An Integrative Review: The Role of the Doula in Breastfeeding Initiation and Duration

Abstract

The objective of this integrative review was to assess birth and postpartum doulas' roles in supporting breastfeeding initiation and duration. The electronic databases Cumulative Index to Nursing and Allied Health Literature (CINAHL), Embase, PubMed, and Scopus were searched using the key terms doula and breastfeeding. Fourteen articles met inclusion criteria. Six key themes were identified. Doulas may acquire only modest amounts of lactation-specific education; however, doula care still enhances the breastfeeding care provided by health-care professionals. Doulas offer prenatal and intrapartum support that encourages breastfeeding initiation in the hospital, as well as providing breastfeeding support in the community and home settings. This reinforces the unique role of the doula in bolstering maternal–infant health. The effect of doulas on breastfeeding duration is less clear.

Human milk is the optimal source of infant nutrition. Infants receive passive immunity via the receipt of maternal immune factors and anti-inflammatory cytokines, as well as decreased risk of infections (otitis media, respiratory infections), decreased incidence of necrotizing enterocolitis, and decreased risk of type II diabetes in adulthood (American Academy of Pediatrics, 2012; Ballard & Morrow, 2013; Lawrence & Pane, 2007). For the mother, breastfeeding provides an opportunity to bond with her child, decreases postpartum blood loss, stimulates uterine involution, aids in postpartum weight loss, and decreases the risk of future breast and ovarian cancer (American Academy of Pediatrics, 2012). A lactating mother's milk also varies in composition to meet the evolving needs of the infant, rendering human milk uniquely suited to the developmental and immunological needs of the child it is made to nourish (Ballard & Morrow, 2013).

For many years, childbirth educators have served as advocates to protect the role which lactation plays in the natural childbearing cycle. Over the past few decades, the medical and research communities have given increasing attention to breastfeeding as well. The U.S. Centers for Disease Control and Prevention (CDC) reiterates the World Health Organization's (WHO) recommendation for human milk as the exclusive source of infant nutrition throughout the first 6 months of life (CDC, 2019; WHO, 2011).

The U.S. government also includes breastfeeding goals among the Healthy People Goals, published every decade. Healthy People 2020 goals aimed at improving maternal–infant health by improving breastfeeding rates included increasing the percentage of infants born in the United States who have ever been breastfed to 81.9%, increasing the percentage of infants still breastfed at 6 months of life to 60.6%, increasing the percentage of infants who were exclusively breastfed though the first 6 months of life to 25.5%, and reducing the number of infants who received infant formula supplements within the first two days of life to 14.2% (Office of Disease Prevention and Health Promotion [ODPHP], n.d.).

National achievement of the Healthy People 2020 goal for breastfeeding initiation was slightly surpassed and the goal for breastfeeding exclusivity through 6 months was nearly met (CDC, 2018). However, stark disparities in rates of breastfeeding initiation and exclusivity through 6 months are revealed when the data is segregated and analyzed by race/ethnicity and socioeconomic status (socioeconomic status as gauged by eligibility for the Supplemental Nutrition Program for Women, Infants, and Children) (CDC, 2019). This reveals that there is still much work to be done in increasing breastfeeding rates in the United States, particularly among low-income and minority populations.

One approach to improving breastfeeding rates is to increase the support provided to pregnant and breastfeeding mothers. A Cochrane synthesis found that extra, organized support of breastfeeding mothers by trained volunteers and community health workers positively influences rates of breastfeeding duration and exclusivity (Mcfadden et al., 2017). This Cochrane synthesis was not included in our integrative review as it did not specifically examine the role of doulas.

As childbirth and postpartum support professionals who reach clients at perinatal phases salient to breastfeeding, doulas may be a form of organized support implicated in the solution to improving low and disparate breastfeeding rates. Therefore, it is important to know if, how, and when doulas influence breastfeeding initiation and duration. An understanding of the doula's role in supporting breastfeeding may also be important to ensuring effective collaboration between doulas and traditional members of the maternal–infant health-care team.

A 2017 Cochrane systematic review of doula care across 17 countries found doulas to have no apparent impact on breastfeeding rates (Bohren et al., 2017). However, some studies conducted in the United States prior to 2010 had revealed that doulas in the United States do improve breastfeeding rates (Mottl-Santiago et al., 2008; Nommsen-rivers et al., 2009). Therefore, the goals of this integrative review were to: (a) explore literature from the past decade to determine if birth and postpartum doula care impacts rates of breastfeeding initiation and duration in the United States and (b) explore the role (if one exists) of the birth and postpartum doula in supporting breastfeeding initiation and duration.

METHODS

In this integrative review, the authors followed the methodological steps outlined by Whittemore and Knafl in 2005. Since the task of reviewing a pool of literature featuring diverse methodologies can be difficult and therefore prone to error, this methodology was proposed as a way to decrease bias and enhance rigor of integrative reviews (Whittemore & Knafl, 2005). The authors also utilized the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) for additional guidance during the review process (Liberati et al., 2009).

