of breastfeeding	0.46 (0.31, 0.70)^c	0.79 (0.16, 0.95)^c
How to breastfeed	0.46 (0.30, 0.70)^c	0.43 (0.28, 0.67)^c
Timing of feedings	0.99 (0.63, 1.55)	0.88 (0.55, 1.41)
Best type of formula to use	3.23 (2.09, 5.00)^c	3.23 (2.06, 5.08)^c
Type of bottles, nipples, etc. to use	1.79 (1.12, 2.87)	1.91 (1.18, 3.11)^c
Breast pump and assorted accessories	0.52 (0.31, 0.85)^c	0.50 (0.30, 0.83)^c
Spitting up, reflux	1.21 (0.72, 2.04)	1.16 (0.68, 1.98)
Feeding history
Previous live birth	2.44 (1.57, 3.80)^c	2.76 (1.72, 4.42)^c
Never breastfed a child^d	11.3 (6.21, 20.6)^c	11.5 (6.00, 22.1)^c
Did not breastfed most recent child^d	13.9 (7.41, 26.1)^c	14.2 (7.12, 28.2)^c

Data are odds ratio (95% confidence interval).

Adjusted for design variable, study site.

Adjusted for the other variables shown in the table.

Significant factor.

Among women who had a previous live birth; sample size of 265.

GED, general educational development test.

Differences in types of information received also persisted. Women who reported receiving information about the benefits of breastfeeding were about 20% less likely to intend to feed only artificial breastmilk substitute instead of any breastmilk. In addition, those who received information on how to breastfeed or information about breast pumps were half (or less) as likely to report intention to feed only artificial breastmilk substitute (Table 4).

Parity and infant feeding history remained strongly associated with intention to feed artificial breastmilk substitute. Women who had a previous live birth (versus nulliparous) were nearly three times as likely to intend to feed artificial breastmilk substitute. Among those who had a previous live birth, women who had never breastfed were more than 11 times as likely to intend to feed artificial breastmilk. The magnitude of the association between not breastfeeding the most recent child and intention to feed only artificial breastmilk substitute was even larger (Table 4).

Discussion

In this cohort of low-income women receiving prenatal care, infant feeding plans had been determined by 95% of women at the time of enrollment during the third trimester of pregnancy. Nearly a third of these women had no intention to feed any breastmilk to their infants. The women in this study most likely to intend to feed artificial breastmilk substitute were non-Hispanic black, had less than a high school education, and were never married or were separated. Among multiparous women, the strongest risk factor for intention to feed only artificial breastmilk was never breastfeeding a previous child.

These results are consistent with previous research, and we confirm many of the well-documented factors affecting breastfeeding intent, such as education, marital status, and race/ethnicity.^1–7,20,21 Thus, many factors shown to be associated with breastfeeding intent in the broader population of U.S. women are also associated with intent among lower–income women. Maternal confidence and self-efficacy also have been shown to be associated with breastfeeding.²² In the current study, however, only women who reported intention to feed any breastmilk were asked about confidence, so the association between confidence and intention could not be examined. However, unlike in previous research, plans to return to work did not affect intended feeding methods in this population.^23–26 Economic pressures and work preferences appear to exert a different influence on feeding plans than among more economically diverse populations.

In this population, the mean duration of planned breastfeeding was only 6 months, half of the time recommended by the American Academy of Pediatrics¹ and the Academy of Breastfeeding Medicine.²⁷ Furthermore, it is unlikely that these women will meet their feeding goals, as only about one-third of U.S. women overall reach their feeding duration goals.²⁸ The first step to increasing the duration of breastfeeding is to provide perinatal education and public health approaches that target increasing the time during which women intend to feed breastmilk. This education, coupled with providing information to create realistic expectations among women about breastfeeding and support to overcome both the challenges of establishing breastfeeding early and maintaining it through transitions such as returning to work, could help women achieve longer durations of breastfeeding. Given that prenatal education aimed at influencing feeding plans in combination with breastfeeding support has been shown to increase initiation, these findings suggest that the same is likely among low-income women.²⁹

