Breastfeeding Among Inner-City Women: From Intention Before Delivery to Breastfeeding at Hospital Discharge

Abstract

Objective:

This study assessed the degree to which women's intention to breastfeed prior to delivery translates to actual breastfeeding at hospital discharge and to investigate predictors of breastfeeding in a minority inner-city population.

Study Design and Methods:

We conducted a retrospective cohort study of consecutive mother–infant dyads born July–September 2010 and discharged from the nursery at an academic community hospital in Philadelphia.

Results:

The demographics of the 578 women who participated included a mean age of 25.3 years (SD 6.1), 61% African American and 18% Latina, 85% covered by Medicaid, and a mean postpartum hospital stay of 2.3 days. Overall, 60% expressed an intention to breastfeed prior to delivery (exclusively or with formula), but only 50% were breastfeeding at discharge. Of those who intended to breastfeed (exclusively or with formula), 75% were breastfeeding at discharge. Of those who intended to breastfeed exclusively, 40% were doing so at discharge. Of those who intended to bottle feed, 11% were breastfeeding at discharge. In multivariable analysis, older mothers and those with lower parity were more likely to breastfeed at discharge and also to breastfeed exclusively, controlling for ethnicity, parity, insurance, pregravida body mass index, score on the Edinburgh Postpartum Depression Scale, type of delivery, infant birth weight and gestational age.

Conclusions:

In a minority inner-city population, only three in four women who intended to breastfeed prior to delivery were breastfeeding at hospital discharge. However, one in 10 women previously not intending to breastfeed did so. Strategies are needed to promote and strengthen women's intention to breastfeed and to help women's breastfeeding outcomes meet their intentions.

Introduction

Breastfeeding is the recommended method of feeding for newborns.^1,2 Nationwide, only 74.6% of women initiate breastfeeding, a figure below the Healthy People 2020's target of 82%.³ Disparities in breastfeeding are attributed to demographic and cultural factors; the lowest rates are among African American women and women of lower educational and socioeconomic status.^2,4

Across population groups, a woman's prenatal intention to breastfeed is the most consistent predictor of initiation and duration of breastfeeding.^5,6 Most women decide how to feed their infant before they are pregnant or early in their pregnancy.⁷ Women who express favorable intentions to breastfeed are more likely to initiate breastfeeding.⁶ Initiation of breastfeeding in the immediate postpartum period is critical to the establishment of successful lactation.

Several studies have examined the relationship between intention to breastfeed and actual breastfeeding at 1 week of age.^6,8,9 However, little is known about breastfeeding patterns—from intention to actual practice—in the immediate postpartum period, a vulnerable period of breastfeeding initiation.¹⁰ To our knowledge, no study has reported on the path from prenatal intention to breastfeeding at hospital discharge. A better understanding of the pathway from intention to actual initiation of breastfeeding could assist in the development of strategies to improve breastfeeding rates. This is particularly important in population groups with low breastfeeding rates.

The objective of this study was to assess the degree to which a woman's intention to breastfeed prior to delivery translates to actual breastfeeding at the time of hospital discharge. A secondary objective was to investigate predictors of breastfeeding in a minority inner-city population.

Subjects and Methods

Study design and study population

We conducted a retrospective cohort study of consecutive mothers of infants who were born July–September 2010 and discharged from the nursery at Albert Einstein Medical Center (Philadelphia, PA), an academic medical center affiliated with a medical school, where residents and medical students participate in patient care. Albert Einstein Medical Center delivers 3,200 babies a year, predominantly from low-income, minority women from the northern census tracts of Philadelphia. Newborns are admitted to the term nursery if they satisfy the following criteria: gestational age ≥36 weeks, birth weight ≥2,000 g, no major malformations, and clinically stable. Over 90% of the women delivering at Albert Einstein Medical Center are enrolled in the Special Supplemental Nutrition Program for Women, Infants and Children. Mothers and newborns are discharged from the hospital on average 2.3 days after delivery. Albert Einstein Medical Center has a full-time lactation consultant, and infants room in. The study was approved by the Institutional Review Board as a quality improvement project with the purpose of collecting baseline data to begin improving hospital practices.