Data Sources

A Boolean literature search was conducted across PubMed, CINAHL, Embase, and Scopus using the key words “breastfeeding” and “doulas.” The initial search yielded a result of 121 records: 30 from PubMed, 42 from CINAHL, 31 from Embase, and 18 from Scopus. After removing duplicates, 84 records remained. Throughout this phase of the review, the PRISMA guidelines were utilized (Liberati et al., 2009); please see Figure 1.

Figure 1. Preferred reporting items for systematic reviews and meta-analyses (PRISMA) review process flow diagram.

Screening and Inclusion Criteria

The first author screened the records' abstracts and full texts for relevance and compliance with inclusion criteria. Overall, articles were included if they were published between January 2010 and January 2020, were written in English, were conducted in the United States or had a U.S. practice focus, discussed doulas and their relationship to breastfeeding, and/or discussed doulas and their relationship to the maternal health-care team.

Of the 84 records, 5 were removed due to lack of availability in full-text formats and one was removed due to its status as an ongoing study without current reportable results. The goal of this integrative review was to evaluate the doula's role in supporting breastfeeding initiation and duration among women living in the United States. Thus, the 18 non-U.S. articles were excluded. Five more were excluded due to unavailability of English full-text formats.

Eight articles were removed for discussing breastfeeding without discussing the doula's involvement, while six were excluded for lack of discussion on the doula's role in supporting families with breastfeeding initiation and duration. Additionally, 10 records that were not from peer reviewed journals were excluded from this review. Seventeen articles were excluded due to containing only limited information or discussion on doulas and their relationship to breastfeeding; records excluded for this reason were categorized as “insufficient evidence” (see Figure 1).

Fourteen articles met criteria for inclusion and are discussed in this review.

Analytic Strategy

The first author scored the 14 articles for level of evidence using the framework developed by Melnyk and Fineout-Overholt and the second author reviewed and verified scoring (Melnyk & Fineout-Overholt, 2019). See Table 1 for Table of Evidence. The analytic strategy employed in this review involved first extracting methodological data from the articles. Extracted data included data relating to the study locations, number of participants, characteristics of interventions, statistical outcomes, and any provided characteristics of doulas and participants (i.e., race, age, socioeconomic status). The data was organized into a table and then compared and categorized. Comparison of categories of data led to the identification of themes which the authors saw reflected throughout the literature.

TABLE 1. Characteristics of Included Articles.

