Effectiveness of an Enhanced Community Doula Intervention in a Safety Net Setting: A Randomized Controlled Trial

1Department of Obstetrics and Gynecology, Boston Medical Center, Boston, Massachusetts, USA.

2Department of Pediatrics, Oregon Health & Science University, Portland, Oregon, USA.

3Department of Biostatistics, Boston University School of Public Health, Boston, Massachusetts, USA.

1Department of Obstetrics and Gynecology, Boston Medical Center, Boston, Massachusetts, USA.

1Department of Obstetrics and Gynecology, Boston Medical Center, Boston, Massachusetts, USA.

1Department of Obstetrics and Gynecology, Boston Medical Center, Boston, Massachusetts, USA.

4Department of Community Health Sciences, Boston University School of Public Health, Boston, Massachusetts, USA.

Racial inequities in maternal health outcomes, the result of systemic racism and social determinants of health, require maternity care systems to implement interventions that reduce disparities. One such approach may be support from a community doula, a health worker who provides emotional support, peer education, navigation, and advocacy for pregnant, birthing, and postpartum people who share similar racial identities, cultural backgrounds, and/or lived experiences. While community support during birth has a long tradition within communities of Black Indigenous and People of Color (BIPOC), the reframing of community doula support as a social intervention that reduces disparities in clinical outcomes is recent.

We conducted a pragmatic randomized trial at an urban safety net hospital, comparing standard maternity care with standard care plus enhanced community doula support. We tested the effectiveness of a community doula program embedded in a safety net hospital in improving birth outcomes and explored the association between community doula support and health equity. Participants were nulliparous, insured by publicly funded health plans, and had lower risk pregnancies. The primary outcome was cesarean birth. Secondary outcomes included preterm birth and breastfeeding outcomes. Exploratory subgroup analysis was conducted by race–ethnicity.

Three hundred sixty-seven participants were included in the primary analysis. In the intent-to-treat analysis, outcomes were similar between groups. There was a trend toward increased breastfeeding initiation (p=0.08). There was a statistically nonsignificant 12% absolute reduction in cesarean birth and 11.5% increase in exclusive breastfeeding during delivery hospitalization among Black non-Hispanic participants.

While outcomes for the study sample were similar between randomization groups, health outcomes were improved for Black birthing people in cesarean and breastfeeding rates.

This study demonstrates the need for larger studies of community doula support for Black birthing people. Clinicaltrials.gov ID: {"type":"clinical-trial","attrs":{"text":"NCT02550730","term_id":"NCT02550730"}}NCT02550730.

Racial inequities in maternal health outcomes are rooted in a long history of systemic racism, sexism, and classism in the United States.1–3 Black birthing people have higher rates of maternal morbidity, mortality, postpartum depression, and poor experience of maternity care and lower rates of breastfeeding than other racial groups.4–11 These outcomes are a result of centuries of macrolevel policies, institutional practices, and cultural norms that reinforce one another and continue to perpetuate social, economic, and political disadvantages for Black individuals and communities.1,2

Structural barriers to health experienced by Black people in the reproductive years include inequitable access to quality health care, housing, employment, education, community resources, and fair policing for themselves and their families.12 Within maternity care services, Black pregnant, birthing, and postpartum people report higher levels of obstetric violence, disrespect and dismissal, and withholding of information about their care.13

Black birth workers have always provided essential holistic health care to their communities in the United States.14 From the 17th to 19th centuries, enslaved Black midwives practiced the skills and traditions of African midwifery in caring for both other enslaved Black people and the White relatives of slave owners.14 In the Jim Crow era, “Grand” midwives in the South (elder Black community midwives) provided essential maternal and infant health care to Black communities.15

The American Medical Association's campaign to regulate the practice of medicine at the turn of the 20th century led to legal and regulatory limitations on the practice of community midwifery, including for Southern Black Grand midwives.15 At the same time, the rise of obstetrics as a field of medicine required training and practice opportunities for physicians, who were primarily White, upper class, and male.15 Childbirth was reframed as a medical event and moved rapidly into hospitals over the mid-20th century, eliminating community midwifery and support for Black birthing people.15

Community doulas have reclaimed this support role over the last few decades.14,16,17 As culturally congruent health workers, they share similar lived experiences and racial, cultural, and other intersectional identities with their client. Doulas provide physical and emotional support during pregnancy, childbirth, and the postpartum period. In the prenatal and postpartum periods, community doulas assist with navigation of health care and social services, provide peer education, and give social support. During labor, they provide continuous presence to promote physical comfort and support the birthing person's emotional needs.

Community doulas practice within a framework of birth justice,16,17 an aspect of reproductive justice, which names “the human right to maintain personal autonomy…” for Black birthing people.17 Doulas accomplish this through a variety of approaches.16–19 They navigate clients through resources essential for healthy social determinants of health (SDoH) such as housing, employment, nutritious food, and health care services that are less accessible compared with White pregnant and birthing people due to structural racism.13,19 They provide affirming and nonjudgmental support that may buffer the effects of relationship stressors, discrimination, and inadequate social support.16,17 In addition, they serve as advocates by amplifying the voice of the birthing person during labor and birth.16,17

A growing literature frames outcomes of community doula support as efficacious in reducing cesarean births,18,20 which may impact morbidity and mortality in both current and future pregnancies.21 Additionally, some studies show that doulas increase breastfeeding and improve the experience of maternity care.18,20,22 For low-income people and Black, Indigenous, People of Color (BIPOC), doulas provide a sense of physical and emotional safety, reduce stress related to experiences of discrimination in health care, and amplify the voice of the pregnant person.16,17

The effectiveness of community doula programs integrated into maternity care has not been studied. To understand the effectiveness of a doula program in improving health outcomes in a racially diverse low-income setting, a pragmatic, randomized controlled trial was conducted. The aims of the trial were to evaluate the effectiveness of doula support in reducing rates of cesarean and preterm births, as well as on breastfeeding outcomes. Additionally, the study aimed to explore the impact of the doula on health equity for Black birthing people, specifically focusing on cesarean birth outcomes.

