ArticlePDF Available

Evaluating Birth Outcomes From a Community-Based Pregnancy Support Program for Refugee Women in Georgia

  • Emory Decatur Hospital

Abstract and Figures

Refugee women face numerous and unique barriers to sexual and reproductive healthcare and can experience worse pregnancy-related outcomes compared with U.S.-born and other immigrant women. Community-based, culturally tailored programs like Embrace Refugee Birth Support may improve refugee access to healthcare and health outcomes, but empirical study is needed to evaluate programmatic benefits. This community-engaged research study is led by the Georgia Doula Access Working Group, including a partnership between academic researchers, Emory Decatur Hospital nurses, and Embrace. We analyzed hospital clinical records ( N = 9,136) from 2016 to 2018 to assess pregnancy-related outcomes of Embrace participants ( n = 113) and a comparison group of women from the same community and racial/ethnic backgrounds ( n = 9,023). We controlled for race, language, maternal age, parity, insurance status, preeclampsia, and diabetes. Embrace participation was significantly associated with 48% lower odds of labor induction (OR = 0.52, p = 0.025) and 65% higher odds of exclusive breastfeeding intentions (OR = 1.65, p = 0.028). Embrace showed positive but non-significant trends for reduced cesarean delivery (OR = 0.83, p = 0.411), higher full-term gestational age (OR = 1.49, p = 0.329), and reduced low birthweight (OR = 0.77, p = 0.55). We conclude that community-based, culturally tailored pregnancy support programs like Embrace can meet the complex needs of refugee women. Additionally, community-engaged, cross-sector research approaches could ensure the inclusion of both community and clinical perspectives in research design, implementation, and dissemination.
Content may be subject to copyright.
published: 17 June 2021
doi: 10.3389/fgwh.2021.655409
Frontiers in Global Women’s Health | 1June 2021 | Volume 2 | Article 655409
Edited by:
Tabassum Firoz,
Yale New Haven Health System,
United States
Reviewed by:
Adanna Chukwuma,
World Bank Group, United States
Bilal Ahmed Usmani,
Aga Khan University, Pakistan
Elizabeth A. Mosley
Specialty section:
This article was submitted to
Maternal Health,
a section of the journal
Frontiers in Global Women’s Health
Received: 18 January 2021
Accepted: 18 May 2021
Published: 17 June 2021
Mosley EA, Pratt M, Besera G,
Clarke LS, Miller H, Noland T,
Whaley B, Cochran J, Mack A and
Higgins M (2021) Evaluating Birth
Outcomes From a Community-Based
Pregnancy Support Program for
Refugee Women in Georgia.
Front. Glob. Womens Health
doi: 10.3389/fgwh.2021.655409
Evaluating Birth Outcomes From
a Community-Based Pregnancy
Support Program for Refugee
Women in Georgia
Elizabeth A. Mosley 1,2
*, Michelle Pratt 3, Ghenet Besera 2, Lasha S. Clarke 2, Heidi Miller 4,
Tracy Noland 4, Bridget Whaley 2, Jennifer Cochran 4, Amber Mack 5and Melinda Higgins 6
1Georgia State University School of Public Health, Atlanta, GA, United States, 2Emory University Rollins School of Public
Health, Atlanta, GA, United States, 3Emory Decatur Hospital, Decatur, GA, United States, 4Embrace Refugee Birth Support,
Clarkston, GA, United States, 5Healthy Mothers, Healthy Babies Coalition of Georgia, Atlanta, GA, United States, 6Emory
University Nell Hodgson Woodruff School of Nursing, Atlanta, GA, United States
Refugee women face numerous and unique barriers to sexual and reproductive
healthcare and can experience worse pregnancy-related outcomes compared with
U.S.-born and other immigrant women. Community-based, culturally tailored programs
like Embrace Refugee Birth Support may improve refugee access to healthcare and
health outcomes, but empirical study is needed to evaluate programmatic benefits.
This community-engaged research study is led by the Georgia Doula Access Working
Group, including a partnership between academic researchers, Emory Decatur Hospital
nurses, and Embrace. We analyzed hospital clinical records (N=9,136) from 2016
to 2018 to assess pregnancy-related outcomes of Embrace participants (n=113)
and a comparison group of women from the same community and racial/ethnic
backgrounds (n=9,023). We controlled for race, language, maternal age, parity,
insurance status, preeclampsia, and diabetes. Embrace participation was significantly
associated with 48% lower odds of labor induction (OR =0.52, p=0.025) and 65%
higher odds of exclusive breastfeeding intentions (OR =1.65, p=0.028). Embrace
showed positive but non-significant trends for reduced cesarean delivery (OR =0.83,
p=0.411), higher full-term gestational age (OR =1.49, p=0.329), and reduced
low birthweight (OR =0.77, p=0.55). We conclude that community-based, culturally
tailored pregnancy support programs like Embrace can meet the complex needs of
refugee women. Additionally, community-engaged, cross-sector research approaches
could ensure the inclusion of both community and clinical perspectives in research
design, implementation, and dissemination.
Keywords: refugees, perinatal health, pregnancy support, birth outcomes, community-based research, doulas
Refugee women resettled in the United States (U.S.) face numerous and unique barriers to sexual
and reproductive healthcare, and can experience worse pregnancy-related outcomes than U.S.-born
and other immigrant women (18). Historically, the U.S. had the largest refugee resettlement
program in the world, with over 3 million refugees resettled there since 1975—half of whom were
Mosley et al. Evaluating Refugee Pregnancy Support Program
women (9,10). After the 2016 presidential election, U.S.
resettlement numbers dropped to a record low of only 22,000
refugees in 2018 (down from 85,000 in 2016) (11). In turn,
federal funding for refugee resettlement has also been reduced,
endangering refugee-centered programming efforts (12,13).
Approximately one-third of the U.S. refugee resettlement
agencies have been forced to close due to budget shortfalls (13).
The antecedent life experiences of refugee women, combined
with the challenges of a hostile sociopolitical environment in
the U.S., are associated with poorer pregnancy-related outcomes.
Compared with U.S.-born and non-refugee immigrant women,
refugee women begin prenatal care later and have fewer
prenatal visits during pregnancy due to socioeconomic and
language barriers, stigma, culturally insensitive health services,
and other challenges (1,46,8). Additionally, a number of
studies have shown that refugee women have higher rates of
labor induction and cesarean delivery (2,14), higher risk of
preterm birth (1,2), lower birthweight babies (1,3), and lower
rates of exclusive breastfeeding (15,16). Although, notably,
some studies have shown inconsistent results—for example,
well-documented advantages in birth outcomes for foreign-
born Latina women relative to U.S.-born women (17) and
other studies that found African refugee women have better
pregnancy outcomes than U.S-born women. Further, previous
studies have found associations between characteristics of refugee
and migrant populations, including race/ethnicity or country of
origin (14,1820), length of time in the host country (19), parity
(3,20), and maternal age (20) and their maternal and child-health
outcomes. For example, in a systematic review and meta-analysis,
researchers found cesarean birth rates were higher among Sub-
Saharan African and South Asian migrants compared with
non-migrant women, but Eastern European and Vietnamese
migrants had lower rates compared with non-migrant women
(20). In addition to considering factors established as having
an association with maternal and child-health outcomes, the
diverse backgrounds (e.g., country of origin and length of time
in the host country) of refugees can influence maternal and
child-health outcomes.
Considerable literature describes the independent and joint
benefits for healthy mothers and infants of spontaneous (vs.
induced), vaginal (vs. cesarean) delivery at 37 weeks gestation or
more (i.e., full term). For instance, induction of labor is associated
with greater risk of maternal post-partum hemorrhage relative
to spontaneous labor, and with greater risk of fetal stress and
respiratory illness (21). Labor induction is also associated with
a significantly higher risk of cesarean delivery (22), which itself is
linked to acute and chronic complications, including postpartum
cardiac arrest (23). Additionally, the increased risk of morbidity
and mortality among infants born preterm or early term is
also well-known (2426). Low birthweight (<2,500 g), which
is often a consequence of preterm birth, confers a higher risk
of infant mortality and morbidity (including cognitive deficits
and motor delays) that can extend throughout the life course
(27). Particularly for low birthweight infants, though the benefit
may extend to infants of normal birthweight, evidence suggests
that initiation of breastfeeding within the first day of life is
associated with a significant reduction in the risk of neonatal
mortality as compared with breastfeeding delayed for >24 h after
birth (28).
Community- and evidence-based pregnancy support
programs—ones that provide support for pregnant women
throughout the duration of their pregnancy and the postpartum
period through strategies such as doula support and group
education—have the potential to provide support and improve
connection to culturally appropriate healthcare, but there is
a dearth of evidence on how these programs impact refugee
birth outcomes (29). In a mixed-methods study of the Refugee
Women’s Health Clinic in Arizona, researchers documented
widespread barriers to prenatal care and high approval of the
specialized clinic among refugee women from diverse ethnic
backgrounds but did not assess birth outcomes (30). Another
study of community-based doulas in New York demonstrated
that having a doula of the same ethnic background improved
Burmese refugee women’s self-advocacy during labor and
delivery (31). However, this qualitative study did not measure
associations with maternal and child health outcomes. Similarly,
the participants in a qualitative study of community-based
prenatal services for refugee women in Perth, Australia,
reported improved social support, greater continuity of care,
increased knowledge about pregnancy, greater confidence to ask
questions, and more assistance with other life challenges such
as transportation and language services (32). Finally, a study
with Burmese refugees in Melbourne, Australia, showed that
group-based prenatal care can help women feel more informed,
prepared, and confident; improve social and emotional support;
and build trusting relationships with healthcare providers
(33). But it is still unclear how group-based prenatal care and
childbirth education translate to improved birth outcomes.