Data collection

Newborns discharged from the nursery during the study period were identified from the daily list of discharges and were all included in the study. We recorded maternal, obstetric, and infant characteristics, including maternal age, ethnicity, parity, insurance, pregravida body mass index, and score on the Edinburgh Postpartum Depression Scale, which was administered the day after delivery. We also collected data on type of delivery and infant birth weight and gestational age.

Prior to delivery, nurses in the labor and delivery suite asked the mothers, “How do you plan to feed your infant?” Answers were recorded as formula feeding, breastfeeding plus formula, or breastfeeding exclusively. Data were collected on type of infant feeding at discharge from the nursing flow sheets in the infant record.

Statistical analyses

In total, 666 infants were born and discharged from the nursery during the study period. The analysis was restricted to the 578 women who had documentation of predelivery feeding intention in their record. The study outcome measure was breastfeeding at discharge.

To assess the degree of change from intention to breastfeed to breastfeeding at discharge, we used a sign test on paired observations from the same woman. We investigated predictors of breastfeeding at discharge using a single multivariable ordinal logistic regression model with the outcome variable having three ranks: formula feeding (no breastfeeding), breastfeeding plus formula, and breastfeeding exclusively. We included maternal, obstetric, and infant covariates in a forced-entry multivariable ordinal logistic regression model. Adjusted odds ratios (ORs) and 95% confidence intervals assessed the strength of the associations. The multivariable regression passed the test of constant ORs across responses categories (proportional odds) for each covariate (p=0.54). Thus we were able to use a single OR to describe the association of any breastfeeding compared with none as well as exclusive breastfeeding compared with breastfeeding and bottle feeding for each covariate. A p value of <0.05 was considered statistically significant. Data were analyzed using Stata version 11 (Stata Corp., College Station, TX).

Results

Description of sample

The study group comprised minority and low-income, predominantly African American women (Table 1). Most women were covered by Medicaid.

Table 1.

Maternal, Obstetric, and Infant Characteristics of the Study Group

Characteristic	Value
Total n	578
Maternal age (years)	25.3 (6.1)
Ethnicity
African American	350 (60.5%)
White	71 (12.3%)
Latina	102 (17.7%)
Other	55 (9.5%)
Parity	1.14 (1.32)
Delivery
Vaginal	390 (67.5%
Cesarean section	188 (32.5%)
Insurance
Medicaid	491 (85%)
Other	87 (15%)
EPDS score	4.6 (1.4)
Gestational age (weeks)	39 (1.4)
Birth weight (g)	3,194 (475)

Data are mean (SD) values or n (%) as indicated.

EPDS, EPDS, Edinburgh Postpartum Depression Scale.

Pathway from intention to breastfeeding at discharge

Overall, 348 of 578 (60%) women expressed intention to breastfeed (exclusive or with formula) (Fig. 1), 54% initiated breastfeeding (data not shown), and 288 of 578 (50%) were breastfeeding (exclusive or with formula) at discharge (p<0.0001 comparing intent to discharge breastfeeding). Nearly half (279 of 578 [48%]) of women expressed an intention to breastfeed exclusively (Fig. 1), but only 123 of 578 (21%) were breastfeeding exclusively at discharge (p<0.0001).

FIG. 1.

Paired analysis of maternal predelivery feeding intent and infant feeding at discharge from the nursery.

Of the women who intended to formula feed, 11% were breastfeeding at discharge, one-third of them exclusively.

How predictive was intent to breastfeed of breastfeeding at discharge? Among the women who expressed an intention to breastfeed prior to delivery, 75% were breastfeeding at discharge. Among those with intention to breastfeed exclusively, 40% were breastfeeding exclusively at discharge. Of those with intention to formula feed, 89% were doing so at discharge.