Author, Year	Article Type and Melnyk and Fineout-Overholt Level of Evidence Grade	Location, Population	Methods	Interventions	Outcomes or conclusion
Ahlemeyer and Mahon (2015)	Literature Review(VII)	United States, no specific region	Review and discussion of birth and postpartum doulas	N/A	Research strongly suggests the potential for doulas to have positive effects on birth outcomes and initiation and duration of breastfeeding. Different certifying bodies require varying amounts of lactation training of doula certification candidates. Across four major doula certifying organizations, the minimum amount of lactation training required for certification ranged from only that which was covered in the doula training course or workshop to completion of at least one book on human lactation from a preselected reading list and attendance at a 3-hour breastfeeding class. This yields doulas with varying degrees of preparedness to provide breastfeeding support to clients. Health-care providers such as maternity nurses should be knowledgeable about doulas and their role in breastfeeding and birth support (and vice versa).
Cattelona et al. (2015)	Case study(VI)	Bloomington, Indiana	Through a community-based birth services program which provides comprehensive birth and postpartum services to breastfeeding mothers, mothers are connected with birth and postpartum doulas. Clients are low income with mental health issues. One client's experience with prenatal and postpartum doula intervention was recorded and reported in this case study.	Mothers are connected with one volunteer birth doula who provides two prenatal visits and is on call for the client's labor and birth. During prenatal visits, birth doulas help mother's prepare for postpartum challenges. The birth doula remains with the client until the first feeding has been established, and then returns to the client for one postpartum visit. The client is then paired with at least one postpartum doula who provides ongoing, in-home breastfeeding and general postpartum support for as many hours as possible/needed.	Doulas accompanied the mother in the case study to various appointments and served as a constant from hospital to home setting (birth and eventually postpartum doulas both provided support in the hospital and postpartum doulas provided support at home). The mother breastfed for 4 months and stopped after the fourth month due to concerns about newly prescribed medications for depression passing into breastmilk.
Deubel et al. (2019)	Qualitative pilot research study; VI	Tampa, Florida; Participants (n = 20) were at least 18 years old and less than 3 months postpartum at the time of enrollment in the study	Twenty postpartum women were interviewed regarding their perceptions and practices related to breastfeeding; Kroeger's medical anthropological framework was applied to understand what factors influenced African American mother's decision-making in regard to breastfeeding	N/A	Factors contributing to mother's perceptions and practices related to breastfeeding include service-related factors (such as clinic wait times, services available, social relations within the clinic setting, etc.). Some factors such as personal or cultural beliefs, expectations about breastfeeding outcomes, individual characteristics, etc.
					were predisposing factors to breastfeeding perceptions that influenced a women's breastfeeding decision-making, while other factors such as socioeconomic status, cost, availability of affordable breast pumps, receipt of paid maternity leave, etc., were enabling factors to breastfeeding. Women were more likely to partially or exclusively breastfeed if they were supported by peers/family and/or a doula, and had positive prenatal education regarding breastfeeding.
Edwards et al. (2013)	Randomized controlled trial; II	Unnamed major urban University hospital in the United States; Low-income, African American mothers (n = 248) under age 21 years	Pregnant women were recruited from clinics affiliated with the urban University at which the study was conducted. Eligible participants who enrolled in the study were randomized to a group receiving standard, clinic-based prenatal health-care services or an intervention group receiving standard services and a community home-visiting doula intervention. Outcomes data were gathered by a combination of chart review and interviews with the mothers. Data was analyzed using X2 test of association.	Women in the intervention group received the services of one of four program doulas. All four doulas were African American and came from the communities surrounding the clinics from which the participants were recruited. Doulas were all experienced in counseling pregnant adolescents, three of the four had personal breastfeeding experience, and three of the four had been adolescent mothers.	Compared to mothers in the control group, women who received the home-visiting doula intervention were significantly more likely to initiate breastfeeding (63.9% compared to 49.6%). Differences in breastfeeding duration between the two groups were unclear, and both groups of mothers had similar, significantly low rates of breastfeeding by 4 months postpartum (8.3% of intervention group mothers and 4.4% of control group mothers).
			Differences between control and intervention groups in occurrence of breastfeeding attempts at the hospital, breastfeeding duration (classified as “never,” “less than 6 weeks,” “six weeks to four months”, and “longer than four months”), and timing of cereal or solid food introduction were analyzed.	All doulas were prepared with a 10-week breastfeeding peer counselor training and a 20-week doula training course. Doulas provided weekly prenatal home visits which focused on development of a relationship with the participant, time spent listening to mothers' breastfeeding related fears or concerns, identification and elimination of breastfeeding myths held by mothers, and normalizing of breastfeeding. Doulas provided labor support at the hospital and immediate postpartum breastfeeding encouragement and support. They also provided an average of 12 postpartum home visits over a three-month period and were available 24/7 by phone. Doulas also encouraged mothers to refrain from premature introduction of cereals or solid foods to the infant's diet. Doulas were provided with a doula-trained registered nurse as an emergency consult should they need one while engaged in program work.
Gruber et al. (2013)	Case-control (nonexperimental) study; IV	Greensboro, North Carolina, United States; pregnant adolescents and women between the ages of 13 and 30 (n = 226)	Study followed a non-experimental design, so clients were not randomized to control (nondoula) and intervention (doula) groups; instead, only mothers who attended at least 3 childbirth class sessions offered at YWCA Greensboro were eligible to be offered doula support. This resulted in two groups that were relatively similar in demographics (a majority of each group self-identified as African American—78.9% of nondoula vs. 77.3% of doula), but the average age of the nondoula group was slightly lower than the average age of the doula group (19.1 years in the nondoula group vs. 20.3 in the doula group). Nondoula mothers were also more likely to be living with family or a guardian (68.8% nondoula vs. 44.3% doula). Birth weight for the baby of one participant in the nondoula group was not recorded, thus the record was excluded. Comparative analysis using z-tests (p < .05) measured differences in outcomes between nondoula and doula group mothers. Outcomes compared were the incidence of low birth weight, type of birth, incidence of complications, and occurrence of initial breastfeeding attempts.	Pregnant women in the intervention group (n = 97) received pre-birth assistance from a certified doula. Doulas had also received continuing education and their services included a minimum of 2 prenatal visits, continuous support during labor and birth, and a minimum of two postpartum visits. Additionally, intervention and control (nondoula group; n = 126) group mothers were offered case management and an 8-week childbirth preparation class which followed a peer-support model and included information on breastfeeding and other infant and maternal health topics such as safe sleeping and nutrition. Doulas and clients were matched by characteristics such as doula's availability near mother's expected date of birth, language, race/ethnicity, and general personality.	Rates of breastfeeding initiation in the doula-assisted group of mothers was found to be significantly higher than rates of breastfeeding initiation in the nondoula group (79.4% in the doula group vs. 67.2 in the nondoula group).
