Social and Institutional Factors that Affect ...

Social and Institutional Factors that Affect Breastfeeding Duration Among WIC Participants in Los Angeles County, California Brent A. Langellier, M. Pia Chaparro & Shannon E. Whaley

Maternal and Child Health Journal ISSN 1092-7875 Volume 16 Number 9 Matern Child Health J (2012) 16:1887-1895 DOI 10.1007/s10995-011-0937-z

1 23

Your article is protected by copyright and all rights are held exclusively by Springer Science+Business Media, LLC. This e-offprint is for personal use only and shall not be selfarchived in electronic repositories. If you wish to self-archive your work, please use the accepted author’s version for posting to your own website or your institution’s repository. You may further deposit the accepted author’s version on a funder’s repository at a funder’s request, provided it is not made publicly available until 12 months after publication.

1 23

Author's personal copy Matern Child Health J (2012) 16:1887–1895 DOI 10.1007/s10995-011-0937-z

Social and Institutional Factors that Affect Breastfeeding Duration Among WIC Participants in Los Angeles County, California Brent A. Langellier • M. Pia Chaparro Shannon E. Whaley

•

Published online: 29 December 2011 Ó Springer Science+Business Media, LLC 2011

Abstract Hospital practices and early maternal return to work are associated with breastfeeding duration; however, research has not documented the long-term effects of many hospital policies or the effect of early return to work on breastfeeding outcomes of WIC participants. This study investigated the impact of in-hospital breastfeeding, receipt of a formula discharge pack, and maternal return to work on the long-term breastfeeding outcomes of 4,725 WIC participants in Los Angeles County, California. Multivariate logistic regression analyses were used to assess determinants of exclusive breastfeeding at 6 months and breastfeeding at 6, 12, and 24 months. In-hospital initiation of breastfeeding, exclusive breastfeeding in the hospital, receipt of a formula discharge pack, and maternal return to work before 3 months were all significantly associated with breastfeeding outcomes after controlling for known confounders. Mothers who exclusively breastfed in the hospital were eight times as likely as mothers who did not breastfeed in the hospital to reach the AAP recommendation of breastfeeding for 12 months or longer (P \ .01). Only 6.9% of the sample reported exclusively breastfeeding for 6 months or more, and just one-third reported any breastfeeding at 12 months. Nine in ten respondents received a formula discharge pack in the hospital. Mothers who received a discharge pack were half as likely to exclusively breastfeed at 6 months as those who did not B. A. Langellier (&) M. Pia Chaparro Department of Community Health Sciences, School of Public Health, University of California, Los Angeles, P.O. Box 951772, Los Angeles, CA, USA e-mail: [email protected] S. E. Whaley Department of Research and Evaluation, Public Health Foundation Enterprises WIC Program, Los Angeles, CA, USA

receive one (P \ .01). Medical providers should educate, encourage, and support WIC mothers to breastfeed in the hospital and refrain from giving formula discharge packs. Keywords Breastfeeding WIC Human milk Hospitals Maternity Infant formula Employment Abbreviations WIC Special Supplemental Nutrition Program for Women, Infants, and Children AAP American Academy of Pediatrics WHO World Health Organization Field Field Research Corporation

Introduction Breastfeeding has well-documented health benefits for mothers and children and has become a top public health priority [1, 2]. The American Academy of Pediatrics (AAP) and World Health Organization (WHO) recommend that infants be exclusively breastfed for the first 6 months of life and that continued breastfeeding with appropriate complementary foods continue for at least 1 year or more [3, 4]. Breastfeeding initiation rates in the United States have recently reached the Healthy People 2010 goal of 75%; however, exclusive breastfeeding and continuation rates remain low: just 13% of all US mothers exclusively breastfeed for 6 months and only 43 and 22% do any breastfeeding at 6 and 12 months, respectively [5, 6]. Breastfeeding rates are somewhat higher in California and Los Angeles County than nationwide: among California mothers, 87% initiate breastfeeding, 17% exclusively breastfeed for 6 months, and 54 and 31% do any