To date, evidence on reproductive health disparities and
interventions for refugee women is sparse and incomplete.
As described above, most studies use qualitative or mixed
methods to evaluate the process and experience of refugee
pregnancy support programs. Much less is known about how
these programs quantitatively impact birth outcomes on a large
scale. Additional research is needed to assess programmatic
effects on reproductive health outcomes for refugees in the U.S.
This requires both adequate sample sizes and statistical methods
that can account for the vast diversity of refugee groups.
The current study is a community-engaged, quantitative
evaluation of maternal health outcomes, child health outcomes,
and breastfeeding intentions among the participants of the
Embrace Refugee Birth Support program in Clarkston, Georgia.
Founded in 2010, Embrace Refugee Birth Support (34) is a
comprehensive, culturally tailored pregnancy support program
offered by the non-profit, refugee support organization Friends
of Refugees. The Embrace participants receive 8 weeks of no-
cost, evidence-based childbirth education classes taught in their
language by community liaisons. They are also matched to an
Embrace volunteer, who provides transportation to prenatal and
postnatal visits, continuous support during labor and childbirth,
and social connection. Embrace uses an evidence-based “Healthy
Moms” curriculum (35) for pregnant refugee women that covers
topics including prenatal health, newborn care, and how to
communicate with providers. In our study, we analyze hospital
Frontiers in Global Women’s Health | 2June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
clinical records to compare birth outcomes for the Embrace
participants and women who did not participate in the program,
controlling for relevant covariates. We ask the following:
Do the Embrace participants have improved maternal health
outcomes, including reduced labor induction, vaginal delivery
compared with similar women who did not participate
in Embrace?
Do the Embrace participants have improved child health
outcomes, including higher birthweight and gestational age at
birth compared with similar women who did not participate
in Embrace?
Do the Embrace participants have greater likelihood of
breastfeeding intentions compared with similar women who
did not participate in Embrace?
Research Approach and Setting
In 2019, the Healthy Mothers, Healthy Babies Coalition of
Georgia (HMHBGA)—a non-profit organization dedicated to
local maternal and child health—convened the Georgia Doula
Access Working Group (GADAWG) (36). The GADAWG
mission is to improve access to full-spectrum doula services in
Georgia, especially for marginalized groups facing the greatest
barriers to high-quality maternal and child healthcare, including
refugees, people of color, and low-income families. This study
is a community-engaged, cross-sector collaboration between
Embrace Refugee Birth Support, Georgia State University School
of Public Health, Emory School of Public Health, Emory School
of Nursing, and Emory Decatur Hospital overseen and supported
by the GADAWG. The members of our research team represent
each of the organizations involved as well as the refugee
communities we serve.
It is important to note that Clarkston, Georgia is a uniquely
diverse community near Atlanta. The town has resettled over
37,000 refugees in the past 25 years, and the current population of
nearly 13,000 is over 31% foreign-born with representation from
150 countries, 760 ethnic groups, and 140 languages all in one
square mile. Emory Decatur Hospital—located only 3 miles from
Clarkston—is where the majority of the Embrace participants
give birth. Inclusive of Clarkston, metro-Atlanta is the ninth
largest metro area in the country at over 6 million residents and
the fourth fastest growing (37). Its population is also diverse with
46% White, 34% Black, 11% Hispanic, and 6% Asian residents.
Yet racial/ethnic and economic inequality persist: Atlanta has the
highest income inequality in the country, 76% of Black children
live in high-poverty neighborhoods compared to 6% of White
children, and Black women are 3 times as likely as White women
to die from pregnancy-related causes (38,39).
All research activities were approved by the Emory University
Institutional Review Board (IRB00109995). The researchers were
trained in research ethics through the Collaborative Institutional
Training Initiative, including human subjects protection.
Data and Measures
To evaluate the Embrace program, we abstracted maternal
clinical records from March 2016 to December 2018 at Emory
Decatur Hospital. There were a total of 9,136 unique clinical
records during this period. We sought to include all covariates
related to refugee maternal and child health outcomes, but we
were limited by the hospital medical records. In the end, we
included the Embrace participation, race and language spoken
(as a proxy for ethnicity), age, parity, insurance status (as a proxy
for socioeconomic status), preeclampsia, and diabetes.
Embrace Participation
We used Embrace’s internal program dataset to identify the
names, dates of birth, and delivery dates for the Embrace program
participants during the same period. Those names were cross-
matched with the Emory Decatur Hospital clinical records to
identify the patients who had participated in the pregnancy
support program. We created a new variable called “Embrace
Participation” where the participants were designated 1, and the
non-participants were designated 0. There were 113 Embrace
participants in the sample.
Race and Language Spoken
We operationalized race/ethnicity as the patients’ race and their
primary language spoken at home—both demographics that were
captured in the electronic medical record system. Race included
White, Black, Asian, American Indian, Hawaiian/Pacific Islander,
Hispanic, other, and unknown. The language variable included
over 50 different languages, which we categorized into English,
Burmese/Karen, African languages, Arabic/Egyptian/Aramaic,
and other/unknown. Race and language were highly colinear,
so we created a composite variable Race-Language with
seven categories: White-non-Arabic, Black-African, Black-Other
(reference because this was the largest subsample), Asian-
Burmese/Karen, Spanish-speaking, Asian-Nepali, Asian-Other,
Arabic/Egyptian/Aramaic-speaking, and other/unknown race.
Additional Covariates
Age was measured as a continuous variable in the electronic
medical records. Parity—the number of pregnancies ending in
live births, stillborns, and miscarriages—was measured as a count
variable that included the current birth. There were 24 different
insurance statuses, which we classified into these categories:
public insurance (Medicaid/Medicare) or other (self-pay, private
insurance, military insurance, veteran’s insurance, or other). We
also included underlying maternal health conditions that are
associated with our pregnancy-related outcomes of interest—
pre-eclampsia and diabetes. The variables were dichotomous,
where having the condition was coded as 1 while not having the
condition was coded as 0.
Zip Code
To consider environmental and neighborhood effects on
maternal and child health, we controlled for the patient’s zip
code of residence. Over 300 zip codes were represented in the
Emory Decatur database. We classified these into a categorical
variable: Clarkston (reference), metro-Atlanta (coded as 2),
Georgia outside of metro-Atlanta (coded as 3), and outside of
Georgia (coded as 4).
Frontiers in Global Women’s Health | 3June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
Labor Induction
When patients are admitted to Labor and Delivery at Emory
Decatur Hospital, the provider orders indicate whether (1) the
patient was in spontaneous labor or (2) the patient was being
induced (i.e., labor is started through medicine like Pitocin). In
our dataset, spontaneous labor was coded as 0, and induction was
coded as 1.
Cesarean Delivery
When a patient gives birth at Emory Decatur Hospital, the nurses
record whether it was a vaginal delivery or a cesarean delivery
(i.e., by surgical operation). In our dataset, we coded vaginal
delivery as 0, and cesarean delivery as 1.
Gestational Age at Birth (and Full-Term)
The nurses also record the gestational age at birth, which is
measured in weeks and days—either estimated since the last
menstrual period of the patient or confirmed by ultrasound. In
our dataset, we measured gestational age as a continuous variable
in days. We also created a dichotomous variable for full-term
gestational age, which is set at 259 days or 37 weeks (19). We
coded all births before or at 259 as 0 (not full term), and then
coded all births after 259 days as 1 (full term).
Birthweight (and Low Birthweight)
The electronic medical record also includes the birthweight of
the baby in kilograms. In our analyses, we used birthweight
as a continuous measure, and then also created a dichotomous
variable for low birthweight, which is defined as <2.5 kg (19). We
coded low birthweight as 1, and anything at or above 2.5 kg as 0.
Breastfeeding Plans
When patients are admitted to the Labor and Delivery Unit
for birth, nurses ask how they plan to feed their babies. The
feeding intention of the patient is recorded in the electronic
medical record as exclusive breastfeeding, bottle feeding with
formula, or a mixture of both. For this study, we coded exclusive
breastfeeding as 1 and any other feeding plans as 0. Notably, this
variable measures the patient’s intention for breastfeeding, not the
actual behavior.
For our analyses, we first looked at descriptive statistics for
the Embrace participants and the comparison group. We then
tested those differences, using the appropriate bivariate tests. For
continuous predictors and outcomes (maternal age, gestational
age in days, and birthweight in kilograms), we used t-tests. For
categorical predictors and outcomes (race language, insurance,
pre-eclampsia, diabetes, zip code, labor induction, cesarean, full
term, low birthweight, and exclusive breastfeeding), we used chi-
squared tests. For parity, which is a count variable, we used a
bivariate Poisson regression.