Having no previous experience with breastfeeding was a risk factor for intention to feed artificial breastmilk substitute. To facilitate breastfeeding promotion and support, taking a detailed breastfeeding history (such as number of previous attempts, goals, challenges, and total number of breastfeeding months) could be integrated into the patient's prenatal record. For example, BPA(L) is a tool currently under development at the University of Virginia that quantifies the number of infants the mother attempted to breastfeed (B), the number of infants for which the mother had problems breastfeeding (P), the number of infants the mother was subjectively able to breastfeed (A), and the number of lifetime months of breastfeeding for the mother (L). The use of this abbreviation may facilitate use of the tool because it mirrors TPAL, an abbreviation routinely used to report obstetric history (term births, preterm births, abortions, and living children). However, any method developed for gathering more detailed breastfeeding history would likely be helpful in identifying those at risk when targeting interventions. Our data show that multiparous women with no previous breastfeeding experience are less likely to intend to breastfeed, although we did not collect information as to why. Studies have shown that some women who have previously fed artificial breastmilk substitute will choose to breastfeed subsequent children, suggesting that interventions can and should be developed to target this specific group of women.³⁰

Consistent with previous work, we confirmed that information about health benefits of breastfeeding is associated with women intending to breastfeed.^1–7,21,31 Although these data are cross-sectional, educational interventions and media campaigns that highlight health benefits appear to have affected the proportion of women intending to breastfeed. Such campaigns, however, may not provide the best approach to reaching the remaining group of women who intend to feed only artificial breastmilk substitute. The majority of women in this study intending to feed artificial breastmilk substitute reported convenience or personal preference as reasons behind their intended feeding choice. Future educational efforts should promote the ease of feeding breastmilk, and future research should use qualitative methods such as focus groups or motivational interviewing to uncover reasons, such as cultural norms, for this personal preference among at-risk populations.

Regardless of infant feeding plan, the two most commonly cited sources of information regarding infant feeding were WIC and healthcare professionals, highlighting the role these sources can and should play in educating women about breastfeeding. Of note is that few women in this population reported receiving information from traditional sources of prenatal education, namely, classes and written materials. These data support the need for the development of innovative strategies that augment traditional provider messages during prenatal care for delivering breastfeeding education to underserved populations and younger mothers, such as Text for Baby.³²

Our findings suggest that women have already made decisions regarding their infant feeding plan by the third trimester of pregnancy. In this study, participants were at least 24 weeks of gestational age at enrollment, and 95% of women had already made a decision regarding how they intended to feed their infant. Breastfeeding promotion needs to occur as early as possible in the prenatal period. Moreover, breastfeeding promotion likely needs to occur prior to conception to have a positive impact on high-risk women, including public health approaches to shift cultural norms about breastfeeding. Although the number of undecided women in this study was small, identifying those who “don't know” and the characteristics of those who could potentially be encouraged to breastfeed may be valuable. Because women were only asked their intended feeding plan once prenatally, it is unknown whether women changed their plan during pregnancy. From clinical experience, we know that some women change their minds about infant feeding choice even up to the time of delivery. A prospective study that queries women repeatedly during the prenatal period could provide insight as to what extent women can be influenced to change their infant feeding plan and what specific interventions (e.g., motivational interviewing) might be effective at changing their feeding plans.

The majority of women in this study who intended to breastfeed also reported intending to feed pumped breastmilk from a bottle. Because some WIC programs now offer free breast pumps to eligible women, assessing intent to pump in this population is potentially important. One study showed that women in WIC who requested a breast pump and received it immediately requested formula later than those who do not receive a pump in a timely manner.³³ Future studies should assess the effectiveness of offering free pumps and the timing of their availability.