123

Author's personal copy 1888

breastfeeding at 6 and 12 months [5, 6]. Among mothers in Los Angeles County, 82% initiate breastfeeding, 16% exclusively breastfeed for 6 months, and 51 and 30% do any breastfeeding at 6 and 12 months [5, 6]. Breastfeeding rates are lower among participants in the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) than among the general population: just 9% of children in the WIC program are exclusively breastfed for 6 months, compared to 18% who are income ineligible [7]. Further, just 34% of WIC children are breastfed at 6 months and 18% are breastfed at 12 months, compared to 54 and 28% who are income ineligible [8]. Several institutional factors have been found to affect breastfeeding outcomes [9–21]. For example, breastfeeding duration is shorter among mothers who return to work within 3–6 months postpartum [9–13]. Hospital practices may also play a role in determining breastfeeding outcomes [14–20]. One hospital practice in particular that may affect breastfeeding outcomes is when hospitals give free ‘gift packs’ containing formula to mothers upon discharge. As discussed by Rosenberg and colleagues, it has been standard practice for decades for formula manufacturers to provide hospitals with free formula to use in the hospital and free formula gift packs to give mothers upon hospital discharge [21]. These gift packs are a low-cost and effective means by which formula companies can advertise their brand and encourage new mothers to try their formula. Further, as the gift packs are given out by hospital staff, some mothers may interpret this practice as being a medical endorsement of formula. In their own study among mothers in Oregon, Rosenberg and colleagues found that in-hospital receipt of a formula discharge pack was associated with significantly lower odds of exclusively breastfeeding at 10 weeks [21]. The objective of the current study is to investigate the impact that specific hospital practices and maternal return to work have on breastfeeding outcomes of WIC participants in Los Angeles County, California. Specifically, we investigate the impact of the following three factors on the breastfeeding outcomes of WIC participants: (1) in-hospital breastfeeding, (2) receipt of a formula discharge pack, and (3) maternal return to work within 3 and 6 months postpartum. Our hypothesis is that not breastfeeding in the hospital, receipt of a formula discharge pack, and early return to work will have a negative impact on the breastfeeding outcomes of WIC participants.

Materials and Methods Survey Design We use data from a 2008 survey of participants in the WIC program in Los Angeles County. This survey was the

123

Matern Child Health J (2012) 16:1887–1895

second in a continuous series of surveys that are conducted every 3 years in order to provide comprehensive data about the WIC population in Los Angeles County. The survey was based on the 2005 Los Angeles County Health Survey and adapted with extensive input from the California State WIC Division and WIC Local Agency staff [22]. Final survey development was aided by Field Research Corporation (Field), an independent public opinion research organization. Field staff provided considerable input and advice to facilitate administration of the survey questions by telephone and to ensure that the survey would average 20–25 min in length. The objectives of the survey were to assess key health indicators and health-related behaviors, as well as home and community indicators of support for families with young children. Surveys were written in English and translated into Spanish by a WIC translator who oversees translation of all WIC documents. During the translation, emphasis was placed on keeping language simple and consistent with the English version. The English and Spanish versions of the survey were each piloted with English- and Spanish-speaking WIC participants to ensure clarity and consistency. Interviews Interviews were conducted by Field staff. Prior to the start of full-scale data collection, both the English and Spanish questionnaires were programmed into Field’s computerassisted telephone interviewing system and several rounds of pre-testing of the survey instrument were performed. Field’s interviewers were trained in both general and specific interviewing techniques and sample record keeping, refusal conversion techniques, and confidentiality procedures. The performance of each member of the interviewing team was closely monitored and evaluated. In order to maintain WIC client confidentiality, auto-dialers were used such that interviewers only knew the first name of the person they were calling and the computer system suppressed the phone number from view. Sampling and Participants The survey was conducted with a random sample of WIC participants who received WIC services in Los Angeles County in January 2008. From the set of all 400,000 families receiving WIC services that month, 10,000 families were randomly identified as potential survey recipients. Prior to beginning the calling effort, WIC program staff mailed out a postcard to each randomly selected household alerting them that a Field interviewer would be calling on behalf of the WIC Program. The survey was completed with a total of 5005 WIC participants between the period of April 8 to July 22, 2008.

Author's personal copy Matern Child Health J (2012) 16:1887–1895

Only persons who could complete the survey in English or Spanish and reported that they or a child in the household was enrolled in the WIC program were eligible for the survey. If respondents reported that more than one child in the household was enrolled in the WIC program, data were collected regarding only the child with the most recent birthday. Up to eight attempts were made to reach and interview eligible respondents from each telephone listing dialed. The survey yielded an overall cooperation rate of 90.3%, indicating that when WIC participants were reached by phone, 9 out 10 completed the survey. The survey yielded a response rate of 59.6%. Only 4.9% of those who did not respond were refusals or partial interviews. Instead, many WIC participants were never able to be reached at home within 8 call attempts. Upon completion of surveys, WIC program staff mailed a $10 gift card to all households completing a survey. Approval from the Independent Review Consulting Institutional Review Board was obtained for all protocols prior to commencement of the study.

Breastfeeding Data and Analyses Breastfeeding data were collected from 4,725 respondents who reported they were the biological mother of a child enrolled in the WIC program (94.4% of the total sample). Four breastfeeding outcomes were used as dependent variables in the analyses: (1) any breastfeeding at 6 months; (2) any breastfeeding at 12 months; (3) any breastfeeding at 24 months; and (4) exclusive breastfeeding at 6 months, defined as an infant’s consumption of human milk with no supplementation of any type except for vitamins, minerals, and medications [23]. Breastfeeding durations were determined with a question asking, ‘‘How old was CHILD when you completely stopped breastfeeding him/her?’’ Exclusive breastfeeding was defined using the above question, a question asking about in-hospital breastfeeding, and a question asking how old the child was the first time he/she ate or drank anything besides breastmilk. Using the most conservative definition, we characterized children as exclusively breastfeeding at 6 months only if the mother reported all of the following: (1) no in-hospital supplementation; (2) breastfeeding of 6 months or more; and (3) no supplementation before 6 months. The focal independent variables in our analyses included: (1) non-exclusive and exclusive breastfeeding in the hospital, (2) receipt of a formula discharge pack, and (3) maternal return to work within 6 months. Breastfeeding in the hospital was determined by a question asking, ‘‘Did you breastfeed CHILD in the hospital?’’ Exclusive breastfeeding in the hospital was determined using a follow-up

