Att 3k_CDC MMWR 2014

Morbidity and Mortality Weekly Report

Racial Disparities in Access to Maternity Care Practices That Support
Breastfeeding — United States, 2011
Jennifer N. Lind, PharmD1,2, Cria G. Perrine, PhD2, Ruowei Li, MD2, Kelley S. Scanlon, PhD2, Laurence M. Grummer-Strawn, PhD2
(Author affiliations at end of text)

Despite the well documented health benefits of breastfeeding
(1), initiation of breastfeeding and breastfeeding duration rates
among black infants in the United States are approximately
16% lower than among whites (2). Although many factors
play a role in a woman’s ability to breastfeed, experiences during the childbirth hospitalization are critical for establishing
breastfeeding (3). To analyze whether the implementation
by maternity facilities of practices that support breastfeeding varied depending on the racial composition of the area
surrounding the facility, CDC linked data from its 2011
Maternity Practices in Infant Nutrition and Care (mPINC)
survey to U.S. Census data on the percentage of blacks living
within the zip code area of each facility. The results of that
analysis indicated that facilities in zip code areas where the
percentage of black residents was >12.2% (the national average during 2007–2011) were less likely than facilities in zip
code areas where the percentage was ≤12.2% to meet five of
10 mPINC indicators for recommended practices supportive
of breastfeeding and more likely to implement one practice;
differences for the other four practices were not statistically
significant. Comparing facilities in areas with >12.2% black
residents with facilities in areas with ≤12.2% black residents,
the largest differences were in the percentage of facilities that
implemented recommended practices related to early initiation
of breastfeeding (46.0% compared with 59.9%), limited use
of breastfeeding supplements (13.1% compared with 25.8%),
and rooming-in (27.7% compared with 39.4%). These findings suggest there are racial disparities in access to maternity
care practices known to support breastfeeding.
The mPINC survey is a biennial census of maternity facilities
(hospitals and free-standing birth centers) in the United States
and its territories (4). The survey is sent to the person at each
facility most knowledgeable about the facility’s maternity care
practices and policies. A total of 2,727 facilities participated
in the 2011 mPINC survey (response rate = 83%). These data
were analyzed for 10 mPINC indicators for recommended

maternity care practices* from the World Health Organization/
United Nations Children’s Fund’s Ten Steps to Successful
Breastfeeding (5). The Ten Steps are evidence-based practices
shown to increase breastfeeding exclusivity and duration, and
are the basis for the Baby-Friendly Hospital Initiative.†
To estimate the prevalence of facilities with recommended
maternity care practices by the percentage of black residents in
their area, zip code level data for the category “non-Hispanic black
or African American alone” were obtained for the period 2007–
2011 from the U.S. Census Bureau’s American Community
Survey (ACS). ACS is a continuous nationwide survey that collects detailed information on demographic, social, economic,
and housing characteristics; these data are only available by zip
code as 5-year estimates (6). ACS and mPINC data were linked
by zip codes; of the 2,727 facilities that participated in the 2011
mPINC survey, 84 (3%) facilities were missing zip code level
racial data in ACS, resulting in a final analytic sample of 2,643
facilities. Facilities were divided into two categories: 1) those in
zip code areas where the percentage of black residents was >12.2%
(the national average during 2007–2011) (6) and 2) those in zip
code areas where the percentage was ≤12.2%. The z-test was used
to compare data from the two categories and determine whether
differences in implementation of recommended maternity care
practices were statistically significant (p<0.05). No other racial
or ethnic groups were examined.
* The 10 mPINC indicators for recommended maternity care practices from the
Ten Steps were as follows: 1) Model breastfeeding policy: hospital has a written
breastfeeding policy that includes 10 model policy elements; 2) Staff competency
assessment: nurses/birth attendants are assessed for competency in basic
breastfeeding management and support at least once per year; 3) Prenatal
breastfeeding education: breastfeeding education is included as a routine element
of prenatal classes; 4) Early initiation of breastfeeding: ≥90% of healthy, fullterm, breastfed infants initiate breastfeeding within 1 hour of uncomplicated
vaginal birth; 5) Teach breastfeeding techniques: ≥90% of mothers who are
breastfeeding or intend to breastfeed are taught breastfeeding techniques (e.g.,
positioning and how to express milk); 6) Limited use of breastfeeding
supplements: <10% of healthy, full-term, breastfed infants are supplemented
with formula, glucose water, or water; 7) Rooming-in: ≥90% of healthy, fullterm infants, regardless of feeding method, remain with their mother for at
least 23 hours per day during the hospital stay; 8) Teach feeding cues: ≥90%
of mothers are taught to recognize and respond to infant feeding cues instead
of feeding on a set schedule; 9) Limited use of pacifiers: <10% of healthy, fullterm, breastfed infants are given pacifiers by maternity care staff members; and
10) Post-discharge support: hospital routinely provides three modes of postdischarge support to breastfeeding mothers (physical contact, active reaching
out, and referrals).
† Additional information available at http://www.babyfriendlyusa.org/about-us/
baby-friendly-hospital-initiative/the-ten-steps.