Hans et al. (2018)	Randomized controlled trial; II	Illinois, United States. Participants were young, low-income mothers who identified as African American (n = 140) or Latina/Hispanic (n = 117).	Young mothers who met screening criteria for the study (no younger than 14 years of age and no older than 26 years of age, living in region served by one of four doula home-visiting programs, and not mentally impaired, not involved in the juvenile justice system) and agreed to study terms were randomly assigned to receive doula-home visiting services (intervention group) or case management (control group). Case management included family screening for need for intervention services for substance abuse, depression, and domestic violence, as well as referral to services for families found to be in want of basic needs such as food, housing, employment, education, and childcare needs. Baseline, 3-week postpartum, and 3-month postpartum interviews were conducted in the mother's preferred language (choice of English or Spanish). Interviews focused on eliciting information relating to Healthy People 2020 Maternal–Infant Health Goals (birth outcomes, vaginal or cesarean birth, infant birthweight and gestational age at birth, NICU admissions, length of hospital stay, mother or infant hospitalizations, and breastfeeding).	Mothers assigned to the intervention group received prenatal and postpartum visits from both a doula and a home visitor. The doulas also provided in-hospital support. Of the two visitors, doulas provided the majority of care related to health during pregnancy, childbirth, breastfeeding, early newborn care, bonding, and postpartum health while nondoula home visitors provided a majority of care relating to educational-work planning, child safety, child-development, mother–infant relationship, and screening of the family's basic needs.	Of the women interviewed at 37-weeks, the average number of visits these women received from a doula prior to the 37-week mark was 8.9 (with a SD = 6.9). Doulas attended 75% of births in the intervention group and by the three-month postpartum interview, doulas had provided at least one visit to 131 of the intervention group mothers. Mothers who received doula home-visiting services were more likely to have initiated breastfeeding than mothers in the control group (breastfeeding initiation was defined to mean breastfeeding at least throughout the duration of the hospital stay). Only about 20% of mothers in the intervention group were still breastfeeding at the 3-month postpartum interview.
Kozhimannil et al. (2013)	Retrospective cohort study; IV	Minnesota, United States; birth records of Minnesota Medicaid recipients who had elected to receive doula support from Everyday Miracles doulas from January 1st, 2010 to April 30th 2012 (n = 1069) and Pregnancy Risk Assessment Data (PRAMS) 2009 and 2010 data of Minnesota Medicaid beneficiaries who did not receive doula support (n = 51,721)	Breastfeeding initiation was treated as a dichotomous variable in both Everyday Miracles and PRAMS records. Rates of breastfeeding initiation for each group (Everyday Miracles doula group and PRAMS group) were calculated and reported using 95% confidence intervals. Breastfeeding initiation rates for each group were also calculated by race, which had been determined by participant self-report in Everyday Miracles birth records and PRAMS data (in Everyday Miracles birth records, race categories were mutually exclusive but in PRAMS data, they were not).	Women whose records comprised the data obtained from Everyday Miracles had received prenatal education, continuous labor support, postpartum visits, and breastfeeding education from DONA trained and/or certified doulas. Doulas came from diverse racial and ethnic backgrounds.	The overall breastfeeding initiation rate among women who had received doula support was significantly higher than the breastfeeding initiation rate of women who had not received doula support (97.9% in doula group vs. 80.8% in the PRAMS sample). Among each racial/ethnic group, breastfeeding initiation rates were also higher for the portion of the racial/ethnic group who received doula support than for the portion who did not (although there was insufficient data to confidently report on results among Native American, Asian, and other race/ethnic groups). African American women had the lowest rates of breastfeeding initiation, but showed a very high level of initiation in the doula support group (70.7% in nondoula vs. 92.7% in the doula group). Breastfeeding initiation rates for women who identified as white in the nondoula and doula groups were 78.7% and 98.2%, respectively.
					Rates among women who identified as Hispanic in the nondoula and doula groups were 92.0% and 99.2%, respectively, and rates among women of African descent in the doula and nondoula groups were 95.2% and 99.5%, respectively.
Richards and Lanning (2019)	Qualitative study; VI	Southeastern United States	Semistructured interviews were conducted with doulas who had provided support to women and newborns during or immediately after cesarean births	N/A	Doula engagement with clients undergoing cesarean included helping to cope with and relieve anxiety, support during placement of anesthesia, postpartum support, and supporting breastfeeding initiation within the first 60 to 90 minutes postcesarean birth
Shlafer et al. (2018)	Qualitative study; VI	Midwestern United States, Incarcerated women in a female-only prison, and the women's doulas	Participants completed a questionnaire regarding personal and incarceration statistics (e.g., length of sentence) upon enrollment in the study and doulas of participants took notes at each of the six meetings they had with participants. Doulas also kept track of and recorded breastfeeding initiation at each birth. Breastfeeding intention before birth was also recorded (as a dichotomous variable, either yes or no). Fisher's exact test and the Mann–Whitney U test were used to examine relationships between breastfeeding intention and variables such as race,	Incarcerated women attended a 12-week parenting support group and were visited by their doulas twice prenatally and twice postpartum, and they received continuous support during labor in the hospital.	Although less than half of the pregnant participants indicated a desire to breastfeed at the beginning of the program, more than half initiated breastfeeding after intervention visits with doulas
			age, prison sentence length, and whether or not the doula discussed breastfeeding with the client before the birth. Qualitative descriptive methodologies were used to examine doulas' experiences supporting incarcerated women to breastfeed.