1889

question that asked, ‘‘Was CHILD fed only breast milk at the hospital?’’ Receipt of a formula discharge pack at the hospital was determined with a question asking, ‘‘Did the hospital staff give you a gift pack with formula?’’ Maternal return to work within 6 months was determined with two question. The first question asked, ‘‘Since the birth of CHILD did you return to work or begin a new job?’’ Mothers who responded ‘yes’ were then asked a follow-up question, ‘‘How old was CHILD when you first returned to work or began work?’’ We used multivariate logistic regression analyses to assess the relationship between the breastfeeding outcomes and in-hospital breastfeeding, receipt of a formula discharge pack, and maternal return to work within 6 months. In the multivariate analyses we controlled for factors that may influence breastfeeding outcomes, including the child’s age and gender, maternal age, maternal race/ethnicity, maternal education, maternal foreign-born status, interview language, whether the other parent lives in the household, and the time period when the mother enrolled in the WIC program. We further controlled for the mother’s prenatal intention to breastfeed, which was assessed with a question asking, ‘‘While you were pregnant with CHILD, which of the following describes what you thought you would do with regard to breast-feeding CHILD?’’ Response options were that mothers knew they would breastfeed, thought they would breastfeed, knew they would not breastfeed, or didn’t know what to do about breastfeeding. We considered mothers to have a prenatal intention to breastfeed if they responded that they knew or thought they would breastfeed. We did not include income as a control variable in our multivariate models because a significant portion of our respondents were missing data on family income (22%) and the family income of all WIC participants is at or below 185% of the federal poverty level. In our multivariate models, we did not include in-hospital breastfeeding as a predictor of exclusive breastfeeding at 6 months due to colinearity, because all mothers who exclusively breastfed for 6 months also exclusively breastfed in the hospital. Each analysis was restricted to data for children as old or older than the duration in question. Therefore, children 5 months of age or younger were not included in any of the analyses, children 11 months and younger were not included in the analysis of ‘any’ breastfeeding at 12 months, and children 23 months or younger were not included in the analysis of any breastfeeding at 24 months. Respondents who answered ‘‘Don’t Know’’ or ‘‘Refused’’ for any of the variables in a model were not included in that analysis (total N shown on each table). We present descriptive statistics for all variables. All analyses were conducted using Stata version 10.0 [24].

123


Results Descriptive statistics are presented in Table 1. Among mothers in our sample, 90.9% were Hispanic, nearly 73.7% were foreign-born, and 70.1% of interviews were conducted in Spanish. Prenatal intention to breastfeed was high, with 85.9% of mothers reporting that they thought they would or might breastfeed. In-hospital breastfeeding was common among mothers in our sample; however, most in-hospital breastfeeding was non-exclusive: 75.9% of mothers breastfed in the hospital, but just 25.5% of all mothers were exclusively breastfeeding in the hospital. The majority of mothers, or 87.3%, reported receiving a formula discharge pack from the hospital. Only 6.9% of mothers reported exclusive breastfeeding at 6 months, the duration recommended by the AAP and WHO. While 53.3% of mothers reported any breastfeeding at 6 months, by 12 months only 34.5% were still breastfeeding and at 24 months only 11.0% continued breastfeeding. About 71% of mothers in the sample were not employed at the time of the survey, however nearly half of those who were employed (13.5% of all mothers) returned to their jobs within 3 months postpartum.

Matern Child Health J (2012) 16:1887–1895 Table 1 Descriptive statistics for a sample of WIC participants in Los Angeles County (Total n = 4,725) No.

Table 2 shows the results of four logistic regression models predicting any breastfeeding at 6, 12, and 24 months and exclusive breastfeeding at 6 months. Maternal race/ethnicity was not significantly associated with ‘any’ breastfeeding outcomes; however, non-Hispanic White mothers have 2.9 times the odds of Hispanic mothers to exclusively breastfeed at 6 months (P \ .01). Foreign-born mothers in our sample were more likely than U.S.-born mothers to breastfeed at both 6 (odds ratio: 2.0, P \ .01) and 12 months (OR: 1.9, P \ .01). Similarly, mothers interviewed in Spanish were more likely to breastfeed at 6 (OR: 1.8, P \ .01), 12 (OR: 1.5, P \ .01), and 24 months (OR: 1.9, P \ .05) than those interviewed in English. Mothers living in a household with the child’s other parent had 21% greater odds of breastfeeding at 6 months (P \ .05) and 31% greater odds of breastfeeding at 12 months (P \ .05) than mothers living in single-parent households. Breastfeeding Intention Breastfeeding intention emerged as a strong predictor of breastfeeding outcomes. After adjusting for demographic factors, in-hospital breastfeeding, receipt of a formula discharge pack, and maternal return to work, mothers reporting a prenatal intention to breastfeed had 3.6 times the odds of breastfeeding at 6 months (P \ .01) and 7.4