MMWR / August 22, 2014 / Vol. 63 / No. 33

725

Morbidity and Mortality Weekly Report

In 2011, three of the 10 mPINC indicators for recommended
practices were met by >75% of the 2,643 facilities surveyed.
The three were providing prenatal breastfeeding education
(92.7%), teaching breastfeeding techniques (90.7%), and
teaching mothers how to recognize and respond to infant
feeding cues (84.7%) (Table).
Facilities in zip code areas with >12.2% black residents
were significantly more likely to assess staff competency
than facilities in zip code areas with ≤12.2% black residents
(59.4% compared with 53.2%) (Table). However, facilities in
zip code areas with >12.2% black residents were significantly
less likely than facilities in zip code areas with ≤12.2% black
residents to meet five of the nine other mPINC indicators
for recommended practices: early initiation of breastfeeding
(46.0% compared with 59.9%), limited use of breastfeeding

supplements (13.1% compared with 25.8%), rooming-in
(27.7% compared with 39.4%), limited use of pacifiers,
(30.5% compared with 37.9%), and post-discharge support
(23.9% compared with 29.9%) (Table).
Discussion

In 2011, implementation of 10 recommended maternity care
practices supportive of breastfeeding among 2,643 maternity
facilities varied widely, ranging from 18.9% to 92.7%, and was
<50% for five practices. For half of the 10 practices, implementation was significantly lower among facilities in zip code areas
with a higher percentage of black residents. These findings are
important because research has shown that U.S. residents usually are admitted to hospitals within a relatively short distance
of where they live, although persons living in rural areas might

TABLE. Prevalence of facilities meeting indicators for recommended maternity care practices,* by racial composition† of the zip code areas
where the facilities were located — Maternity Practices in Infant Nutrition and Care Survey (mPINC), United States, 2011
Total
facilities
surveyed
(N = 2,643§)

≤12.2%
(n = 2,030§)

>12.2%
(n = 613§)

%

%

%

Model breastfeeding policy: hospital has a written breastfeeding
policy that includes 10 model policy elements.

18.9

18.5

20.3

-1.8

1.87

0.33

Staff competency assessment: nurses/birth attendants are assessed
for competency in basic breastfeeding management and support at
least once per year.

54.6

53.2

59.4

-6.2

2.28

0.01¶

Prenatal breastfeeding education: breastfeeding education is
included as a routine element of prenatal classes.

92.7

92.9

91.8

1.1

1.25

0.38

Early initiation of breastfeeding: ≥90% of healthy, full-term,
breastfed infants initiate breastfeeding within 1 hour of
uncomplicated vaginal birth.

56.7

59.9

46.0

13.9

2.31

<0.01¶

Teach breastfeeding techniques: ≥90% of mothers who are
breastfeeding or intend to breastfeed are taught breastfeeding
techniques (e.g., positioning and how to express milk).

90.7

91.2

89.2

2.0

1.41

0.16

Limited use of breastfeeding supplements: <10% of healthy,
full-term, breastfed infants are supplemented with formula,
glucose water, or water.