Thurston et al. (2019)	Retrospective cohort study; IV	Jefferson County, Alabama, United States; records of all Medicaid-covered births in Jefferson county (n = 3,782) from 2013–2014 and records of women supported by BWPdoulas in Central Alabama from 2013–2014 (n = 120)	Birth records of Medicaid beneficiaries in Jefferson County, Alabama (reference cohort) and birth records of women who were supported by doulas from an organization in Central Alabama (study cohort) between January 2013 and December 2014 were analyzed using descriptive statistics. Because many of the women supported by BWP doulas were Medicaid beneficiaries, many of their birth and breastfeeding information was also included in the reference cohort. Descriptive statistics were used to calculate the percentage of each cohort that had low birth weight babies, preterm labors, inductions, epidural anesthesia, unscheduled Cesarean births, and the number of mother–infant dyads who initiated breastfeeding during the hospital stay.	All of the women whose birth records comprised the study cohort had been assisted by BWP doulas. The BWP doulas had taken a 3-day long, DONA International approved birth doula training workshop, a day long childbirth preparation class, a 4-hour class focused on issues of maternal–child health on a local and national level, and a 4-hour breastfeeding class. In total, 45 doulas supported the doula-assisted mothers. There were 36 White doulas, 18 Black doulas, and one Hispanic doula. Some doulas provided up to six prenatal visits to clients while others did not provide any at all, with the average number of prenatal visits being 1.9. All doulas provided continuous support throughout labor, and some doulas provided up to three postpartum visits while some did not provide any postpartum visits.	The women who were supported by BWP doulas were 10.5 times more likely to initiate breastfeeding in the hospital than the women who did not receive doula support. Black mothers who were supported by doulas were 7.8 times more likely to initiate breastfeeding than nondoula supported mothers, and White mothers who received doula support were 13 times more likely to initiate breastfeeding in the hospital than White non-doula-assisted mothers.
			After removing twin birth records (due to the nature of twins to be born preterm and at a low birth rate), percentages for each category were compared between the cohorts. The same data was also grouped and analyzed by race.	The average number of postpartum visits was 1.5. Both prenatal and postpartum visits ranged from 1 to 2 hours in length.
Torres (2013)	Qualitative study; VI	Michigan, United States; Internationally Board-Certified Lactation Consultants (n = 18) and DONA Certified Birth Doulas (n = 16)	The researcher interviewed the lactation consultants and birth doulas, transcribed, and analyzed the data from the interviews using open and focused coding. Although the data was published in 2013, the interviews were conducted in 2008 and 2011. Data from postpartum doulas was not included due to the limited interactions that postpartum doulas typically have with medical professionals.	N/A	Lactation consultants have been integrated into the maternity care team and thus are more likely to have the respect of their physician and other medical colleagues. Doulas, on the other hand, work outside of the maternity care team (though they often work with the maternity care team) in order to support and advocate for the natural birth wishes of their clients. The researcher found that because doulas are not always viewed as colleagues, they often have more work to do to earn the respect of maternity care team than a lactation consultant might.
Torres (2015a)	Qualitative study; VI	Michigan, United States; lactation consultants (n = 19), birth doulas (n = 18), clients/patients (n = 17), health-care professionals (doctors, nurses, and midwives; n = 18)	The researcher conducted 150 hours of ethnographic observation of lactation consultants over a period of 9 months. She did this by shadowing two lactation consultants who practiced in-hospital settings (including inpatient care and outpatient breastfeeding care clinics) and one who practiced in the community and clinic setting. Additionally, information gained in interviews with lactation consultants and doulas during a related Michigan-based project were used. The interviewed lactation consultants included 10 who practiced in the hospital setting only, three who worked in private practice only, three who worked in both private practice and hospital settings, and three who reportedly worked in “other” settings. The researcher recruited clients of the doulas and lactation consultants whose (previous interview data was being used in this study) to be interviewed for this study.	N/A	The researcher found that the separate spheres ideology can perpetuate relative low wages for paid caring work that doulas provide. Many doulas offer their services for free or on a sliding scale when their clients cannot afford the full cost of their services, and this can make it difficult for doulas to make their practice their full-time job, as they may struggle to earn livable wage. Lactation consultants were found to be in a similar predicament, yet made more per year on average than doulas, perhaps due to their more formal integration into the maternity care team.
Torres (2015b)	Qualitative study; VI	Michigan, United States; lactation consultants (n = 19), birth doulas (n = 18), clients/patients (n = 17), health-care professionals (doctors, nurses, and midwives; n = 18)	The same body of data was used for this article as for the previous one (Torres, 2015a). The researcher evaluated the body of research with the aim of understanding how participants viewed the roles of doulas and lactation consultants within the context of the larger maternity care system.	N/A	Participants were found to view the use of doulas and lactation consultants as outsourcing of care that had been traditionally provided by family.
					Due to the gendered and caring nature of lactation consultants' and doulas' work, they may be undervalued as care providers, as their work is often viewed as an innate quality of women. The researcher found that the skill and knowledge of both doulas and lactation consultants sets their work at a higher level of expertise than it would if it were simply the natural work of women. In the current, medicalized system of maternity care, the knowledge and skill of lactation consultants and doulas is deeper and more objective than that which would traditionally have been provided by families. This legitimizes the work of lactation consultants and doulas as a service worthy of a significant fee, yet many lactation consultants and doulas often feel conflict in charging for their services, especially when they know that clients' financial backgrounds may be strained.
					This leads to many doulas and private practice lactation consultants charging on a sliding scale and often doing pro bono work. While charitable, this is an unsustainable model of care, especially when the doula or lactation
					consultant is the sole income earner in a household. The dilemma created is one in which these professionals must decide whether to turn away clients who cannot afford services or lose needed household revenue themselves by putting aside time to take on such clients.
Zandt et al. (2016)	Retrospective cohort (IV)	John's Hopkins University in Baltimore, Maryland; 1,511 records of women who were served by BCs Program doulas in urban and suburban communities in a metropolitan area on the Eastern Coast of the United States	Birth records of 1,511 women who had been supported by BCs from 1998–2014 were reviewed. Clients were separated into mutually exclusive “vulnerable” (n = 522) and “nonvulnerable” (n = 989) groups based age, income, education level, English proficiency, and refugee status. The vulnerable group was also divided into five subgroups based on demographic factors that led to their “vulnerable” classification (teenagers, newly resettled refugees, non-English speakers, low-income, and low-educational level).	BCs—nursing students who were trained as birth doulas using a combination of classroom content and a DONA International approved birth doula workshop were provided to women who self-referred themselves to the BCs Program. BCs provided physical, emotional, and informational support to pregnant and laboring women through a combination of a prenatal visit, continuous labor support, and a postpartum visit.	Although breastfeeding attempts by nonvulnerable women supported by BCs were high (consistently above 80%), breastfeeding rates among vulnerable recipients of BCP doula care were lower than expected. Researchers speculated that this may be due to inadequate BCP doula breastfeeding training, lack of doula confidence in breastfeeding support skills, and limited time spent with postpartum clients.
			The records contained information about specific doula support services provided to each woman as well as demographic information, maternal health indicators, and newborn health indicators. Descriptive statistics was used to analyze records and chi-squared, t-, and Fisher's exact tests were used analyze and compare the birth outcomes/indicators from each group.