123

Mean (SD)

Child gender Female

2,347

49.7

Male

2,378

50.3

Child age (months)

4,725

28.2 (17.2)

Mother race/ethnicity Hispanic NH White NH Black Other/Multi Mother age (years)

4,285

90.9

101 214

2.1 4.5

115

2.4

4,722

29.9 (6.8)

Mother immigrant status Foreign-born

3,481

73.7

US-born

1,240

26.3

English

1,414

29.9

Spanish

3,311

70.1

\ High school

2,513

53.9

High school

1,167

25.0

986

21.1

1,208 3,516

25.6 74.4

Before birth

3,411

72.2

Birth-6 months

1,019

21.6

294

6.2

Interview language

Mother education

[ High school

Demographics

%

Other parent living in household Yes No Mother WIC enrollment

[6 Months postpartum Mother intended to breastfeed Yes

4,045

85.9

666

14.1

None

1,132

24.1

Non-exclusive

2,363

50.4

Exclusive

1,196

25.5

No Hospital breastfeeding

Received formula discharge pack at hospital Yes No

4,032

87.3

588

12.7

Maternal return to work (months) C7

421

8.9

4–6

295

6.4

0–3

636

13.5

3,372

71.4

Not employed Any breastfeeding at 6 months Yes

2,244

53.3

No

1,969

46.7


1891

Table 1 continued No.

%

Yes

1,257

34.5

No

2,392

65.6

291

11.0

2,363

89.0

Mean (SD)

Any breastfeeding at 12 months

Any breastfeeding at 24 months Yes No

discharge pack, mothers who returned to work within 3 months postpartum were 27% less likely to breastfeed at 6 months (P \ .05), 37% less likely to breastfeed at 12 months (P \ .01), and 51% less likely to breastfeed at 24 months (P \ .05) than mothers who returned to work after 7 months postpartum or later. Mothers who returned to work within 3 months postpartum were also 54% less likely to exclusively breastfeed at 6 months (P \ .01) than mothers who returned to work after 7 months postpartum.

Exclusive breastfeeding at 6 months Yes No

290

6.9

3,912

93.1

All categories do not add up to 4,725 due to missing data

times the odds of exclusively breastfeeding at 6 months (P \ .01) as mothers reporting no intention to breastfeed. In-Hospital Breastfeeding In-hospital breastfeeding had a significant impact on breastfeeding outcomes after adjusting for demographic factors, intention to breastfeed, receipt of a formula discharge pack, and maternal return to work. Mothers who breastfed non-exclusively in the hospital had 4.3 times the odds of mothers who did not breastfeed in the hospital to breastfeed at 6 months (P \ .01) and 3.5 times the odds of achieving the AAP recommendation of breastfeeding for 12 months or more (P \ .01). Compared to mothers who did not breastfeed in the hospital, those who exclusively breastfed in the hospital had 9.9 times the odds of breastfeeding at 6 months (P \ .01), 8.0 times the odds of breastfeeding at 12 months (P \ .01), and 5.7 times the odds of breastfeeding at 24 months (P \ .01). Receipt of Formula Discharge Pack After adjusting for demographic factors, intention to breastfeed, in-hospital breastfeeding, and maternal return to work, mothers who received a formula discharge pack in the hospital were 45% less likely than mothers who did not receive a formula discharge pack to reach the AAP recommendation of exclusive breastfeeding for 6 months (P \ .01). Receipt of a formula discharge pack in the hospital was not significantly associated with ‘any’ breastfeeding outcomes at 6, 12, or 24 months. Maternal Return to Work The timing of mothers’ return to work had a significant negative impact on all breastfeeding outcomes. After adjusting for demographic factors, intention to breastfeed, in-hospital breastfeeding, and receipt of a formula

Discussion Our study demonstrated generally high levels of nonexclusive breastfeeding among WIC mothers in Los Angeles County: about one in three mothers in our sample continued to breastfeed at 12 months, compared to 22.4% of all mothers nationally and 17.5% of WIC mothers nationally [8]. This is consistent with previous research, which has found higher rates of non-exclusive breastfeeding among all mothers in California and Los Angeles County than in the nation as a whole [5]. Our findings are also consistent with research suggesting that breastfeeding rates are higher among foreign-born and less-acculturated Latinas than among their US-born and highly-acculturated counterparts [25, 26]. On the other hand, just 7% of mothers in our sample exclusively breastfed for the full 6 months recommended by the AAP, compared to 13.3% of all mothers nationally and 9.2% of WIC mothers nationally [3, 7]. These findings are somewhat alarming when considering that 17.2% of all mothers in California and 16.2% of all mothers in Los Angeles County exclusively breastfeed for 6 months [6]. Despite the generally low levels of exclusive breastfeeding among mothers in our sample, nearly nine in ten intended to breastfeed prior to giving birth. Mothers with the prenatal intention to breastfeed were much more likely than other mothers to breastfeed either exclusively or nonexclusively for 6 months. The finding that breastfeeding intention predicts breastfeeding behavior is consistent with previous research and public health theory [10, 27–31]. Our study also demonstrated that, for those mothers who achieved it, exclusive breastfeeding in the hospital had a large and significant impact on breastfeeding outcomes at 6, 12, and 24 months. Mothers who exclusively breastfed in the hospital were about eight times more likely than mothers who did not breastfeed in the hospital to reach the AAP recommendation of continuing to breastfeed for at least 12 months. Previous studies have also shown that hospital practices and in-hospital initiation of exclusive breastfeeding can have a significant impact on breastfeeding continuation rates at between 1 and 20 weeks [14, 17, 18, 20]. We found that 25.5% of mothers in our sample