We then conducted multivariate analyses to test for
differences between the Embrace participants and the
comparison group after controlling for covariates (race-language,
maternal age, parity, insurance status, preeclampsia, diabetes,
and zip code). For the continuous outcomes of gestational age at
birth and birthweight, we used multiple linear regression models.
For the dichotomous outcomes (labor induction, cesarean, low
birthweight, full term, and exclusive breastfeeding), we used
multiple logistic regression models. We then assessed marginal
effects of Embrace participation, using adjusted predicted
probabilities. We checked all assumptions for our regression
models, including multicollinearity. We also conducted Wald
tests to test the hypothesis that Embrace was associated with
changes in the maternal and child health outcomes.
Descriptive Statistics by Embrace
Demographics and Predictors
We analyzed 9,136 unique clinical records from Emory Decatur
Hospital (Table 1). The majority of the patients in the sample
(62.2%) were Black and spoke a language other than Arabic,
Spanish, or a Bantu African language—most often, English. On
average, the patients in our study were 29 years old, had two prior
births, and were on public insurance (65.8%). There was a small
but important minority of women who had pre-eclampsia (6.9%)
and diabetes (8.6%).
Out of the total sample, 113 (1.2%) patients had participated
in the Embrace program and 9,023 (98.8%) patients were used
for the comparison group. The two groups differed by race and
language spoken (χ²8=598.3, p<0.001), parity (bPoisson =0.4,
p<0.001), insurance (χ²1=45.1, p<0.001, and zip codes (χ²3
=355.3, p<0.001). The Embrace participants were more likely
to be Asian and speak Burmese/Karen or Black and speak African
language. They were also more likely to have higher parity, public
insurance, and live in Clarkston than the comparison group.
Pregnancy-Related Outcomes
There is evidence that the Embrace participants had better
pregnancy-related outcomes than the comparison group (see
Table 2). Labor was induced for 15.0% of the Embrace
participants compared with 24.7% of the comparison group
(p=0.02). Twenty-six percent of the Embrace participants had
cesarean deliveries compared with 33.8% of the comparison
group (p=0.07). Embrace babies had significantly higher
gestational age at birth: 277 days compared with 272 days
in the comparison group (p=0.01), and nearly 94% of the
Embrace babies were born full-term compared with 88.9% in the
comparison group (p=0.05). Only 5.3% of the Embrace babies
were low birthweight compared with 10.1% in the comparison
group (p=0.09). Finally, 62.9% of the Embrace participants
planned to exclusively breastfeed compared with 63% of the
comparison group (p=0.93).
Multivariate Analyses
Regression Models
Results from our multiple logistic regression models similarly
show some significantly improved pregnancy-related outcomes
for the Embrace participants. Table 3 includes our outcomes
of interest: labor induction, cesarean, full-term gestation, low
birthweight, and exclusive breastfeeding intentions. We have
included the total number of observations (n) and the model’s
fit-test statistics (likelihood ratio, LR, and chi-square). All models
Frontiers in Global Women’s Health | 4June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
TABLE 1 | Descriptive statistics of the Embrace participants and comparison group at Emory Decatur Hospital from 2016 to 2018.
Descriptor Comparison Embrace Total Statistic p-value
Total 9,023 98.76 113 1.24 9,136 100
Race and Language*** χ²8=598.30 <0.001
White Non-Arabic speaking 1,370 15.57 2 1.77 1,372 15.4
Black African language 307 3.49 37 32.74 344 3.86
Black-Other 5,537 62.94 5 4.42 5,542 62.2
Asian-Burmese/Karen 360 4.09 34 30.09 394 4.42
Spanish-speaking 325 3.69 0 0 325 3.65
Asian-Nepali 236 2.68 1 0.88 237 2.66
Asian-Other 367 4.17 12 10.62 379 4.25
Arabic-Speaking 85 0.97 9 7.96 94 1.05
Other/Unknown race 210 2.39 13 11.5 223 2.5
Maternal age 28.81 (M) 6.01 (SD) 28.89 (M) 6.20 (SD) 28.81 (M) 6.01 (SD) t9,124 = −0.16 0.88
Parity*** 2.24 (M) 1.45 (SD) 3.24 (M) 2.20 (SD) 2.26 (M) 1.47 (SD) b=0.37 <0.001
Insurance Status*** χ²1=45.1 <0.001
Other Insurance/Self-Pay 3,120 34.59 5 4.42 3,125 34.21
Public insurance: medicaid/medicare 5,901 65.41 108 95.58 6,009 65.79
Pre-Eclampsia624 6.92 3 2.65 627 6.86 χ²1=3.16 0.075
Accucheck 773 8.57 12 10.62 785 8.59 χ²1=0.60 0.44
Residence zip code*** χ²3=355.30 <0.001
Clarkston, Georgia 937 10.39 75 66.37 1,012 11.08
Metro-Atlanta, Georgia 7,945 88.07 38 33.63 7,983 87.4
Other Georgia 62 0.69 0 0 62 0.68
Outside of Georgia 77 0.85 0 0 77 0.84
The bparity coefficient is from a bivariate Poisson regression due to the count distribution of the parity variable; p<0.01; *p<0.05; **p<0.01; ***p<0.001.
TABLE 2 | Bivariate statistics by Embrace participation/comparison group at Emory Decatur Hospital from 2016 to 2018.
Outcome Comparison Embrace Total Statistic p-value
Labor induction* 2,225 24.66 17 15.04 2,242 24.54 χ²1=5.57 0.02
Cesarean delivery3,041 33.76 29 25.66 3,070 33.66 χ²1=3.27 0.07
Gestational age (days)* 271.69 (M) 18.33 (SD) 276.54 (M) 10.02 (SD) 271.75 (M) 18.26 (SD) t9,009 = −2.76 0.01
Full term gestation7,929 87.88 106 93.81 8,035 87.95 χ²1=3.7 0.05
Birthweight (kilos)3.17 (M) 0.75 (SD) 3.30 (M) 0.49 (SD) 3.17 (M) 0.75 (SD) t8,662 = −1.82 0.07
Low birthweight909 10.07 6 5.31 915 10.02 χ²1=2.81 0.09
Exclusive breastfeeding 5,191 63.33 61 62.89 5,252 63.32 χ²1=0.01 0.93
p<0.01; *p<0.05; **p<0.01; ***p<0.001.
were statistically significant (p<0.001), indicating we can reject
the null hypothesis that all predictors in the models are equal
to zero.
Relative to the comparison group, the Embrace participants
had 48% lower odds of labor induction (OR =0.52, p=
0.025), and 65% higher odds of planning to breastfeed exclusively
(OR =1.65, p=0.028). The first row corresponds with
Embrace participation (our independent variable). Following
rows correspond with our covariates: race-language, maternal
age, parity, insurance status, preeclampsia, diabetes, and zip
code. Other results, while not statistically significant, also showed
trends in the positive direction. The Embrace participants had
17% lower odds of cesarean delivery (OR =0.83, p=0.411),
49% higher odds of full-term gestational age (OR =1.49, p=
0.329), and 23% lower odds of low birthweight (OR =0.77, p=
0.552). Not shown in Table 3, our results also showed improved
but non-significant improvements in gestational age at birth and
birthweight: a 2.26-day increase in gestational age at birth (b=
2.26, p=0.204) and a 0.06 kg increase in birthweight (b=0.06,
Holding other variables at their mean, we found that
Embrace participation was significantly associated with a 9-
percentage point reduction in labor induction from 23% (95% CI:
22.7–24.5%) to 14% (95% CI: 0.07–20.6%) and a 10-percentage
Frontiers in Global Women’s Health | 5June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
TABLE 3 | Multivariate analyses of pregnancy-related outcomes for the Embrace participants and the comparison group at Emory Decatur Hospital from 2016 to 2018.