We conducted a parallel-group, pragmatic single-center trial with 1:1 randomization to assess the impact of community doula support on cesarean birth, preterm birth, and breastfeeding outcomes for nulliparous, lower-risk pregnant people with public insurance coverage. The study was conducted at an urban safety net hospital serving a racially and linguistically diverse population of ∼2700 births per year, of which 85% are publicly financed.

The institutional review board of the study site approved this study on April 29, 2015. Participants provided written informed consent. The study follows the Consolidated Standards of Reporting Trials (CONSORT) guidelines. We submitted our registration to Clinicaltrials.gov on June 23, 2015 ({"type":"clinical-trial","attrs":{"text":"NCT02550730","term_id":"NCT02550730"}}NCT02550730). We enrolled participants from August 2015 through November 2017. Data collection occurred through June 2018.

After completing the baseline survey, participants were randomized 1:1 in blocks of eight to either the Birth Sisters Best Beginnings for Babies (BBB) enhanced doula intervention or routine care. Computer-generated randomization was performed by an outside statistician and group allocation was placed in sequentially numbered opaque envelopes. The envelope was opened by the research assistant in front of the participant. The sequence was not revealed until enrollment was complete. Follow-up survey assessors were blinded to the allocation assignment of each participant.

People pregnant with their first child, insured by Medicaid or other public insurance, and between 16 and 24 weeks of pregnancy were eligible. Research assistants assessed eligibility during routine ultrasound visits and obtained informed consent. Exclusion criteria were age Open in a separate windowFIG. 1.

Flow diagram.

Three hundred thirteen of those who remained in the analysis for birth outcomes also completed a postpartum survey (85%). One hundred sixty of those participants completed the survey before 12 weeks postpartum. One hundred fifty-three completed the survey after 12 weeks of giving birth. The proportion of participants completing the survey before 12 weeks was similar between groups (51.6% in the intervention group compared with 52.6% in the control group).

We follow standard guidelines for reporting randomized trials.29 Our design achieved balanced groups, as noted in Table 1. Baseline characteristics for those who were lost to follow-up, dropped out, or excluded were also balanced between groups (Supplementary Table S1). All participants qualified for MassHealth, with an income of 200% of the federal poverty level. To measure differences in SDoH beyond income strata, we report baseline data on housing, food, and energy security,30–32 as well as social isolation.33

Baseline Characteristics

OB, obstetrician; SD, standard deviation.

Of 187 people included in the intervention group analysis, 172 (92%) participants received a prenatal visit, 142 (76%) received labor support, and 132 (71%) had a Birth Sister postpartum visit. Primary reasons for no labor support are as follows: 14 delivered at another hospital, 1 was lost to follow-up, 9 declined services after enrollment in the study, 4 delivered precipitously before the Birth Sister arrived, 12 gave birth without the Birth Sister being notified, and 5 were unknown.

The mean number of prenatal meeting hours was 5.3 (range 0–18.8), mean number of hours of labor support was 10.7 (range 0–25.5), and mean number of postpartum meeting hours was 3.1 (range 0–12.5). There was no difference in program fidelity by individual Birth Sisters. The mean caseload by Birth Sister was 13.2 (range 1–34) over the 28 months of enrollment. One hundred twenty-seven (91%) participants were matched with a doula who was racially congruent.

In the control group, 25 participants received a referral for the hospital's Birth Sisters Program through their clinical provider, although these participants did not receive the full BBB intervention, including MLP | Boston, since they were not enrolled in the intervention arm of the study. Thirteen received a Birth Sister prenatal visit, five had Birth Sister labor support, and eight had a Birth Sister postpartum visit. Typical program uptake is more reflective of the BBB intervention uptake described above.

As shown in Table 2, there were no significant difference between randomization groups in the primary outcome of cesarean birth (p=0.72). Overall breastfeeding initiation rates were high, but there was a trend (defined as a p-value of ≤0.1) toward increased breastfeeding initiation (p=0.08). Differences in Apgar scores Table 2.

Intent-to-treat Outcomes

Exploratory Analysis for Non-Hispanic Black Race/Ethnicity

Per-protocol Analysis

The authors thank the participants, the Birth Sisters, and the Medical Legal Partnership | Boston for their contributions to this study. Medical Legal Partnership received grant funding for their work.

J.M.-S. was involved in conceptualization (lead), funding acquisition (lead), formal analysis (equal), writing—original draft (lead), and writing—review and editing (equal); D.D. was involved in conceptualization (supporting), writing—original draft (supporting), and writing—review and editing (equal); H.C. was involved in methodology (supporting) and writing—review and editing (equal); D.R. and L.S. were involved in review and editing (equal); E.A.V. was involved in software (lead), data curation (lead), formal analysis (equal), and writing—review and editing (equal); and E.F. was involved in conceptualization (supporting), methodology (lead), writing—original draft (supporting), and writing—review and editing (equal).

No competing financial interests exist.

Funding for the study was provided by the W.K. Kellogg Foundation [Project number: P3030989]. They had no role in the design of the study, in collection, analysis, and interpretation of data, in the writing of the report, or in the decision to submit the article for publication.

Supplementary Table S1

Cite this article as: Mottl-Santiago J, Dukhovny D, Cabral H, Rodrigues D, Spencer L, Valle EA, Feinberg E (2023) Effectiveness of an enhanced community doula intervention in a safety net setting: a randomized controlled trial, Health Equity 7:1, 466–476, DOI: 10.1089/heq.2022.0200.