123



Table 2 Four multivariate logistic regression models predicting any breastfeeding at 6, 12, and 24 months and exclusive breastfeeding at 6 months among WIC children in Los Angeles County OR [95% CI] (1) Any 6 months

(2) Any 12 months

(3) Any 24 months

(4) Exclusive 6 months

0.901

0.920

0.816

0.951

[0.782,1.039]

[0.789,1.072]

[0.628,1.062]

[0.742,1.218]

1.005 [1.000,1.009]

1.003 [0.997,1.009]

1.010 [0.998,1.023]

1.007 [0.999,1.015]

Hispanic

Ref.

Ref.

Ref.

Ref.

NH White

0.861

1.627

1.932

2.859**

[0.501,1.479]

[0.864,3.064]

[0.528,7.065]

[1.292,6.329]

Male Child age (months) Mother race

NH Black Other/Multi Mother age (years) Mother foreign born Interview in Spanish

1.050

1.121

0.487

1.128

[0.703,1.569]

[0.671,1.873]

[0.109,2.172]

[0.506,2.518]

0.871

0.674

1.251

1.164

[0.533,1.424]

[0.360,1.261]

[0.399,3.919]

[0.465,2.918]

1.019**

1.010

1.021

0.986

[1.007,1.031]

[0.997,1.023]

[0.999,1.043]

[0.966,1.007]

1.970**

1.857**

1.427

1.353

[1.564,2.481]

[1.416,2.436]

[0.834,2.443]

[0.871,2.102]

1.753**

1.498**

1.889*

1.336

[1.381,2.225]

[1.141,1.967]

[1.098,3.250]

[0.848,2.105]

\ High school

Ref.

Ref.

Ref.

Ref.

High school

0.977

0.925

0.996

1.123

[0.818,1.166]

[0.765,1.120]

[0.722,1.374]

[0.829,1.520]

[ High school

1.114

1.048

0.861

1.159

Mother education

Other parent in household

[0.906,1.368]

[0.837,1.312]

[0.580,1.276]

[0.814,1.652]

1.209*

1.307*

1.254

1.162

[1.008,1.451]

[1.064,1.606]

[0.861,1.827]

[0.827,1.633]

Mother WIC Enrollment Before birth

Ref.

Ref.

Ref.

Ref.

Birth-6 months

1.041

0.953

1.011

1.127

[0.870,1.246]

[0.780,1.163]

[0.716,1.428]

[0.827,1.535]

[ 6 Months postpartum

1.211

1.138

0.757

1.350

[0.891,1.645]

[0.834,1.554]

[0.450,1.276]

[0.860,2.119]

3.559**

2.801**

2.074*

7.401**

[2.789,4.542]

[2.053,3.821]

[1.148,3.747]

[3.269,16.75]

None

Ref.

Ref.

Ref.

Non-exclusive

4.256**

3.523**

2.719**

Breastfeeding intention Hospital breastfeeding

[3.519,5.148]

[2.753,4.510]

[1.644,4.495]

9.937**

8.043**

5.718**

[7.927,12.46]

[6.169,10.49]

[3.435,9.519]

0.815

1.012

0.753

0.541**

[0.651,1.020]

[0.800,1.281]

[0.520,1.092]

[0.393,0.744]

C7

Ref.

Ref.

Ref.

Ref.

4–6

0.931

0.680

0.502

0.842

[0.652,1.330]

[0.458,1.009]

[0.246,1.022]

[0.451,1.575]

Exclusive Formula discharge pack at hospital Maternal return to work (months)

123


1893

Table 2 continued OR [95% CI]

0–3 Not employed Observations

(1) Any 6 months

(2) Any 12 months

(3) Any 24 months

(4) Exclusive 6 months

0.728*

0.627**

0.491*

0.456**

[0.538,0.986]

[0.448,0.877]

[0.276,0.875]

[0.251,0.828]

0.932

1.101

0.686

0.889

[0.726,1.197]

[0.854,1.420]

[0.470,1.003]

[0.590,1.339]