Variable Labor induced* Cesarean delivery Full term gestation Low birthweight Exclusive breastfeeding*
n=8,871 n=8,860 n=8,871 n=8,871 n=8,071
LR chi2(17) =44.56 LR chi2(17) =394.11 LR chi2(17) =494.16 LR chi2(17) =422.59 LR chi2(17) =1,051.76
p= ≤0.001 p= ≤0.001 p= ≤0.001 p= ≤0.001 p= ≤0.001
OR p95% CI OR p95% CI OR p95%CI OR p95% CI OR p95% CI
Embrace participation 0.52 0.025 0.29 0.92 0.83 0.411 0.53 1.30 1.49 0.329 0.67 3.34 0.77 0.552 0.33 1.82 1.65 0.028 1.06 2.57
Race and language (reference: black-english)
White, non-Arabic 0.82 0.014 0.70 0.96 0.67 <0.001 0.57 0.77 1.69 <0.001 1.33 2.14 0.47 <0.001 0.35 0.62 3.60 <0.001 2.92 4.44
Black-African 0.65 0.003 0.48 0.87 0.89 0.392 0.69 1.15 1.67 0.019 1.09 2.56 0.61 0.041 0.38 0.98 0.99 0.931 0.76 1.28
Asian-Burmese 0.49 <0.001 0.36 0.67 0.58 <0.001 0.44 0.76 1.72 0.013 1.12 2.65 0.43 0.001 0.25 0.72 0.93 0.564 0.73 1.19
Any-Spanish 0.79 0.076 0.61 1.03 0.78 0.047 0.61 1.00 1.36 0.109 0.93 1.97 0.60 0.023 0.39 0.93 1.03 0.797 0.81 1.32
Asian-Nepali 0.69 0.043 0.48 0.99 1.05 0.73 0.78 1.41 2.20 0.009 1.22 3.96 0.79 0.37 0.48 1.32 0.58 <0.001 0.43 0.78
Asian-Other 0.81 0.118 0.63 1.05 0.77 0.03 0.61 0.97 1.06 0.746 0.75 1.49 1.05 0.803 0.73 1.50 1.14 0.292 0.89 1.47
Any, Arabic 1.10 0.699 0.69 1.74 1.09 0.715 0.70 1.68 1.89 0.12 0.85 4.20 0.93 0.834 0.45 1.89 1.11 0.646 0.71 1.74
Other/unknown race 0.61 0.007 0.43 0.88 0.89 0.434 0.66 1.19 1.36 0.214 0.84 2.19 0.66 0.137 0.38 1.14 1.59 0.005 1.15 2.20
Maternal age 1.04 <0.001 1.03 1.05 1.06 <0,001 1.05 1.07 1.01 0.123 1.00 1.02 0.99 0.077 0.97 1.00 1.04 <0.001 1.03 1.05
Parity 1.17 <0.001 1.13 1.22 0.91 <0,001 0.88 0.94 0.88 <0.001 0.84 0.92 1.05 0.057 1.00 1.11 0.77 <0.001 0.74 0.80
Medicaid insurance 1.13 0.058 1.00 1.28 1.23 0.001 1.09 1.38 1.05 0.528 0.89 1.25 1.20 0.054 1.00 1.44 0.49 <0.001 0.43 0.56
Pre-Eclampsia 1.63 <0.001 1.36 1.95 2.29 <0,001 1.93 2.72 0.17 <0.001 0.14 0.20 5.71 <0.001 4.73 6.89 0.91 0.328 0.75 1.10
Diabetes 1.36 <0.001 1.16 1.61 1.61 <0,001 1.37 1.88 0.55 <0.001 0.45 0.68 1.28 0.039 1.01 1.63 0.77 0.004 0.65 0.92
Zipcode (reference: clarkston)
Metro-Atlanta 1.02 0.878 0.83 1.24 0.97 0.769 0.81 1.16 0.90 0.448 0.68 1.19 1.18 0.276 0.88 1.60 0.96 0.660 0.80 1.15
Other Georgia 1.60 0.121 0.88 2.91 0.96 0.901 0.53 1.75 0.79 0.595 0.33 1.87 1.50 0.388 0.60 3.75 1.65 0.212 0.75 3.62
Outside Georgia 1.34 0.310 0.76 2.34 0.87 0.615 0.50 1.50 0.63 0.203 0.30 1.29 2.05 0.06 0.97 4.35 0.68 0.19 0.39 1.21
*p<0.05 for Embrace participation.
Frontiers in Global Women’s Health | 6June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
FIGURE 1 | Marginal effects of Embrace participation on the probability of pregnancy-related outcomes at Emory Decatur Hospital from 2016 to 2018 after adjusting
for covariates. *p<0.05; probabilities are adjusted for race, language spoken at home, parity, maternal age, insurance status, underlying conditions (pre-eclampsia
and diabetes), and zip code.
point increase in exclusive breastfeeding plans (66 vs. 76%)
(Figure 1). Our results also showed non-significant but positive
trends on other outcomes. Embrace participation was associated
with a 4-percentage point decrease in cesarean delivery from
33% (95% CI: 32.1–34.1%) to 29% (95% CI: 19.8–38.2%), a 4-
percentage point increase in full-term gestational age from 89%
(88.6–90%) to 93% (95% CI: 87.1–98.1%), and a 2-percentage
point decrease in low birthweight from 9% (95% CI: 8.1–9.4%)
to 7% (95% CI: 1.4–12.3%).
Our Wald tests indicate that we can reject the null hypothesis
that Embrace participation was not associated with lower
likelihood of labor induction (Wald test χ²1=5, p<0.025) and
greater likelihood of exclusive breastfeeding intentions (Wald test
χ²1=4.83, p<0.028). We cannot reject that null hypothesis for
Embrace participation and cesarean delivery (Wald test χ²1=
0.68, p=0.411), full-term gestational age (Wald test χ²1=0.85,
p=0.329), or low birthweight (Wald test χ²1=0.35, p=0.552).
Refugee women are particularly vulnerable to adverse pregnancy
outcomes (2). Community-based pregnancy programs represent
a promising strategy to address some of the challenges refugee
women face when giving birth in a new country. Our study
suggests that one such program, Embrace Refugee Birth Support
in Georgia, might be effectively improving pregnancy-related
outcomes for refugees, particularly reduced labor induction and
increased plans for exclusive breastfeeding. Labor induction is,
in turn, associated with higher gestational age and birthweight
as well as a lower risk of cesarean delivery (22). While exclusive
breastfeeding is associated with health benefits for mothers and
babies across their life courses (28).
A core value of Embrace is cultural sensitivity; through
their programming, education, and birth support, Embrace
seeks to integrate a refugee’s home culture into her birth
experience in America. Additionally, the Embrace volunteers
help alleviate many barriers women face in accessing and
navigating pregnancy-related services, particularly as it relates
to relationships and communication with providers during the
pregnancy, birth, and postpartum periods (40). As highlighted
by Khan and DeYoung (41), culturally sensitive programs
and strategies such as these are needed to assist refugee
women with accessing maternity services that can improve
outcomes. Furthermore, prior studies have demonstrated that
incorporating culturally sensitive strategies has been successful in
addressing barriers, promoting perinatal health service use, and
improving outcomes among pregnant women (29,30). Findings
from our study reinforce the importance of community-based,
pregnancy support programs to improve refugee maternal and
child health.
Given the benefits of breastfeeding to the health and
development of infants, providing culturally relevant education
on breastfeeding is important to improve breastfeeding-related
outcomes among resettled refugee populations (15,16). In
Embrace’s Healthy Babies and Breastfeeding module, community
liaisons teach women about the health benefits of breastfeeding
and navigating challenges with breastfeeding. While we are not
able to directly assess the relationship in our study, previous
studies have found women who attend prenatal classes focused
on breastfeeding had higher intentions to exclusively breastfeed
and were more likely to exclusively breastfeed, compared with
women who did not attend classes (42,43). This may explain the
greater intention to exclusively breastfeed among the Embrace
Education surrounding obstetrical interventions is critical for
refugee populations, given many refugee women come from
settings and cultures where such interventions are uncommon or
unfamiliar, which may lead to avoiding medical care and distrust
of providers (4446). As found in another study, childbirth
education may be an effective strategy for reducing elective
inductions (47). Consistent with this finding, our study found
Frontiers in Global Women’s Health | 7June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
that women who participated in Embrace had a significantly
lower likelihood of labor induction, which, in turn, supports
the findings that the women who participated in Embrace
might have higher gestational age and birthweight as well as
a lower risk of cesarean delivery. Furthermore, during many
births, an Embrace volunteer is present with the moms to
provide continuous support during labor and childbirth and
to advocate for them and serve as a bridge between patient
and provider, especially in instances where interventions are
recommended. Our results align with findings from a systematic
review that found women with continuous support during
birth, including from doulas, were more likely to have a
spontaneous vaginal delivery and less likely to have a Cesarean
delivery (48). However, unlike our study, this review found
that there was no impact of continuous support on the use of
synthetic oxytocin (e.g., induction and augmentation) during
labor or breastfeeding.
Findings from our study have several implications for
providing services for pregnant refugee women and future
research. Participating in culturally sensitive and tailored support
services may have positive influences on refugee women’s
pregnancy outcomes and social support (19,44). Similar
programs can be developed to meet the pregnancy, birth, and
postpartum needs of diverse resettled refugee populations. The
Embrace participant sample was very diverse and can be used
to generalize to many racial/ethnic groups. The Embrace sample
was African (32.7%), Karen/Burmese (30.1%), other Asians who
were not Nepali (10.6%), and Arab refugees (8.0%). Furthermore,
our sample’s diverse countries of origin are comparable to the
resettled refugee population in the US. In Fiscal Year 2020, 35%
of refugees resettling in the US came from Africa, 18% from East
Asia, and 17% from Near East/South Asia (49). Therefore, our
findings may be generalizable to the broader resettled refugee
population in the US.
While the study findings align with Embrace’s program
components, future studies could assess Embrace’s program
evaluation data to discern possible mechanisms through which
the program is improving pregnancy-related outcomes. In
particular, this study was not able to assess how much doula
support women received from Embrace or how many prenatal
visits the Embrace and comparison women attended. Future
research should include dose-response measures to assess those
effects. Additionally, this was an innovative study, using big
data to rigorously evaluate a community-based pregnancy
support program for refugee women, whose health disparities
are particularly difficult to quantify. Previous studies typically
rely on qualitative designs or compare pregnancy outcomes
between refugee and U.S.-born or immigrant populations. Future
studies could adopt a similar design to quantitatively assess
the effectiveness of services across resettled refugee populations.