Multivariable analysis

Older women were more likely to breastfeed and also to breastfeed exclusively at discharge (both p<0.001) (Table 2). After controlling for age and the other covariates, mothers with higher parity were less likely to breastfeed and also to breastfeed exclusively at discharge (OR=0.80, p=0.002). There was no statistically significant product interaction between age and parity (p=0.39) so that neither modified the effect of the other on breastfeeding at discharge (all p≥0.10). Women with higher body mass index, women delivering vaginally, and those on Medicaid were marginally more likely to breastfeed and to breastfeed exclusively, but results did not reach statistical significance. There were no statistically significant differences in breastfeeding outcomes at discharge by ethnicity, Edinburgh Postpartum Depression Scale scores, infant weight, or gestational age. The multivariable ordinal logistic regression satisfied the proportional odds assumption with p=0.54, indicating the same relationship held between each covariate and both breastfeeding at discharge and exclusive breastfeeding.

Table 2.

Multivariable Ordinal Logistic Regression Model with Three Ranked Infant Feeding Categories at Discharge: Formula Feeding, Breastfeeding and Formula Feeding, and Exclusive Breastfeeding

	OR	95% CI	p value
Maternal age	1.05	1.03–1.09	<0.001
Ethnicity
Latina	Reference
African American	1.04	0.69–1.58	0.22
White	1.21	0.67–2.17	0.52
Parity	0.80	0.69–0.92	0.002
BMI	1.01	0.99–1.04	0.125
Medicaid	1.39	0.88–2.19	0.152
EPDS	0.97	0.93–1.01	0.19
Cesarean section	0.74	0.52–1.05	0.10
Birth weight	1.00	0.99–1.00	0.20
Gestational age	0.96	0.84–1.105	0.61

BMI, body mass index; CI, confidence interval; EPDS, Edinburgh Postpartum Depression Scale; OR, odds ratio.

Discussion

This study describes the pathway of breastfeeding from maternal intention prior to delivery to actual breastfeeding at discharge at a hospital providing care to inner-city minority women. This is the first study to examine intention to breastfeed and outcome specifically at discharge. Studies have shown a woman's intention is the strongest predictor of initiation of breastfeeding.^5,10 Intention to breastfeed also was an important predictor in our study. Among those who intended to breastfeed, 75% were breastfeeding at discharge. Among those who intended to breastfeed exclusively, 40% were doing so at discharge. In addition, 11% of women who did not intend to breastfeed ended up breastfeeding, one-third of them exclusively.

The stages of change model has been used to examine the strength of maternal breastfeeding intentions.^11,12 Using that framework, it is possible that many women in our study were in the “contemplation” or “preparation” stage rather than in the “action” or “maintenance” stage. Successful breastfeeding has been described as a learned skill that requires both confidence and commitment.¹³ Without “confident commitment,” the decision to breastfeed may fall apart once challenged.¹³ A qualitative study found pregnant African American inner-city women well informed about the benefits of breastfeeding but concerned with life-style issues and pain.¹⁴ In a survey of a large multiethnic sample, comfort with formula feeding, comfort with breastfeeding, and breastfeeding self-efficacy all independently predicted intention.¹⁵ Comfort with formula feeding not only had the largest effect but also strongly predicted and substantially mediated ethnic disparity in breastfeeding intention.¹⁵

Our study found one in four women who expressed an intention to breastfeed was not breastfeeding at discharge. Moreover, most of those breastfeeding were supplementing with formula. The intention to breastfeed is modulated by early breastfeeding experiences and perceptions of infant preferences.^8,9,16,17