This study has several strengths, including a fairly large sample size of high-risk women. The study included return-to-work plans as well as intention to pump breastmilk. These results further define risk factors for artificial breastmilk substitute feeding in an already at-risk population in need of targeted interventions. The limitations include a geographically restricted population of only English speakers. This study enrolled only a small number of participants who were not either non-Hispanic white or black. Furthermore, given the cross-sectional nature of the study, the temporal relationship between the risk factors that we identified and the decision regarding intended feeding plan is unclear. For example, it is unknown whether the information related to infant feeding received by women prenatally is a reflection of their reported feeding plan (i.e., the provider presented different types of information because of the reported feeding method) or whether the information influenced their feeding plan decisions. In addition, women may have recalled information provided to them differently depending on their intended feeding plan.

Conclusions

We aimed to identify risk factors for intention to feed artificial breastmilk substitute among a cohort of WIC-eligible, pregnant women with two long-term goals: to inform the development of interventions to promote breastfeeding and to identify subgroups to which to target such interventions in this high-risk population. We also assessed additional potential influences on feeding plans, including sources of information and reasons for feeding plan choice. Based on these findings, breastfeeding promotion should target women early in pregnancy and prior to conception. Preconception community outreach and individual prenatal counseling should not only include the benefits of breastfeeding but also aim to normalize breastfeeding and highlight its convenience. A transformation in breastfeeding promotion among preconception and prenatal healthcare providers and WIC offices is needed, given they have been shown to have an impact on the choice to breastfeed. Additional study is indicated to identify sensitive ways to effectively promote breastfeeding among women who have not previously breastfed their children.

Footnotes

Acknowledgments

This study was supported by a contract with the Virginia Department of Health.

Disclosure Statement

No competing financial interests exist.

References

Section on Breastfeeding. Breastfeeding and the use of human milk. Pediatrics, 2012; 129:e827–e841.

Stuebe

, Schwarz

. The risks and benefits of infant feeding practices for women and their children. J Perinatol, 2010; 30:155–162.

Gartner

, Morton

, Lawrence

, et al. Breastfeeding and the use of human milk. Pediatrics, 2005; 115:496–506.

, Chung

, Raman

, et al. Breastfeeding and maternal and infant health outcomes in developed countries. Evid Rep Technol Assess, 2007;(153):1–186.

Duijts

, Jaddoe

, Hofman

, et al. Prolonged and exclusive breastfeeding reduces the risk of infectious diseases in infancy. Pediatrics, 2010; 126:e18–e25.

Ladomenou

, Kafatos

, Tselentis

, et al. Predisposing factors for acute otitis media in infancy. J Infect, 2010; 61:49–53.

Godfrey

, Lawrence

. Toward optimal health: The maternal benefits of breastfeeding. J Womens Health (Larchmt), 2010; 19:1597–1602.

Centers for Disease Control and Prevention. Progress in increasing breastfeeding and reducing racial/ethnic differences—United States, 2000–2008 births. MMWR Morb Mortal Wkly Rep, 2013; 62:77–80.

Ryan

, Zhou

. Lower breastfeeding rates persist among the Special Supplemental Nutrition Program for Women, Infants, and Children participants, 1978–2003. Pediatrics, 2006; 117:1136–1146.

10.

Centers for Disease Control and Prevention. National Immunization Survey. Atlanta: Centers for Disease Control and Prevention, U.S. Department of Health and Human Services, 2007. www.cdc.gov/breastfeeding/data/NIS_data/2007/socio-demographic_any.htm (accessed June 23, 2014).

11.

Bai

, Middlestadt

, Peng

, et al. Predictors of continuation of exclusive breastfeeding for the first six months of life. J Hum Lact, 2010; 26:26–34.

12.

Merewood

, Brooks

, Bauchner

, et al. Maternal birthplace and breastfeeding initiation among term and preterm infants: A statewide assessment for Massachusetts. Pediatrics, 2006; 118:e1048–e1054.