Note. BCs = birth companions; BWP = birth-well partners.

For studies which reported on outcomes beyond breastfeeding outcomes, only the breastfeeding outcomes and doula interventions were highlighted in this table.

RESULTS

Key Themes Identified

Six key themes relating to the doula's role in supporting breastfeeding initiation and duration stood out. They are as follows.

Impact of Doula Support on Breastfeeding Outcomes

The first key theme was that doula support has in fact been shown to significantly improve rates of breastfeeding initiation. The impact of doula support on breastfeeding duration is less clear, however (Edwards et al., 2013; Gruber et al., 2013; Hans et al., 2018; Kozhimannil et al., 2013; Thurston et al., 2019).

Lactation Training Acquired by Doulas

Second, it was observed that the amount of lactation-specific education which doulas acquire varies based on several factors: certification status, certifying organization, and lactation education requirements of community organization or study.

Certified doulas obtain lactation training per the requirements of their certifying bodies. Across four major certifying organizations, the amount of lactation education required ranged from only that which was covered during the doula training workshop to that which could be obtained from the combination of attendance at a 4-hour breastfeeding basics class and reading of an approved book on lactation (Ahlemeyer & Mahon, 2015).

Doulas who are neither certified nor pursuing certification have full discretion regarding the amount of lactation training they chose to pursue. They may also determine for themselves the credibility of the lactation information which they obtain. There are currently many practicing doulas who are not certified, the exact number of which is unknown; the amount of lactation training obtained by all such doulas is also unknown (Ahlemeyer & Mahon, 2015).

In one study, doulas also obtained additional lactation training through the community organization with which they were volunteering or working (Edwards et al., 2013).

In all but one of the studies examined, women supported by doula care showed significantly higher rates of breastfeeding initiation, regardless of whether the doula had additional lactation training or not. In the study in which doula support was not shown to have any effect on breastfeeding rates, the doulas were novice doulas and had low self-efficacy regarding their role as breastfeeding advocates and breastfeeding supports (Zandt et al., 2016).

Time Points of Breastfeeding Support

Across the articles reviewed, doulas were shown to provide care supportive of breastfeeding during the prenatal, intrapartum, and postpartum periods.

During prenatal visit(s), the birth doula's role in supporting breastfeeding largely surrounds educational and informational support (Gruber et al., 2013; Shlafer et al., 2018). One study described how doulas work during the prenatal period to address clients' beliefs and attitudes about breastfeeding by “listen[ing] to mother's ideas and concerns about breastfeeding, and work[ing] to dispel any myths that the mothers had” (Edwards et al., 2013).

The doula's role in intrapartum and immediate postpartum support surrounds promotion of breastfeeding friendly practices such as skin-to-skin care (Shlafer et al., 2018). Doula support of breastfeeding in the extended postpartum period includes breastfeeding counseling (Hans et al., 2018). In-home breastfeeding counseling from postpartum doulas occurs at a time when mothers are navigating the transition from hospital to home, and this is the time in which mothers are more likely to experience problems latching, discomfort, and concerns about milk supply, and often switch to infant formula or partial infant formula feeding (Deubel et al., 2019).