In the 2000–2001 Pregnancy Risk Assessment and Monitoring System survey of 10 states, 8% of women stopped breastfeeding within the first week because of breastfeeding difficulties and concerns with infant satisfaction.⁹ Among minority women in New Haven, CT, 27% stopped breastfeeding by 1 week because of lack of confidence in breastfeeding and the belief that the baby preferred formula.⁸ Low-income minority women in New York City felt breastfeeding was the best way to feed their infants, but their commitment turned to ambivalence in the face of their perceptions of the virtues of formula and the practical and sociocultural challenges of breastfeeding.¹⁸ Lastly, paternal involvement appears integral to the success of breastfeeding.¹⁹

In our study the independent predictors of breastfeeding at discharge were maternal older age and paradoxically, lower parity, but not ethnicity. In a 2000–2002 study of inner-city women in Philadelphia, Lee et al.²⁰ found a higher intention to breastfeed among African American compared with whites. However, there were no statistically significant differences in initiation of breastfeeding by ethnicity,²¹ suggesting the inner-city context blunted ethnic differences described at the national level. In national samples, younger and low-income women and primiparas were more likely to stop breastfeeding earlier.^6–9 In exploratory analysis, the association between intention and actual breastfeeding at discharge appeared to be stronger among multiparas compared with primiparas (Spearman correlation, 0.68 vs. 0.56, respectively). If so, this would suggest that prior experience strengthens the intention of whether to breastfeed or not.

The drop from breastfeeding intention to actual practice in the immediate postpartum period represents an important missed opportunity. High failure rates may be the result of hospital practices. In a 2006 national survey, mothers found the most helpful hospital practices with breastfeeding were help getting started, no supplementation, breastfeeding resources, and feeding on demand.⁶ A Cochrane review found certain institutional changes in maternity care practices effectively increased breastfeeding initiation and duration.²² Baby-Friendly hospitals in the United States have reported elevated rates of breastfeeding that persist regardless of demographic factors traditionally linked with low breastfeeding rates.²³

In our post hoc analysis (data not provided), breastfeeding in the delivery room and as the infants' first feeding in the nursery both were positively associated with breastfeeding at discharge. These findings emphasize the value of implementing change to support breastfeeding within the hospital setting.

The potential limitations of this study include a social desirability bias toward breastfeeding with regard to the woman's response to the nurse's question on infant feeding intention prior to delivery. We did not assess other possible confounders impacting the decision to breastfeed, such as social support, maternal education level, and marital status. The study was conducted at one urban community hospital serving an inner-city population, and findings may not be generalizable to other hospital populations. The strengths of the study include its large sample size, the combination of data on intention and actual behavior at discharge, and the relative homogeneity of the study group with respect to the sociodemographic profile.

Our study shows two major areas for action among our population of inner-city women: increase the proportion of women who show a positive intention to breastfeed and decrease the drop from intention to actual practice. This is particularly pertinent to exclusive breastfeeding, where intention and outcome are most divided. Women's positive intentions might be interpreted as a reflection of their openness to the idea to breastfeed—although they may not be ready to take action. In evidence-based reviews, strategies found to increase breastfeeding initiation compared with usual care included targeted prenatal peer support,²⁴ pre- and postnatal interventions, and layperson support in multicomponent interventions.²⁵

Additional research is needed to explore why prenatal breastfeeding intentions did not meet outcomes at hospital discharge. Qualitative studies are needed to customize breastfeeding promotion strategies and make them culturally and ethnically appropriate to inner-city women. Lastly, maternity practices to enhance women's level of comfort and confidence in breastfeeding during the vulnerable times before and after delivery are necessary.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Gartner

, Morton

, Lawrence

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

U.S. Department of Health and Human Services. Executive summary: The Surgeon General's call to action to support breastfeeding. Breastfeed Med, 2011; 6:3–5.

U.S. Department of Health and Human Services. Healthy People 2020. www.healthypeople.gov/2020/. 2012 February 24.

Jones

, Kogan

, Singh

et al. Factors associated with exclusive breastfeeding in the United States. Pediatrics, 2011; 128:1117–1125.