22.8

25.8

13.1

12.7

1.69

<0.01¶

Rooming-in: ≥90% of healthy, full-term infants, regardless of feeding
method, remain with their mother for at least 23 hours per day
during the hospital stay.

36.7

39.4

27.7

11.7

2.12

<0.01¶

Teach feeding cues: ≥90% of mothers are taught to recognize and
respond to infant feeding cues instead of feeding on a set schedule.

84.7

85.1

83.2

1.9

1.71

0.26

Limited use of pacifiers: <10% of healthy, full-term, breastfed infants
are given pacifiers by maternity care staff members.

36.2

37.9

30.5

7.4

2.16

<0.01¶

Post-discharge support: hospital routinely provides three modes of
post-discharge support to breastfeeding mothers (physical contact,
active reaching out, and referrals).

28.5

29.9

23.9

6.0

2.00

<0.01¶

mPINC indicators for recommended maternity care practices

Percentage of black residents in the facility zip code area
Percentage- Standard error
point
of the
difference
difference

p-value

* mPINC indicators for recommended maternity care practices are from Ten Steps to Successful Breastfeeding, available at http://www.babyfriendlyusa.org/about-us/
babyfriendly-hospital-initiative/the-ten-steps.
† Zip code areas in which the percentage of “non-Hispanic black or African American” residents was >12.2% (the national average during 2007–2011), compared with
≤12.2%, according to data from the U.S. Census Bureau’s American Community Survey.
§ Number of respondents varied slightly from the total for each of the prevalence estimates.
¶ Statistically significant percentage-point difference by z-test.

726

MMWR / August 22, 2014 / Vol. 63 / No. 33

Morbidity and Mortality Weekly Report

What is already known on this topic?
Breastfeeding has many health benefits for infants, yet there are
persistent gaps in breastfeeding rates between black and white
infants in the United States. Maternity care practices experienced during the hospital stay have a major impact on the
establishment of breastfeeding.
What is added by this report?
Facilities located in zip code areas with higher percentages of
blacks were less likely to meet five indicators for recommended
maternity care practices supportive of breastfeeding and more
likely to meet one indicator, than facilities in areas with a lower
percentage of blacks. The largest differences were for indicators
related to early initiation of breastfeeding, limited use of
breastfeeding supplements, and rooming-in.
What are the implications for public health practice?
Interventions are needed to ensure that all maternity care
facilities are implementing the recommended policies and
practices known to be important for the establishment of
breastfeeding. Facilities located in areas with higher percentages of blacks might need additional support.

travel farther than those in cities (7). Therefore, women living
in zip code areas with a higher percentage of blacks might have
less access to facilities implementing recommended maternity
care practices, which might contribute to lower breastfeeding
rates among blacks compared with other racial groups.
The reasons for the differences in maternity care practices by
racial composition of the areas are not clear. Further research
is needed on barriers to implementing recommended practices
in these areas, on whether poorer maternity care practices are
linked to lower breastfeeding rates in these areas, and on evaluating other factors that might be contributing to these disparities.
This is the first report based on national data showing that
practices at maternity facilities vary with the racial composition
of the zip code area in which the facility is located. However,
similar findings were observed in a previous study in North
Carolina that assessed whether there were differences in breastfeeding support services available through the Supplemental
Nutrition Program for Women, Infants, and Children (WIC)
program based on the county level racial/ethnic composition
of the WIC sites. It was found not only that breastfeeding
initiation by WIC site was negatively associated with the percentage of black clients, but also that WIC sites with higher
percentages of black clients were less likely to offer clinic-based
breastfeeding support services (8).
In a review of U.S.-based randomized trials evaluating
breastfeeding interventions targeting minorities, interventions
to change hospital or WIC policies, including enhanced practices and services, were among the public health approaches