4,020

3,474

2,518

4,022

All odds ratios are adjusted for the effects of all other variables in the model * P \ .05; ** P \ .01

were able to exclusively breastfeed in the hospital, a figure that is actually slightly greater than the 24.4% of all mothers in Los Angeles County who were able to do so [32]. Despite this, it is clear that efforts need to be made to increase in-hospital exclusive breastfeeding rates among not only the WIC population, but also among all lowincome and minority mothers. In Los Angeles County, for example, 54% of White mothers exclusively breastfeed in the hospital, compared to just 17% of Latina mothers and 19% of Black mothers [32]. Given our finding that inhospital exclusivity is strongly associated with longer-term breastfeeding outcomes, increasing support for low-income and minority mothers to exclusively breastfeed in the hospital may represent a meaningful step towards erasing socioeconomic disparities in breastfeeding duration [33]. We found that hospital provision of formula discharge packs is pervasive among hospitals where WIC participants in Los Angeles County deliver. Further, receipt of a formula discharge pack had a large negative impact on exclusive breastfeeding rates at 6 months. The percentage of mothers in our sample who reported receiving a discharge pack was considerably higher than in other studies. Nine of every ten mothers in our sample received a discharge pack, compared to about two-thirds of mothers in both a national study and a study in Oregon [17, 21]. We were somewhat surprised that so many mothers in our sample reported receiving a formula discharge pack, given that in 2007 only 60% of hospitals in Los Angeles County claimed to give out discharge packs containing formula samples to breastfeeding patients [34]. While two other recent studies investigated the impact of receiving a formula discharge pack on breastfeeding outcomes at 1 and 10 weeks [17, 21], we believe our study is the first to show that the negative impact of formula discharge packs persists as long as 6 months. These results underline the importance of legislative and advocacy efforts that encourage hospitals to adopt breastfeeding-friendly policies [35, 36]. About 14% of all mothers in our sample returned to work within the first 3 months postpartum. We found that

these mothers were less likely than mothers who returned to work after 7 months or longer to breastfeed at 6, 12, and 24 months and to exclusively breastfeed at 6 months. This finding is consistent with previous research [9–13] and underscores the importance of ensuring that women in lowincome jobs have maternity leaves of 3 months or longer. It is further important that workplaces and child care centers provide adequate resources and support so that mothers can continue breastfeeding once they do return to work. This study has inherent limitations that must be considered when interpreting the findings. Much of our data, including data on breastfeeding outcomes, was collected from mothers up to 5 years postpartum so there is a high potential for recall bias. We attempted to address this concern by treating breastfeeding duration as a dichotomous rather than a continuous variable. Although we could have dropped data on children older than 1 or 2 years from our analyses, this would have restricted our ability to assess determinants of breastfeeding at 24 months. Given that the WHO recommends continued breastfeeding for 24 months or longer, we believe that this is an important outcome that merits inclusion in our study. The breastfeeding intention variable may be particularly subject to bias, not only because the data was collected retrospectively but also because mothers’ recall of breastfeeding intention may be influenced by their actual breastfeeding behavior. Given this potential for bias, findings regarding breastfeeding intention should be taken with caution but the consistency of these findings with previous research lessens this concern. Another limitation is that our study population consists exclusively of WIC participants in Los Angeles County and respondents were predominately Latina and foreignborn. Further, the fact that about 70% of mothers in our sample preferred to conduct interviews in Spanish implies a generally low level of acculturation. Several studies have shown that breastfeeding rates vary by race/ethnicity [33, 37] and, among Latinas, by acculturation status [25, 26]. Our findings may therefore be more relevant to the WIC population in California, which is 78% Latino, than to the entire WIC population in the US [38, 39].

123


A further limitation of our study is that we may have omitted relevant variables. For instance, we lack specific knowledge about why certain mothers may not have initiated breastfeeding in the hospital. Some mothers may never have intended to breastfeed in the hospital, while others may have been unable to initiate breastfeeding due to specific birthing practices, lack of lactation support, or complications at birth. Similarly, we lack information on the prenatal breastfeeding education that mothers received prior to giving birth and how this may have affected their breastfeeding behavior both in the hospital and longer term. We also lack information regarding whether or not mothers were employed prior to giving birth and thus are unable to determine how this may have impacted the timing of returns to work or breastfeeding outcomes. As with any cross-sectional survey, it is possible that the omission of key variables such as these may have confounded our results. We attempted to limit the potential for confounding by including a wide a range of covariates. Despite these limitations, our study has several strengths. One is the large sample size, which is considerably larger than other studies investigating the effects of hospital practices on breastfeeding outcomes [17, 18, 20, 40]. Our study used data from a sample of WIC mothers, a population at particular risk of early discontinuation of breastfeeding [41]. Our findings, therefore, give a different perspective of determinants of breastfeeding than many other studies, which have commonly used state- or nationally-representative data or samples of mothers with a prenatal intention to breastfeed [16–21]. By concentrating on a low-income, largely Latina, and WIC-enrolled population, our study shows that hospital practices and maternal return to work impact an already-vulnerable population. Our study therefore highlights a tremendous opportunity for hospital practitioners and public health professionals to address one of the nation’s foremost public health concerns.