13.

Lee

, Gould

. Factors influencing breast milk versus formula feeding at discharge for very low birth weight infants in California. J Pediatr, 2009; 155:657–662.e651–e652.

14.

Mitra

, Khoury

, Hinton

, et al. Predictors of breastfeeding intention among low-income women. Matern Child Health J, 2004; 8:65–70.

15.

Hill

, Arnett

, Mauk

. Breast-feeding intentions among low-income pregnant and lactating women. Am J Health Behav, 2008; 32:125–136.

16.

McInnes

, Love

, Stone

. Independent predictors of breastfeeding intention in a disadvantaged population of pregnant women. BMC Public Health, 2001; 1:10.

17.

Chertok

, Luo

, Culp

, et al. Intent to breastfeed: A population-based perspective. Breastfeed Med, 2011; 6:125–129.

18.

Shealy

, Li

, Benton-Davis

, et al. The CDC Guide to Breastfeeding Interventions. Atlanta: Centers for Disease Control and Prevention, U.S. Department of Health and Human Services, 2005.

19.

Taveras

, Capra

, Braveman

, et al. Clinician support and psychosocial risk factors associated with breastfeeding discontinuation. Pediatrics, 2003; 112:108–115.

20.

Virginia Department of Health. Eligibility. May 20, 2014. www.vahealth.org/dcn/General%20Info/eligibility.htm (accessed May 22, 2014).

21.

Kramer

, Kakuma

. Optimal duration of exclusive breastfeeding. Cochrane Database Syst Rev, 2012; 8:CD003517.

22.

Blyth

, Creedy

, Dennis

, et al. Effect of maternal confidence on breastfeeding duration. Birth, 2002; 29:178–284.

23.

Arora

, McJunkin

, Wehrer

, et al. Major factors influencing breastfeeding rates: Mother's perception of father's attitude and milk supply. Pediatrics, 2000; 106:E67.

24.

Hawkins

, Griffiths

, Dezateux

, et al. Maternal employment and breast-feeding initiation: Findings from the Millennium Cohort Study. Paediatr Perinat Epidemiol, 2007; 21:242–247.

25.

Ogbuanu

, Glover

, Probst

, et al. The effect of maternity leave length and time of return to work on breastfeeding. Pediatrics, 2011; 127:e1414–e1427.

26.

Noble

; ALSPAC Study Team. Avon Longitudinal Study of Pregnancy and Childhood. Maternal employment and the initiation of breastfeeding. Acta Paediatr, 2001; 90:423–428.

27.

Academy of Breastfeeding Medicine Board of Directors. Position on breastfeeding, Breastfeed Med, 2008; 3:267–270.

28.

Perrine

, Scanlon

, Li

, et al. Baby-Friendly hospital practices and meeting exclusive breastfeeding intention. Pediatrics, 2012; 130:54–60.

29.

Guise

, Palda

, Westhoff

, et al. The effectiveness of primary care-based interventions to promote breastfeeding: Sytematic evidence review and meta-analysis for the US Preventive Services Task Force. Ann Fam Med, 2003; 1:70–78.

30.

Kruse

, Denk

, Feldman-Winter

, et al. Longitudinal patterns of breastfeeding initiation. Matern Child Health J, 2006; 10:13–18.

31.

Stuebe

, Bonuck

. What predicts intent to breastfeed exclusively? Breastfeeding knowledge, attitudes, and beliefs in a diverse urban population. Breastfeed Med, 2011; 6:413–420.

32.

Evans

, Wallace

, Snider

. Pilot evaluation of the text4baby mobile health program. BMC Public Health, 2012; 12:1031.

33.

Meehan

, Harrison

, Afifi

, et al. The association between an electric pump loan program and the timing of requests for formula by working mothers in WIC. J Hum Lact, 2008; 24:150–158.