However, those studies will need to conduct careful power
calculations to ensure the sample sizes are adequate to detect
statistically significant results. Additionally, as found by several
studies, the lack of cultural awareness and receptiveness among
providers to refugee women’s beliefs and practices surrounding
birth creates disconnects between patients and providers and
feelings of fear and marginalization among refugee women
(41,44,50). In addition to birth education for refugee
women, Embrace works closely and establishes relationships
with healthcare providers in the community whom refugee
women regularly see. Through these relationships, Embrace has
worked to provide education, such as Lunch and Learns, for
providers to orient them to the refugee community and different
cultural practices. Future qualitative research is needed with
data collection at the organizational and community levels to
understand provider and community perspectives surrounding
providing pregnancy-related services for refugee women.
Our study demonstrates the potential success of Embrace in
pregnancy-related outcomes; however, important limitations
must be noted. For one, women were not randomized into the
Embrace program. Furthermore, causality cannot be established
because we were unable to control for additional potential
confounding factors or temporality. Nevertheless, we have
controlled for a number of covariates that could have possibly
explained the improved outcomes for the Embrace patients.
Another related limitation is that we have no definitive
data on the refugee status of women in the comparison
group. We have made the comparison group as similar to
the Embrace population as possible, but there could still be
underlying unmeasured differences between the two groups.
The use of secondary data from Emory Decatur Hospital
limited which predictors and outcomes we could assess for
this study. For example, data on length of time in the
U.S. were not available for analysis. By using the hospital
records, however, we were able to analyze a large sample,
including the Embrace participants and a comparison group,
using consistent, clinical measurements. Furthermore, due to
limitations on data availability, only data on breastfeeding
intentions were available rather than actual breastfeeding.
Because behavior intentions are the closest predictor of
actual behavior (51), however, this remains an important and
valuable breastfeeding indicator. Finally, these results might not
generalize to Spanish-speaking refugees in the US, given that
there were no Spanish-speaking Embrace participants in this
sample. Nevertheless, Embrace program evaluation data with the
Spanish-speaking participants, who delivered at other hospitals,
indicates the program is similarly valued by and beneficial
for Latinas.
Community-based, culturally tailored pregnancy support
programs like Embrace are needed to meet the complex
needs of refugee women, who are at greater risk of barriers
to maternal health services and negative birth outcomes.
In a time when anti-refugee social contexts and COVID-19
present additional challenges, programs such as Embrace
are especially needed to support refugee women when
navigating health services and giving birth in their new
country. Community-engaged, cross-sector research approaches
like the one we took in this study under the guidance
of Georgia Doula Access Working Group are needed to
ensure community and clinical perspectives are included
Frontiers in Global Women’s Health | 8June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
in research design, implementation, and dissemination
of results.
The data analyzed in this study is subject to the following
licenses/restrictions: Data come from electronic medical records
and cannot be made publicly available. Requests to access these
datasets should be directed to
The studies involving human participants were reviewed
and approved by Emory University IRB. Written informed
consent for participation was not required for this study
in accordance with the national legislation and the
institutional requirements.
HM, TN, and JC: pregnancy support program development and
implementation. EM, MP, HM, AM, and MH: study design. AM,
HM, TN, and JC: community engagement. EM, MP, HM, TN,
and JC: data collection. EM, MP, BW, and MH: data analysis.
EM, GB, BW, and LC: first draft of manuscript, revisions, and
second draft of manuscript. All authors contributed to the article
and approved the submitted version.
This work was also made possible through the support of
Collaborative for Gender and Reproductive Equity, a sponsored
project of Rockefeller Philanthropy Advisors.
The Georgia Doula Access Working Group, which oversaw
this study, was funded by the United Way of Greater Atlanta
through the Healthy Mothers, Healthy Babies Coalition of
Georgia. Dr. Mosley’s contributions were funded by the Center
for Reproductive Health Research in the Southeast (RISE) and
an anonymous foundation. The team also acknowledges Mary
M. Gullate Ph.D., RN, ANP-BC, AOCN, FAAN, Corporate
Director Nursing Research and Evidence Based Practice, Emory
Healthcare, Atlanta, Georgia for her research guidance and
1. Carolan M. Pregnancy health status of sub-Saharan refugee women who have
resettled in developed countries: a review of the literature. Midwifer. (2010)
26:407–14. doi: 10.1016/j.midw.2008.11.002
2. Wanigaratne S, Shakya Y, Gagnon AJ, Cole DC, Rashid M,
Blake J, et al. Refugee maternal and perinatal health in Ontario,
Canada: a retrospective population-based study. BMJ Open. (2018)
8:e018979. doi: 10.1136/bmjopen-2017-018979
3. Gagnon AJ, Merry L, Robinson C. A systematic review of
refugee women’s reproductive health. Refuge Can J Refug. (2002)
21:6–17. doi: 10.25071/1920-7336.21279
4. Gibson-Helm M, Teede H, Block A, Knight M, East C, Wallace
EM, et al. Maternal health and pregnancy outcomes among women
of refugee background from African countries: a retrospective,
observational study in Australia. BMC Pregnancy Childbirth. (2014)
14:392. doi: 10.1186/s12884-014-0392-0
5. Gibson-Helm M, Boyle J, Cheng IH, East C, Knight M, Teede H.
Maternal health and pregnancy outcomes among women of refugee
background from Asian countries. Int J Gynecol Obstet. (2015) 129:146–
51. doi: 10.1016/j.ijgo.2014.10.036
6. Gibson-Helm ME, Teede HJ, Cheng IH, Block AA, Knight M, East CE,
et al. Maternal health and pregnancy outcomes comparing migrant women
born in humanitarian and nonhumanitarian source countries: a retrospective,
observational study. Birth. (2015) 42:116–24. doi: 10.1111/birt.12159
7. Correa-Velez I, Ryan J. Developing a best practice model of refugee maternity
care. Women Birth. (2012) 25:13–22. doi: 10.1016/j.wombi.2011.01.002
8. Kentoffio K, Berkowitz SA, Atlas SJ, Oo SA, Percac-Lima S. Use of maternal
health services: comparing refugee, immigrant and US-born populations.
Matern Child Health J. (2016) 20:2494–501. doi: 10.1007/s10995-016-
9. United Nations High Commissioner for Refugees. Refugees in America.
Refugee Facts. (2020). Available online at:
10. Mossaad N, Baugh R. Refugees and Asylees: 2016. (2018). Available online
11. Pew Research Center. Key Facts About Refugees to the U.S. Fact Tank: News in
the Numbers. (2019). Available online at:
tank/2019/10/07/key-facts- about-refugees-to- the-u-s/
12. U.S. Department of Health and Human Services. Health and Human Services
FY 2018 Budget in Brief. (2017). Available online at: https://www.
13. Nawyn SJ. Lower Refugee Limits are Weakening resettLement in the US. The
Conversation. (2019). Available online at:
refugee-limits- are-weakening-resettlement- in-the-us- 124970
14. Gagnon AJ, Van Hulst A, Merry L, George A, Saucier JF, Stanger E, et al.
Cesarean section rate differences by migration indicators. Arch Gynecol
Obstet. (2013) 287:633–9. doi: 10.1007/s00404-012-2609-7
15. Bader LR, Ward J, Fouts HN, Jaekel J. Infant care practices among
resettled refugee mothers from east and central Africa. Children. (2020)
7:63. doi: 10.3390/children7060063
16. Steinman L, Doescher M, Keppel GA, Pak-Gorstein S, Graham E, Haq
A, et al. Understanding infant feeding beliefs, practices and preferred
nutrition education and health provider approaches: an exploratory study
with Somali mothers in the USA. Matern Child Nutr. (2010) 6:67–
88. doi: 10.1111/j.1740-8709.2009.00185.x
17. DeSisto CL, McDonald JA. Variation in birth outcomes by mother’s country
of birth among hispanic women in the United States, 2013. Public Health Rep.
(2018) 133:318–28. doi: 10.1177/0033354918765444
18. Urquia ML, Glazier RH, Blondel B, Zeitlin J, Gissler M, Macfarlane A, et al.
International migration and adverse birth outcomes: role of ethnicity, region
of origin and destination. J Epidemiol Community Health. (2010) 64:243–
51. doi: 10.1136/jech.2008.083535
19. Heslehurst N, Brown H, Pemu A, Coleman H, Rankin J. Perinatal health
outcomes and care among asylum seekers and refugees: a systematic review of
systematic reviews. BMC Med. (2018) 16:89. doi: 10.1186/s12916-018-1064-0
20. Merry L, Small R, Blondel B, Gagnon AJ. International migration and
caesarean birth: a systematic review and meta-analysis. BMC Pregnancy
Childbirth. (2013) 13:27. doi: 10.1186/1471-2393-13-27
21. Gregory KD, Jackson S, Korst L, Fridman M. Cesarean versus vaginal
delivery: whose risks? Whose benefits? Am J Perinatol. (2012) 29:7–
18. doi: 10.1055/s-0031-1285829
Frontiers in Global Women’s Health | 9June 2021 | Volume 2 | Article 655409
Mosley et al. Evaluating Refugee Pregnancy Support Program
22. Ehrenthal DB, Jiang X, Strobino DM. Labor induction and the risk of a
cesarean delivery among nulliparous women at term. Gestation Diabetes.