Conclusions In conclusion, our findings indicate that early maternal return to work, in-hospital breastfeeding, and receipt of a formula discharge pack each play a critical role in determining WIC participants’ long-term breastfeeding outcomes. It is very important that WIC participants receive adequate maternity leave after giving birth and that, once they do return to the workplace, they receive the resources and support necessary to continue breastfeeding. About nine in ten mothers in our sample reported they had a prenatal intention to breastfeed their children, and over three-quarters of mothers did at least some breastfeeding in the hospital. However, only one-quarter of mothers

123


exclusively breastfed in-hospital and nearly nine in ten left the hospital with a formula gift pack. Recent changes in the WIC program that support breastfeeding, which include an enhanced food package and expanded peer counseling support, could be undermined by the extremely high prevalence of supplementation and discharge packs given out at hospitals. The findings of our study suggest that hospital policies and practices need to reflect the fact that in-hospital breastfeeding affects long-term breastfeeding outcomes. Hospital providers should provide all mothers with the education, encouragement, and support necessary to achieve successful breastfeeding outcomes. Further, hospitals should stop giving out free formula discharge packs, which appear to have significant damaging effects on exclusive breastfeeding outcomes. Acknowledgments We gratefully acknowledge Judy Gomez, Lu Jiang, Nelly Mallo, Denise Gee, Eloise Jenks, Mike Whaley, Eric Fried and Armando Jimenez for their unique and significant contributions to this work and the WIC participants for their willingness to complete the survey. Funding for this study was provided by First 5 LA. Conflict of interest interest to report.

None of the authors have any conflicts of

References 1. Satcher, D. S. (2001). DHHS blueprint for action on breastfeeding. Public Health Reports, 116(1), 72–73. 2. Ip, S., Chung, M., Raman, G., et al. (2007). Breastfeeding and maternal and infant health outcomes in developed countries. Evidence Report/Technology Assessment (Full Rep), 153, 1–186. 3. Gartner, L. M., Morton, J., Lawrence, R. A., et al. (2005). Breastfeeding and the use of human milk. Pediatrics, 115(2), 496–506. 4. World Health Organization. (2001). The optimal duration of exclusive breastfeeding: Report of an expert consultation. Switzerland: Geneva. 5. National Immunization Survey, Centers for Disease Control and Prevention, Department of Health and Human Services. (2007). Provisional Geographic-specific Breastfeeding Rates among Children born in 2007. Available from: http://www.cdc.gov/breast feeding/data/NIS_data/2007/state_any.htm. Accessed: November 20, 2011. 6. National Immunization Survey, Centers for Disease Control and Prevention, Department of Health and Human Services. (2007). Provisional Geographic-specific Exclusive Breastfeeding Rates among Children born in 2007. Available from: http://www. cdc.gov/breastfeeding/data/NIS_data/2007/state.htm. Accessed: November 20, 2011. 7. National Immunization Survey, Centers for Disease Control and Prevention, Department of Health and Human Services. (2007). Provisional Exclusive Breastfeeding Rates by Socio-demographic Factors Among Children Born in 2007. Available from: http:// www.cdc.gov/breastfeeding/data/NIS_data/2007/socio-demographic. htm. Accessed: November 20, 2011. 8. National Immunization Survey, Centers for Disease Control and Prevention, Department of Health and Human Services. (2007).


9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24. 25.

26.

Provisional Breastfeeding Rates by Socio-demographic Factors Among Children Born in 2007. Available from: http://www.cdc. gov/breastfeeding/data/NIS_data/2007/socio-demographic_any. htm. Accessed: November 20, 2011. Roe, B., Whittington, L. A., Fein, S. B., et al. (1999). Is there competition between breast-feeding and maternal employment? Demography, 36(2), 157–171. Thulier, D., & Mercer, J. (2009). Variables associated with breastfeeding duration. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 38(3), 259–268. Taveras, E. M., Capra, A. M., Braveman, P. A., et al. (2003). Clinician support and psychosocial risk factors associated with breastfeeding discontinuation. Pediatrics, 112(1), 108–115. Berger, L. M., Hill, J., & Waldfogel, J. (2005). Maternity leave, early maternal employment and child health and development in the US. Economical Journal, 115(501), F29–F47. Kimbro, R. T. (2006). On-the-job moms: Work and breastfeeding initiation and duration for a sample of low-income women. Maternal and Child Health Journal, 10(1), 19–26. Merewood, A., Mehta, S. D., Chamberlain, L. B., et al. (2005). Breastfeeding rates in US Baby-Friendly hospitals: Results of a national survey. Pediatrics, 116(3), 628–634. Bartington, S., Griffiths, L. J., Tate, A. R., et al. (2006). Are breastfeeding rates higher among mothers delivering in Baby Friendly accredited maternity units in the UK? International Journal of Epidemiology, 35(5), 1178–1186. Murray, E. K., Ricketts, S., & Dellaport, J. (2007). Hospital practices that increase breastfeeding duration: Results from a population-based study. Birth, 34(3), 202–211. Declercq, E., Labbok, M. H., Sakala, C., et al. (2009). Hospital practices and women’s likelihood of fulfilling their intention to exclusively breastfeed. American Journal of Public Health, 99(5), 929–935. DiGirolamo, A. M., Grummer-Strawn, L. M., & Fein, S. (2001). Maternity care practices: Implications for breastfeeding. Birth, 28(2), 94–100. Rosenberg, K. D., Stull, J. D., Adler, M. R., et al. (2008). Impact of hospital policies on breastfeeding outcomes. Breastfeeding Medicine, 3(2), 110–121. DiGirolamo, A. M., Grummer-Strawn, L. M., & Fein, S. B. (2008). Effect of maternity-care practices on breastfeeding. Pediatrics, 122(Suppl 2), S43–S49. Rosenberg, K. D., Eastham, C. A., Kasehagen, L. J., et al. (2008). Marketing infant formula through hospitals: The impact of commercial hospital discharge packs on breastfeeding. American Journal of Public Health, 98(2), pp 290–295. Los Angeles County Department of Public Health. (2005). The Los Angeles County Health Survey. Available at: http://www. publichealth.lacounty.gov/ha/hasurveyintro.htm Accessed July 30, 2010. Institute of Medicine. (1991). Subcommittee on Nutrition During Lactation, United States Health Resources and Services Administration. Nutrition During Lactation. Washington, DC: National Academy Press. StataCorp. Stata/MP 10.0. College Station, TX. 2007. Guendelman, S., & Siega-Riz, A. M. (2002). Infant feeding practices and maternal dietary intake among latino immigrants in California. Journal of Immigrant Health, 4(3), 137–146. Harley, K., Stamm, N., & Eskenazi, B. (2007). The effect of time in the U.S. on the duration of breastfeeding in women of Mexican Descent. Maternal and Child Health Journal, 11(2), 119–125.