Donath

, Amir

. ALSPAC Study Team. Relationship between prenatal infant feeding intention and initiation and duration of breastfeeding: A cohort study. Acta Paediatr, 2003; 92:352–356.

Declercq

, Labbok

, Sakala

et al. Hospital practices and women's likelihood of fulfilling their intention to exclusively breastfeed. Am J Public Health, 2009; 99:929–935.

Dix

. Why women decide not to breastfeed. Birth, 1991; 18:222–225.

Ertem

, Votto

, Leventhal

. The timing and predictors of the early termination of breastfeeding. Pediatrics, 2001; 107:543–548.

Ahluwalia

, Morrow

, Hsia

. Why do women stop breastfeeding? Findings from the Pregnancy Risk Assessment and Monitoring System. Pediatrics, 2005; 116:1408–1412.

10.

Scott

, Landers

, Hughes

et al. Psychosocial factors associated with the abandonment of breastfeeding prior to hospital discharge. J Hum Lact, 2001; 17:24–30.

11.

Kloeblen

, Thompson

, Miner

. Predicting breast-feeding intention among low-income pregnant women: A comparison of two theoretical models. Health Educ Behav, 1999; 26:675–688.

12.

Humphreys

, Thompson

, Miner

. Assessment of breastfeeding intention using the Transtheoretical Model and the Theory of Reasoned Action. Health Educ Res Theory Pract, 1998; 13:331–341.

13.

Avery

, Magnus

. Expectant fathers' and mothers' perceptions of breastfeeding and formula feeding: A focus group study in three US cities. J Hum Lact, 2011; 27:147–154.

14.

Alexander

, Dowling

, Furman

. What do pregnant low-income women say about breastfeeding? Breastfeed Med, 2010; 5:17–23.

15.

Nommsen-Rivers

, Chantry

, Cohen

et al. Comfort with the idea of formula feeding helps explain ethnic disparity in breastfeeding intentions among expectant first-time mothers. Breastfeed Med, 2010; 5:25–33.

16.

DiGirolamo

, Thompson

, Martorell

et al. Intention or experience? Predictors of continued breastfeeding. Health Educ Behav, 2005; 32:208–226.

17.

Hurley

, Black

, Papas

et al. Variation in breastfeeding behaviours, perceptions, and experiences by race/ethnicity among a low-income statewide sample of Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) participants in the United States. Matern Child Nutr, 2008; 4:95–105.

18.

Kaufman

, Deenadayalan

, Karpati

. Breastfeeding ambivalence among low-income African American and Puerto Rican women in north and central Brooklyn. Matern Child Health J, 2010; 14:696–704.

19.

Alexander

, O'Riordan

, Furman

. Do breastfeeding intentions of pregnant inner-city teens and adult women differ? Breastfeed Med, 2010; 5:289–296.

20.

Lee

, Elo

, McCollum

et al. Racial/ethnic differences in breastfeeding initiation and duration among low-income, inner-city mothers. Soc Sci Q, 2009; 90:1251–1271.

21.

Lee

, Rubio

, Elo

et al. Factors associated with intention to breastfeed among low-income, inner-city pregnant women. Matern Child Health J, 2005; 9:253–261.

22.

Donnelly

, Snowden

, Renfrew

et al. Commercial hospital discharge packs for breastfeeding women. Cochrane Database Syst Rev, 2000; 2:CD002075.

23.

Merewood

, Mehta

, Chamberlain

et al. Breastfeeding rates in US Baby-Friendly hospitals: Results of a national survey. Pediatrics, 2005; 116:628–634.

24.

Ingram

, MacArthur

, Khan

et al. Effect of antenatal peer support on breastfeeding initiation: A systematic review. CMAJ, 2010; 182:1739–1746.

25.

Chung

, Raman

, Trikalinos

et al. Interventions in primary care to promote breastfeeding: An evidence review for the U.S. Preventive Services Task Force. Ann Intern Med, 2008; 149:565–582.