(2010) 116:8. doi: 10.1097/AOG.0b013e3181e10c5c
23. Liu S, Liston RM, Joseph KS, Heaman M, Sauve R, Kramer MS. Maternal
mortality and severe morbidity associated with low-risk planned cesarean
delivery versus planned vaginal delivery at term. Can Med Assoc J. (2007)
176:455–60. doi: 10.1503/cmaj.060870
24. Jarjour IT. Neurodevelopmental outcome after extreme prematurity:
a review of the literature. Pediatr Neurol. (2015) 52:143–
52. doi: 10.1016/j.pediatrneurol.2014.10.027
25. Sengupta S, Carrion V, SheltonJ, Wynn RJ, Ryan RM, Singhal K, et al. Adverse
neonatal outcomes associated with early-term birth. JAMA Pediatr. (2013)
167:1053. doi: 10.1001/jamapediatrics.2013.2581
26. Engle WA, Kominiarek MA. Late preterm infants, early term
infants, and timing of elective deliveries. Clin Perinatol. (2008)
35:325–41. doi: 10.1016/j.clp.2008.03.003
27. Li CI, Daling JR, Emanuel I. Birthweight and risk of overall and cause-
specific childhood mortality. Paediatr Perinat Epidemiol. (2003) 17:164–
70. doi: 10.1046/j.1365-3016.2003.00487.x
28. Edmond K, Bahl R. Optimal Feeding of Low-Birth-Weight Infants: Technical
Review. World Health Organization (2007).
29. Rogers HJ, Hogan L, Coates D, Homer CSE, Henry A. Responding to
the health needs of women from migrant and refugee backgrounds—
models of maternity and postpartum care in high-income countries: a
systematic scoping review. Health Soc Care Community. (2020) 28:1343–
65. doi: 10.1111/hsc.12950
30. Thomas AB, Gieszl S, Nizigiyimana J, Johnson-Agbakwu C. Experiences
of refugee women in accessing and utilizing a refugee-focused pre- natal
clinic in the United States: a mixed methods study. Global Womens Health.
(2018) 1:14–20. Available online at:
uploads/2016/07/GWH-01- 000102.pdf
31. LaMancuso K, Goldman RE, Nothnagle M. “Can I ask that?”: perspectives
on perinatal care after resettlement among karen refugee women, medical
providers, community-based doulas. J Immigr Minor Health. (2016) 18:428–
35. doi: 10.1007/s10903-015-0172-6
32. Owens C, Dandy J, Hancock P. Perceptions of pregnancy experiences when
using a community-based antenatal service: a qualitative study of refugee and
migrant women in Perth, Western Australia. Women Birth. (2016) 29:128–
37. doi: 10.1016/j.wombi.2015.09.003
33. Riggs E, Muyeen S, Brown S, Dawson W, Petschel P, Tardiff W, et al.
Cultural safety and belonging for refugee background women attending
group pregnancy care: an Australian qualitative study. Birth. (2017) 44:145–
52. doi: 10.1111/birt.12272
34. Embrace Refugee Birth Support. Embrace Refugee Birth Support.
Embrace Refugee Birth Support. (2020). Available online at: https://
35. Evidence Based Birth. Belly Blooms: Jenny Cochran. Evidence Based Birth.
(2016). Available online at:
36. Healthy Mothers, Healthy Babies Coalition of Georgia. Healthy Mothers,
Healthy Babies Working Groups. Healthy Mothers, Healthy Babies Coalition of
Georgia. (2020). Available online at:
37. U.S. Census Bureau. Most of the Counties with Largest Population Gains
Since 2010 are in Texas. (2020). Available online at:
newsroom/press-releases/2020/pop- estimates-county-metro.html
38. Annie E. Casey Foundation. Changing the Odds: Progress and Promise in
Atlanta. (2019).
39. Department of Public Health Maternal Mortality Review Committee.
Maternal Mortality Report 2014. (2019). Available online at: https://dph.
40. Fuller TJ, Leonard M, Cochran J, Hutchins RJ, Brioche R, Kitiabi H, et al.
Collaborative development and implementation of a video-based curriculum
expansion to improve refugee women’s birth experiences in the United States.
Health Promot.Pract. (2020) 22:177–80. doi: 10.1177/15248399209
41. Khan A, DeYoung S. Maternal health services for refugee
populations: exploration of best practices. Glob Public Health. (2019)
14:362–374. doi: 10.1080/17441692.2018.1516796
42. Tender JA, Janakiram J, Arce E, Mason R, Jordan T, Marsh J, et al. Reasons
for in-hospital formula supplementation of breastfed infants from low-
income families. J Hum Lact. (2009) 25:11–7. doi: 10.1177/089033440832
43. McDonald SD, Pullenayegum E, Chapman B, Vera C, Giglia L, Fusch C, et al.
Prevalence and predictors of exclusive breastfeeding at hospital discharge.
Obstet Gynecol. (2012) 119:1171–9. doi: 10.1097/AOG.0b013e31825
44. Pangas J, Ogunsiji O, Elmir R, Raman S, Liamputtong P, Burns E, et al.
Refugee women’s experiences negotiating motherhood and maternity care
in a new country: a meta-ethnographic review. Int J Nurs Stud. (2019)
90:31–45. doi: 10.1016/j.ijnurstu.2018.10.005
45. Brown E, Carroll J, Fogarty C, Holt C. “They get a C-section they gonna
die”: somali women’s fears of obstetrical interventions in the United States.
J Transcult Nurs. (2010) 21:220–7. doi: 10.1177/1043659609358780
46. Agbemenu K, Banke-Thomas A, Ely G, Johnson-Agbakwu
C. Avoiding obstetrical interventions among US-based somali
migrant women: a qualitative study. Ethn Health. (2019)
1–16. doi: 10.1080/13557858.2019.1613519
47. Simpson KR, Newman G, Chirino OR. Patient education to
reduce elective inductions. Am J Mater Nurs. (2010) 35:188–
94. doi: 10.1097/NMC.0b013e3181d9c6d6
48. Bohren MA, Hofmeyr GJ, Sakala C, Fukuzawa RK, Cuthbert A. Continuous
support for women during childbirth. Cochrane Database Syst Rev. (2017)
7:CD003766. doi: 10.1002/14651858.CD003766.pub6
49. Refugee Processing Center. FY2020 Arrivals by Region. (2020). Available
online at:
50. Wojnar DM. Perinatal experiences of somali couples in the
United States. J Obstet Gynecol Neonatal Nurs. (2015) 44:358–
69. doi: 10.1111/1552-6909.12574
51. Montano DE, Kasprzyk D. Theory of reasoned action, theory of planned
behavior, and the integrated behavioral model. In: Health Behavior and
Health Education: Theory, Research, and Practice. Jossey-Bass (2015).
p. 95–124.
Conflict of Interest: The authors declare that the research was conducted in the
absence of any commercial or financial relationships that could be construed as a
potential conflict of interest.
Copyright © 2021 Mosley, Pratt, Besera, Clarke, Miller, Noland, Whaley, Cochran,
Mack and Higgins. This is an open-access article distributed under the terms
of the Creative Commons Attribution License (CC BY). The use, distribution or
reproduction in other forums is permitted, provided the original author(s) and the
copyright owner(s) are credited and that the original publication in this journal
is cited, in accordance with accepted academic practice. No use, distribution or
reproduction is permitted which does not comply with these terms.
Frontiers in Global Women’s Health | 10 June 2021 | Volume 2 | Article 655409
Maternal health in the United States is an area of immediate concern. The compounded disadvantages and unique global positions of refugee women highlight the need for research that explores the experiences of refugee women during pregnancy and childbirth. The present study examines how contextual factors shape pregnancy and childbirth experiences for Syrian, Afghan, Congolese, and Karen women living in Clarkston, Georgia, US. Qualitative data were collected via focus groups facilitated by community interpreters. We used a reproductive justice framework to center women's desires, needs, and experiences, and to highlight the importance of structural factors in the findings and analysis of this study. Using codebook thematic analysis, three themes were developed: (1) isolation and alienated knowledge, (2) gendered disparities and structural inequities, and (3) community support and precarity. The findings reflect both the diversity and constancy of women's experiences and highlight how the context of the US impacts women's ability to exercise agency during pregnancy and childbirth. Systemic change is needed to improve women's access to tools that increase their capacity to exercise agency both during pregnancy and childbirth, and beyond.
Full-text available
Refugees often parent under extreme circumstances. Parenting practices have implications for child outcomes, and parenting in the context of refugee resettlement is likely to be dynamic as parents negotiate a new culture. This study examined African origin mothers’ infant care values and practices related to feeding, carrying, and daily activities following resettlement in the Southeastern region of the U.S. Ten African origin mothers were asked about their infant care practices through semi-structured interviews. Results indicated that mothers valued breastfeeding but often chose to use formula as a supplement or instead of breastfeeding. In addition, participants valued carrying their infants close to the body but used equipment such as strollers. Mothers expressed that perceptions of American culture and rules, social support, interactions with community agencies, and the need to engage in formal employment were factors that influenced their infant care practices.