Doula Care Beyond the Hospital Setting

Specifically, several articles mention the doula's presence in the home and community setting, and this was shown to have positive impacts on breastfeeding initiation (Deubel et al., 2019; Edwards et al., 2013). One study examined the effectiveness of doula support of incarcerated women and found that in the prison setting as well, doulas have positive impacts on intent to breastfeed and subsequent rates of breastfeeding initiation (Shlafer et al., 2018).

Cultural Cohesion

Many articles made reference to effective doula care being culturally sensitive and appropriate, noting that clients may form closer bonds with doulas with whom they can identify (Cattelona et al., 2015; Deubel et al., 2019; Edwards et al., 2013; Gruber et al., 2013;). Doulas with whom clients may identify may be from the same community as their clients, or they may come from similar backgrounds (e.g., same ethnicity or shared background of being or having been teen parents) (Edwards et al., 2013). One study called for further research on potential therapeutic effects of matching clients with doulas of the same race (Shlafer et al., 2018).

Doulas and the Maternity Care System

Overall, doulas were shown to work largely independently of, yet collaboratively with, the system of maternal health care. Unlike lactation consultants, doulas have not become fully integrated into the maternity care team (Torres, 2013; Torres, 2015b). This is likely because the care they provide, though evidence based, is nonmedical in nature (Torres, 2013).

Effective doula care is that which is provided in close collaboration with other providers of breastfeeding and postpartum care services. Lactation consultants, nurses, midwives, and physicians provide the basis of breastfeeding education and care, and doulas function to bolster this with physical, emotional, and personal yet objective breastfeeding support (Gruber et al., 2013; Kozhimannil et al., 2013; Torres, 2015a, 2015b).

Across the studies, doulas displayed use of heterogeneous models of collaboration with perinatal professionals. Some doulas simply interact with health-care professionals while in the hospital for a client's labor while other doulas relied on close relationships with health-care professionals. In one study, nurses and doulas displayed intimate collaboration: an interprofessional, doula-trained, registered nurse served as an emergency source of clinical consultation, information, and support for intervention program doulas (Edwards et al., 2013).

Doulas are also able to be present and provide support and encouragement for breastfeeding for hours at a time, while medical professionals are often more constrained by time limits (Shlafer et al., 2018). In some cases, when nurses are quick to offer infant formula in the early stages of breastfeeding initiation, doulas serve as advocates and supports to protect the breastfeeding goals of mothers (Shlafer et al., 2018).

DISCUSSION

This integrative review adds to the small body of older research pertaining to the role which doulas in the United States play in supporting breastfeeding initiation and duration, and it reinforces earlier findings that doulas do positively impact the practice of breastfeeding. Doulas significantly improve rates of breastfeeding initiation, despite not necessarily receiving extensive lactation education or training. Doulas do not give clinical breastfeeding care to their clients (e.g., doulas do not carry out pre- and postweights, prescribe galactagogues, diagnose, and treat mastitis). Rather, they play a role similar to peer counselors by engaging in conversations around breastfeeding, encouraging breastfeeding-friendly practices, and supporting their clients with basic breastfeeding knowledge. Doulas provide such breastfeeding support in hospital, home, and community settings throughout the prenatal, intrapartum, and postpartum periods. This level of support truly makes a difference to families.

Since doula care has shown strong ability to improve breastfeeding initiation rates, including among African American mothers, doulas are an important resource to utilize in breastfeeding care networks (Deubel et al., 2019; Kozhimannil et al., 2013; Thurston et al., 2019). Currently, African American mothers breastfeed at lower rates than the national average, indicating that this group of mothers may benefit from additional, culturally appropriate breastfeeding interventions (CDC, 2019). African Americans face particular health-care challenges and barriers to breastfeeding such as cultural beliefs that breastfeeding may spoil a child or decrease the child's independence, the need to return to work soon after the child is born, and lack of peer support or breastfeeding role models to provide support and advice (Deubel et al., 2019). Given the positive impacts of breastfeeding on both maternal and infant health, improving rates of breastfeeding initiation and duration are critically important health-care goals.

Most of the significantly positive impacts of doula care described in the studies impacted breastfeeding initiation. While breastfeeding initiation is important and the provision of colostrum alone has beneficial effects on an infant's naive immune system, breastfeeding duration and breastfeeding exclusivity are important as well (Ballard & Morrow, 2013). One study found that African American mothers who were supported by doulas were more likely to breastfeed and do so exclusively, while another study which used birth doulas who provided up to two postpartum visits showed little impact on breastfeeding duration and exclusivity rates at 3 months postpartum (Edwards et al., 2013; Hans et al., 2018). Therefore, this integrative review cannot draw strong conclusions regarding the role of doulas on breastfeeding duration.