1895 27. Donath, S., Amir, L., & The, A. S. T. (2003). Relationship between prenatal infant feeding intention and initiation and duration of breastfeeding: A cohort study. Acta Paediatrica, 92(3), 352–356. 28. Avery, M., Duckett, L., Dodgson, J., et al. (1998). Factors associated with very early weaning among primiparas intending to breastfeed. Maternal and Child Health Journal, 2(3), 167–179. 29. Chezem, J., Friesen, C., & Boettcher, J. (2003). Breastfeeding knowledge, breastfeeding confidence, and infant feeding plans: Effects on actual feeding practices. Journal of Obstetric, Gynecologic, and Neonatal Nursing, 32(1), 40–47. 30. DiGirolamo, A., Thompson, N., Martorell, R., et al. (2005). Intention or experience? Predictors of continued breastfeeding. Health Education & Behaviour, 32(2), 208–226. 31. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179–211. 32. California Department of Public Health. (2007). California inhospital breastfeeding as indicated on the newborn screening test form statewide and maternal county of residence by race/ethnicity: 2007. Available from: http://www.cdph.ca.gov/data/statistics/ Documents/MO-CountyOfResidence-xRaceEthnicityReport2007. pdf. Accessed: November 20, 2011. 33. Singh, G. K., Kogan, M. D., & Dee, D. L. (2007). Nativity/ immigrant status, race/ethnicity, and socioeconomic determinants of breastfeeding initiation and duration in the United States, 2003. Pediatrics, 119, S38–S46. 34. The Centers for Disease Control and Prevention. (2008). Maternity Practices in Infant Nutrition and Care (mPINC) Survey, 2007: Los Angeles County Benchmark Report. Atlanta, GA. Available from: http://www.cdph.ca.gov/data/statistics/Documents/MO-BFPmPINC-LA-LAC.pdf. Accessed: November 20, 2011. 35. The State of California. (2011). California Senate Bill 502–Hospital Infant Feeding Act. Sacramento, CA. Available from: http:// calwic.org/storage/sb_502_bill_20111006_chaptered.pdf. Accessed: November 20, 2011. 36. California Department of Public Health. (2010). Introduction to the On-Line Toolkit for Providing Breastfeeding Support: Model Hospital Policy Recommendations. Sacramento, CA. Available from: http://www.cdph.ca.gov/HealthInfo/healthyliving/childfamily/ Pages/IntroductiontotheModelHospitalPolicyRecommendations Toolkit.aspx. Accessed: March 20, 2009. 37. Li, R. W., & Grummer-Strawn, L. (2002). Racial and ethnic disparities in breastfeeding among United States infants: Third National Health and Nutrition Examination Survey, 1988–1994. Birth-Issue Perinat Care, 29(4), 251–257. 38. California WIC Association. (2010). WIC Facts & Figures. Sacramento. http://calwic.org/about-us/wic-facts-and-figures. Accessed December 12 2010. 39. Connor, P., Bartlett, S., Mendelson, M., et al. (2008). WIC participant and program characteristics, 2008. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service, Office of Research and Analysis. 40. Dabritz, H. A., Hinton, B. G., & Babb, J. (2009). Evaluation of lactation support in the workplace or school environment on 6-month breastfeeding outcomes in Yolo County, California. Journal of Human Lactation, 25(2), 182–193. 41. US Centers for Disease Control and Prevention. (2007). Division of Nutrition Physical Activity and Obesity, National Center for Chronic Disease Prevention and Health Promotion. Breastfeeding Among U.S. Children Born 1999—2007, CDC National Immunization Survey. Atlanta, GA.

123