Full-text available
Many refugee and immigrant women in the United States experience cultural and language barriers when seeking pregnancy-related medical care. Such barriers may delay needed care and adversely impact birth outcomes. Embrace Refugee Birth Support (Embrace) in Clarkston, Georgia, supports pregnant refugee women by offering birth education classes in the women’s primary languages. Our academic–practice partnership designed and implemented a series of birth education videos for Embrace participants. Based on input from former participants, the partnership team created video scenarios that could be embedded into Embrace’s existing didactic curriculum. The videos addressed common challenges and learning needs identified by previous participants. All videos were filmed in the participant’s primary languages (Swahili and Kinyarwanda) and featured actual Embrace graduates who spoke the languages. Then, Embrace trainers used the video scenarios to augment teaching on birth preparedness and foster participant discussion during class sessions. After implementation, a focus group with participants in the video-expanded class reported the videos were well received, understood, and practically related to their pregnancy needs. Overall, participants reported that video scenarios were an important part of their learning and skill development, as well as a positive experience. Embrace has plans to continue creating native language educational videos for additional languages and birth-related topics. The academic partner’s attempts to measure video impact with standardized quantitative instruments at baseline were terminated. The substantive revisions in data collection strategies highlight the need for cross-cultural flexibility and the potential for unforeseen barriers when using quantitative research tools among non–English-speaking participants.
Full-text available
Pregnant women from migrant and refugee backgrounds living in high‐income countries (HIC) are at increased risk of adverse perinatal outcomes compared with women born in the host country. Women from migrant and refugee background have perinatal healthcare needs that are recognised internationally as a public health priority. The aim of this study was to identify, appraise and synthesise available evidence on the effectiveness of models of care in pregnancy or first 12 months postpartum for women from migrant and refugee backgrounds living in HIC. Care models were mapped in terms of (a) effectiveness at improving service access, (b) effectiveness at improving maternal and infant health outcomes, (c) acceptability and appropriateness from the perspective of women and (d) acceptability and appropriateness from the perspective of service providers. Using systematic scoping review methodology, qualitative, quantitative, and mixed methods research published in English 2008–2019 were included. The databases MEDLINE, Embase, Emcare, PubMed, Scopus, CINAHL, PsycINFO, Web of Science, Google Scholar, Cochrane Database of Systematic Reviews and Joanna Briggs Institute were searched between 27 February 2019 and updated 27 December 2019. Qualitative and quantitative data were analysed narratively. Seventeen studies, involving 1,499 women and 203 service providers, were included. A diverse range of interventions were identified, including bilingual/bicultural workers, group antenatal care and specialised clinics. All identified interventions were acceptable to women, and improved access, however, few provided evidence of improved perinatal outcomes. Gaps identified for future research include the use of qualitative and quantitative approaches to ascertain the experiences of women, their families, service providers and impact on perinatal outcomes. Synthesis of the included studies indicates the key elements of acceptable and accessible models, which were as follows: culturally responsive care, continuity of care, effective communication, psychosocial and practical support, support to navigate systems, flexible and accessible services.
Full-text available
The full scope of women’s health needs is not necessarily addressed in refugee camps and after resettlement, particularly pregnancy and postnatal services. The aims of this research are to examine the maternal care services provided to refugee women in camps and after resettlement to the United States, and to analyse organisational successes and challenges in service provision. With this understanding, policies can improve service delivery for refugee women. We interviewed respondents from five organisations, ranging from local non-profits to international non-governmental organisations. Most of the organisations do not provide direct medical care, but rather education and social service support to clients, and in some cases midwife training. Their success stemmed from a focus on client capacity building, individualised support, effective partnerships, and cultural competency. Respondents described the need for physical resources, effective leadership, and additional personnel, especially with linguistic capabilities. The dialogue in the interviews supports themes of education as empowerment and client self-sufficiency. Respondents emphasised the importance of funding and policies that support their work. This knowledge can lead to improved models of service delivery and inform the development of best practices and policies in maternal and reproductive health for refugee women.
Full-text available
Background: Global migration is at an all-time high with implications for perinatal health. Migrant women, especially asylum seekers and refugees, represent a particularly vulnerable group. Understanding the impact on the perinatal health of women and offspring is an important prerequisite to improving care and outcomes. The aim of this systematic review was to summarise the current evidence base on perinatal health outcomes and care among women with asylum seeker or refugee status. Methods: Twelve electronic database, reference list and citation searches (1 January 2007-July 2017) were carried out between June and July 2017. Quantitative and qualitative systematic reviews, published in the English language, were included if they reported perinatal health outcomes or care and clearly stated that they included asylum seekers or refugees. Screening for eligibility, data extraction, quality appraisal and evidence synthesis were carried out in duplicate. The results were summarised narratively. Results: Among 3415 records screened, 29 systematic reviews met the inclusion criteria. Only one exclusively focussed on asylum seekers; the remaining reviews grouped asylum seekers and refugees with wider migrant populations. Perinatal outcomes were predominantly worse among migrant women, particularly mental health, maternal mortality, preterm birth and congenital anomalies. Access and use of care was obstructed by structural, organisational, social, personal and cultural barriers. Migrant women's experiences of care included negative communication, discrimination, poor relationships with health professionals, cultural clashes and negative experiences of clinical intervention. Additional data for asylum seekers and refugees demonstrated complex obstetric issues, sexual assault, offspring mortality, unwanted pregnancy, poverty, social isolation and experiences of racism, prejudice and stereotyping within perinatal healthcare. Conclusions: This review identified adverse pregnancy outcomes among asylum seeker and refugee women, representing a double burden of inequality for one of the most globally vulnerable groups of women. Improvements in the provision of perinatal healthcare could reduce inequalities in adverse outcomes and improve women's experiences of care. Strategies to overcome barriers to accessing care require immediate attention. The systematic review evidence base is limited by combining heterogeneous migrant, asylum seeker and refugee populations, inconsistent use of definitions and limited data on some perinatal outcomes and risk factors. Future research needs to overcome these limitations to improve data quality and address inequalities. Systematic registration: Systematic review registration number: PROSPERO CRD42017073315 .
Full-text available
Objectives Immigrants are thought to be healthier than their native-born counterparts, but less is known about the health of refugees or forced migrants. Previous studies often equate refugee status with immigration status or country of birth (COB) and none have compared refugee to non-refugee immigrants from the same COB. Herein, we examined whether: (1) a refugee mother experiences greater odds of adverse maternal and perinatal health outcomes compared with a similar non-refugee mother from the same COB and (2) refugee and non-refugee immigrants differ from Canadian-born mothers for maternal and perinatal outcomes. Design This is a retrospective population-based database study. We implemented two cohort designs: (1) 1:1 matching of refugees to non-refugee immigrants on COB, year and age at arrival (±5 years) and (2) an unmatched design using all data. Setting and participants Refugee immigrant mothers (n=34 233), non-refugee immigrant mothers (n=243 439) and Canadian-born mothers (n=615 394) eligible for universal healthcare insurance who had a hospital birth in Ontario, Canada, between 2002 and 2014. Primary outcomes Numerous adverse maternal and perinatal health outcomes. Results Refugees differed from non-refugee immigrants most notably for HIV, with respective rates of 0.39% and 0.20% and an adjusted OR (AOR) of 1.82 (95% CI 1.19 to 2.79). Other elevated outcomes included caesarean section (AOR 1.04, 95% CI 1.00 to 1.08) and moderate preterm birth (AOR 1.08, 95% CI 0.99 to 1.17). For the majority of outcomes, refugee and non-refugee immigrants experienced similar AORs when compared with Canadian-born mothers. Conclusions Refugee status was associated with a few adverse maternal and perinatal health outcomes, but the associations were not strong except for HIV. The definition of refugee status used herein may not sensitively identify refugees at highest risk. Future research would benefit from further refining refugee status based on migration experiences.