Most of the doulas represented in the literature are birth doulas (although some function as both birth and postpartum doulas). Since birth doulas mainly provide support up to, throughout, and for a limited time after birth, we find it unsurprising that their impacts are concentrated on improving breastfeeding initiation. More research specific to postpartum doulas and their impact on breastfeeding duration would add to the literature surrounding potentially effective methods of improving breastfeeding exclusivity and duration rates.

Though increasing doulas' presence in communities has the potential to greatly improve maternal–infant health outcomes, many families which may benefit most from doula care may not be able to afford it (Kozhimannil & Hardeman, 2016). Many doulas offer their services for free or on a sliding scale, but this reduces the doula's ability to earn a livable income (Torres, 2015a, 2015b). Some research suggests that by including provisions for Medicaid funding to cover the cost of doula services for eligible families, states can help abate this problem while simultaneously decreasing the amount spent on care related to adverse birth outcomes (Kozhimannil & Hardeman, 2016).

It is important to note that the Cochrane synthesis noted that birth support appeared most effective when provided by someone who was neither a part of the woman's social network nor a member of the hospital staff (Bohren et al., 2017). This may have implications for the doula profession as doulas continue to be recognized by and interact with hospital-based maternity care systems. As the doula profession continues to grow and awareness and utilization of doulas spreads, updated research as to the relationship that doulas have with traditional members of the maternity care team would greatly add to the literature.

Limitations

There were several limitations to this review. First, one author is a trained doula and therefore may have unconsciously introduced personal bias into the review process when developing and discussing the themes identified during the review. The second author helped to limit the addition of personal bias from the first author.

This review included data from studies with varying study designs, the review was limited in its rigor. Although both quantitative and qualitative studies contribute valuable findings to bodies of knowledge, it can be difficult to integrate data and findings from such an array of study designs, and error can be introduced during integrative reviews (Whittemore & Knafl, 2005).

Additionally, 50% (7/14) of the studies were conducted in the midwestern United States, and none were confirmed to have been conducted in the western United States (one was conducted in an unspecified urban location within the United States). Though it would be ideal for our conclusions to be representative of doulas' impacts on breastfeeding across the country, due to the limited locations of the studies included, this may not be the case.

Suggested Implications for Childbirth Educators

Childbirth educators (CBEs) are able to share with expectant parents the benefits that doula support can provide to their childbearing and lactation experiences. It was mentioned earlier that there is an unknown number of uncertified doulas who also acquire unknown amounts of breastfeeding information. Many new doulas audit childbirth education classes, thus CBEs are likely to have the opportunity to interact with a variety of new doulas who may or may not be pursuing certification. This allows CBEs the opportunity to remind new doulas of the importance of gaining an evidence-based breastfeeding knowledge base with which to support their clients. In this way, childbirth educators have the opportunity to both increase expectant parents' awareness of doulas as a breastfeeding resource and increase certified and uncertified doulas' decisions to seek out well-rounded, accurate breastfeeding knowledge.

CBEs must be advocates for increasing access to affordable doula training programs to ensure that doulas are from a diverse array of backgrounds to better serve diverse families. Maintaining professional relationships with doulas as they work to serve communities may be equally important to increasing pregnant women's access to well-prepared, culturally sensitive doulas. Finally, CBEs can also support their local and state efforts to provide Medicaid coverage of doula services. This may expand doula's abilities to reach women of all socioeconomic backgrounds.

CONCLUSION

The evidence strongly suggests that doulas have varying levels of breastfeeding preparedness, however, doulas provide important information, and emotional and physical support which improves breastfeeding practices in the United States. Doulas were shown to support breastfeeding in various ways and settings throughout the childbearing year, and their support has a promising ability to abate the intensity of disparities in breastfeeding rates among African American families.

While doula support does not replace any portion of breastfeeding care by nurses, midwives, obstetricians, or lactation consultants, several studies acknowledge the unique benefits that women gain from doula support during breastfeeding initiation and duration, thus doula support is complementary to the work of other professionals who provide breastfeeding care. An interprofessional approach will ensure that every mother and family has the knowledge and resources necessary to set and achieve breastfeeding goals. Lactation consultants, midwives, nurses, physicians, and doulas work together to support optimal infant feeding and human lactation. The combined breastfeeding and lactation support provided by this interprofessional network offers a multilayered effort to surmounting obstacles to achieving breastfeeding goals.

Biographies

STEPHANIE N. ACQUAYE is a BSN candidate at the University of Pennsylvania School of Nursing. She is a graduate of nursing 361-a whole semester course on breastfeeding and human lactation. Stephanie is a Hillman Scholar and will be continuing into the PhD program upon completion of her BSN.

DIANE L. SPATZ is a professor of perinatal nursing and the Helen M. Shearer professor of nutrition at the University of Pennsylvania School of Nursing with a joint appointment as a nurse scientist at Children's Hospital of Philadelphia. She is on the Executive Committee of the International Society for Research in Human Milk & Lactation.

DISCLOSURE

The authors have no relevant financial interest or affiliations with any commercial interests related to the subjects discussed within this article.

FUNDING

The author(s) received no specific grant or financial support for the research, authorship, and/or publication of this article.