found to successfully improve breastfeeding outcomes among
minority women (9). CDC currently is funding a project that
addresses the need for quality improvement in maternity care
practices. In June 2012, CDC awarded a 3-year cooperative
agreement to the National Initiative for Children’s Healthcare
Quality to assist 89 hospitals, mostly located in states that
have lower breastfeeding rates and that serve low-income and
minority women, with improving maternity care practices to
support breastfeeding and to move toward the Baby-Friendly
designation. Detailed descriptions of the cooperative agreement
program have been published (2,10).
The findings in this report are subject to at least four limitations. First, one mPINC indicator for each of the Ten Steps was
selected; these indicators are consistent with the Ten Steps,§
but might not encompass all aspects of each step. Second,
although the mPINC survey was sent to the person identified
as the most knowledgeable about the facility’s policies and
practices and facilities were encouraged to get input from key
staff members as needed, responses might not accurately reflect
actual practices. Third, the racial composition of the patients
served at each facility is not collected in the mPINC survey.
However, because most U.S. residents are admitted to hospitals
close to where they live and most hospital service areas have
only one local hospital, the data in this report for zip code
areas are likely reasonable estimates for the racial composition
of hospital patients, assuming overall hospital admission patterns (7) apply to births. Finally, only facilities with zip code
level race data were included in this analysis. Excluded facilities
might have had different percentages of blacks and maternity
care practices. However, only 3% of facilities were excluded,
which is not likely to have affected results.
The findings suggest that the implementation of maternity
care practices supportive of breastfeeding vary based on the
racial composition of the area, which means women living in
areas with higher percentages of blacks might have less access
to these services. Although the reasons for these disparities are
unclear, the results might provide some insight into why there
has been a persistent gap in breastfeeding initiation and duration rates between black and white infants in the United States.
All facilities, regardless of the racial/ethnic composition of the
populations they serve, can support the breastfeeding decisions
of their patients by implementing evidence-based policies and
practices shown to be critical for establishing breastfeeding, so
that more infants are able to reap the numerous health benefits
of breastfeeding.
§ Additional

information available at https://www.babyfriendlyusa.org/getstarted/the-guidelines-evaluation-criteria.

MMWR / August 22, 2014 / Vol. 63 / No. 33

727

Morbidity and Mortality Weekly Report

1Epidemic Intelligence Service, CDC; 2Division of Nutrition, Physical Activity,

and Obesity, National Center for Chronic Disease Prevention and Health
Promotion, CDC (Corresponding author: Jennifer N. Lind, [email protected],
770-498-4339)

References
1. Ip S, Chung M, Raman G, et al. Breastfeeding and maternal and infant
health outcomes in developed countries [Review]. Evid Rep Technol Assess
2007;153.
2. CDC. Progress in increasing breastfeeding and reducing racial/ethnic
differences—United States, 2000–2008 births. MMWR 2013;62:77–80.
3. DiGirolamo AM, Grummer-Strawn LM, Fein SB. Effect of maternity-care
practices on breastfeeding. Pediatrics 2008;122(Suppl 2):S43–9.
4. CDC. Maternity Practices in Infant Nutrition and Care (mPINC) survey.
Atlanta, GA: US Department of Health and Human Services, CDC;
2013. Available at http://www.cdc.gov/breastfeeding/data/mpinc.

728

MMWR / August 22, 2014 / Vol. 63 / No. 33

5. CDC. Vital signs: hospital practices to support breastfeeding—United
States, 2007 and 2009. MMWR 2011;60:1020–5.
6. US Census Bureau. American Community Survey. Washington, DC:
US Department of Commerce, US Census Bureau; 2008. Available at
http://www.census.gov/acs/www.
7. Wennberg JE, Cooper MM, eds; Dartmouth Atlas of Health Care
Working Group. The Dartmouth Atlas of Health Care in the United
States. Chicago, IL: American Hospital Publishing; 1998. Available at
http://www.dartmouthatlas.org/downloads/atlases/98atlas.pdf.
8. Evans K, Labbok M, Abrahams SW. WIC and breastfeeding support
services: does the mix of services offered vary with race and ethnicity?
Breastfeed Med 2011;6:401–6.
9. Chapman DJ, Perez-Escamilla R. Breastfeeding among minority women:
moving from risk factors to interventions. Adv Nutr 2012;3:95–104.
10. Grummer-Strawn LM, Shealy KR, Perrine CG, et al. Maternity care
practices that support breastfeeding: CDC efforts to encourage quality
improvement. J Womens Health (Larchmnt) 2013;22:107–12.