Full-text available
Background: Historically, women have generally been attended and supported by other women during labour. However, in hospitals worldwide, continuous support during labour has often become the exception rather than the routine. Objectives: The primary objective was to assess the effects, on women and their babies, of continuous, one-to-one intrapartum support compared with usual care, in any setting. Secondary objectives were to determine whether the effects of continuous support are influenced by:1. Routine practices and policies in the birth environment that may affect a woman's autonomy, freedom of movement and ability to cope with labour, including: policies about the presence of support people of the woman's own choosing; epidural analgesia; and continuous electronic fetal monitoring.2. The provider's relationship to the woman and to the facility: staff member of the facility (and thus has additional loyalties or responsibilities); not a staff member and not part of the woman's social network (present solely for the purpose of providing continuous support, e.g. a doula); or a person chosen by the woman from family members and friends;3. Timing of onset (early or later in labour);4. Model of support (support provided only around the time of childbirth or extended to include support during the antenatal and postpartum periods);5. Country income level (high-income compared to low- and middle-income). Search methods: We searched the Cochrane Pregnancy and Childbirth Group's Trials Register (31 October 2016),, the WHO International Clinical Trials Registry Platform (ICTRP) (1 June 2017) and reference lists of retrieved studies. Selection criteria: All published and unpublished randomised controlled trials, cluster-randomised trials comparing continuous support during labour with usual care. Quasi-randomised and cross-over designs were not eligible for inclusion. Data collection and analysis: Two review authors independently assessed trials for inclusion and risk of bias, extracted data and checked them for accuracy. We sought additional information from the trial authors. The quality of the evidence was assessed using the GRADE approach. Main results: We included a total of 27 trials, and 26 trials involving 15,858 women provided usable outcome data for analysis. These trials were conducted in 17 different countries: 13 trials were conducted in high-income settings; 13 trials in middle-income settings; and no studies in low-income settings. Women allocated to continuous support were more likely to have a spontaneous vaginal birth (average RR 1.08, 95% confidence interval (CI) 1.04 to 1.12; 21 trials, 14,369 women; low-quality evidence) and less likely to report negative ratings of or feelings about their childbirth experience (average RR 0.69, 95% CI 0.59 to 0.79; 11 trials, 11,133 women; low-quality evidence) and to use any intrapartum analgesia (average RR 0.90, 95% CI 0.84 to 0.96; 15 trials, 12,433 women). In addition, their labours were shorter (MD -0.69 hours, 95% CI -1.04 to -0.34; 13 trials, 5429 women; low-quality evidence), they were less likely to have a caesarean birth (average RR 0.75, 95% CI 0.64 to 0.88; 24 trials, 15,347 women; low-quality evidence) or instrumental vaginal birth (RR 0.90, 95% CI 0.85 to 0.96; 19 trials, 14,118 women), regional analgesia (average RR 0.93, 95% CI 0.88 to 0.99; 9 trials, 11,444 women), or a baby with a low five-minute Apgar score (RR 0.62, 95% CI 0.46 to 0.85; 14 trials, 12,615 women). Data from two trials for postpartum depression were not combined due to differences in women, hospitals and care providers included; both trials found fewer women developed depressive symptomatology if they had been supported in birth, although this may have been a chance result in one of the studies (low-quality evidence). There was no apparent impact on other intrapartum interventions, maternal or neonatal complications, such as admission to special care nursery (average RR 0.97, 95% CI 0.76 to 1.25; 7 trials, 8897 women; low-quality evidence), and exclusive or any breastfeeding at any time point (average RR 1.05, 95% CI 0.96 to 1.16; 4 trials, 5584 women; low-quality evidence).Subgroup analyses suggested that continuous support was most effective at reducing caesarean birth, when the provider was present in a doula role, and in settings in which epidural analgesia was not routinely available. Continuous labour support in settings where women were not permitted to have companions of their choosing with them in labour, was associated with greater likelihood of spontaneous vaginal birth and lower likelihood of a caesarean birth. Subgroup analysis of trials conducted in high-income compared with trials in middle-income countries suggests that continuous labour support offers similar benefits to women and babies for most outcomes, with the exception of caesarean birth, where studies from middle-income countries showed a larger reduction in caesarean birth. No conclusions could be drawn about low-income settings, electronic fetal monitoring, the timing of onset of continuous support or model of support.Risk of bias varied in included studies: no study clearly blinded women and personnel; only one study sufficiently blinded outcome assessors. All other domains were of varying degrees of risk of bias. The quality of evidence was downgraded for lack of blinding in studies and other limitations in study designs, inconsistency, or imprecision of effect estimates. Authors' conclusions: Continuous support during labour may improve outcomes for women and infants, including increased spontaneous vaginal birth, shorter duration of labour, and decreased caesarean birth, instrumental vaginal birth, use of any analgesia, use of regional analgesia, low five-minute Apgar score and negative feelings about childbirth experiences. We found no evidence of harms of continuous labour support. Subgroup analyses should be interpreted with caution, and considered as exploratory and hypothesis-generating, but evidence suggests continuous support with certain provider characteristics, in settings where epidural analgesia was not routinely available, in settings where women were not permitted to have companions of their choosing in labour, and in middle-income country settings, may have a favourable impact on outcomes such as caesarean birth. Future research on continuous support during labour could focus on longer-term outcomes (breastfeeding, mother-infant interactions, postpartum depression, self-esteem, difficulty mothering) and include more woman-centred outcomes in low-income settings. -- This review is published as a Cochrane Review in the Cochrane Database of Systematic Reviews 2017, Issue 7. Cochrane Reviews are regularly updated as new evidence emerges and in response to comments and criticisms, and the Cochrane Database of Systematic Reviews should be consulted for the most recent version of the Review."
Objective: Somali refugee women are known to have poor health-seeking behavior with a higher proportion of adverse pregnancy outcomes compared to US-born women. Yet unknown is how they avoid obstetrical interventions. This study sought to identify perceived protective mechanisms used to avoid obstetric interventions as well as the underpinning factors that influence aversion to obstetrical interventions by Somali refugee women. Design: A descriptive, exploratory qualitative study purposively sampled Somali refugee women recruited via snowball technique in Franklin County, Ohio, United States. Data were collected through audio-recordings of individual interviews and focus groups conducted in English and Somali languages. The collected data were transcribed and analyzed using thematic analyses. Results: Forty Somali refugee women aged 18–42 years were recruited. Participants reported engaging in four perceived protective mechanisms to avoid obstetrical interventions during pregnancy and childbirth: (1) intentionally not seeking or misleading prenatal care, (2) changing hospitals and/or providers, (3) delayed hospital arrival during labor, and (4) refusal of care. Underpinning all four avoidance mechanisms were their significant fear of obstetrical interventions, and perceived lack of choice in their care processes as influenced by cultural and/or religious beliefs, feeling judged or undervalued by service providers, and a lack of privacy provided to them while receiving care. Conclusion: Like every woman, Somali women also have a right to choose or refuse care. If the intention is to improve access to and experiences with care for this population, building trust, addressing their fears and concerns, and respecting their culture is a critical first step. This should be well established prior to the need for critical decisions surrounding pregnancy and childbirth wherein Somali women may feel compelled to refuse necessary obstetrical care. Bridging gaps between Somali women and their providers is key to advance health equity for this vulnerable population.
Objectives: The aim of this meta-ethnographic review was to examine refugee women's experiences negotiating motherhood and maternity services in a new country with a view to identifying the specific needs of refugee women accessing maternity care in high income countries. Design: A meta-ethnographic synthesis of qualitative research. Data sources: Five databases were searched for papers published in English between January 2000 and January 2017. Review methods: The synthesis process was guided by the seven steps of meta-ethnography. The quality of included studies was assessed using the COREQ tool. Results: One overarching theme and three major themes emerged from the synthesis. The overarching theme "Living between two cultures" conveyed women's experience of feeling "in between" cultures and described refugee women's experience of striving to maintain a strong cultural identity from their country of origin while simultaneously adapting to their new context and country. This theme permeated the following three major themes: 1) "Constructing maternal identity across cultures" which discusses the cultural conflict experienced by refugees accessing maternity services in their host country; 2) "Understanding in practice" which describes reciprocal issues in communication between women and health professionals; and 3) "Negotiating care" which illustrates a mix of coping mechanisms which refugee women utilise to navigate health services in the context of high income countries. Conclusion: Liminality is a ubiquitous experience for refugee women seeking maternity care in high income countries. It impacts feelings of belonging and connection to services and society. It is often a challenging experience for many women and a time in which they reformulate their identity as a citizen and a mother. This review found that the experience of liminality could be perpetuated by social factors, and inequality of healthcare provision, where communication and cultural barriers prevented women accessing care that was equal, accessible, and meaningful. Findings revealed both positive and negative experiences with maternity care. Continuity, culturally appropriate care, and healthcare relationships played an important role in the positive experiences of women. The review also revealed the damaging effects of disparities in care experienced by refugee women.
Objectives: Despite knowledge that the Hispanic population is growing in the United States and that birth outcomes may vary by maternal country of birth, data on birth outcomes by maternal country of birth among Hispanic women are scant. We compared the rates of 3 birth outcomes for infants born in the United States-preterm birth, low birth weight, and small for gestational age-between foreign-born Hispanic women and US-born Hispanic women, and then we examined these birth outcomes by mother's country of birth for foreign-born Hispanic women. Methods: Using the 2013 natality file from the National Vital Statistics System of the National Center for Health Statistics, we examined data on the 3 birth outcomes and maternal characteristics by maternal country of birth. We used log binomial models to calculate unadjusted and adjusted relative risks for preterm birth, low birth weight, and small for gestational age for US-born Hispanic women compared with foreign-born Hispanic women. We also compared the relative risk of each adverse birth outcome for foreign-born Hispanic women by country of birth. Results: US-born Hispanic women had higher rates of the 3 birth outcomes than did foreign-born Hispanic women (preterm birth: 8.0% vs 7.0%; low birth weight: 6.1% vs 5.2%; small for gestational age: 9.2% vs 7.9%). These higher rates persisted after adjusting for maternal characteristics. The rates for these 3 birth outcomes varied significantly by country of birth for foreign-born Hispanic women, with Puerto Rican women consistently having the poorest birth outcomes. Conclusions: Our results demonstrated heterogeneity in rates of adverse birth outcomes by country of birth for foreign-born Hispanic women. Presenting rates for foreign-born mothers as a group masks differences by country. To understand possible changes in data on birth outcomes, states should stratify data by maternal country of birth.