Systematic Review of Evidence for Baby-Friendly Hospital Initiative Step 3

Abstract

Background:

The Ten Steps to Successful Breastfeeding outline maternity practices that protect, promote, and support breastfeeding and serve as the foundation for the Baby-Friendly Hospital Initiative.

Research aim:

This systematic review describes interventions related to Step 3 of the Ten Steps, which involves informing pregnant women about the benefits and management of breastfeeding. Our main objective was to determine whether prenatal clinic- or hospital-based breastfeeding education increases breastfeeding initiation, duration, or exclusivity.

Methods:

The electronic databases MEDLINE and CINAHL were searched for peer-reviewed manuscripts published in English between January 1, 2000, and May 5, 2016. Bibliographies of relevant systematic reviews were also screened to identify potential studies.

Results:

Thirty-eight studies were included. The research studies were either randomized controlled trials or quasi-experimental studies conducted in developed or developing countries. Findings suggest that prenatal interventions, delivered alone or in combination with intrapartum and/or postpartum components, are effective at increasing breastfeeding initiation, duration, or exclusivity where they combine both education and interpersonal support and where women’s partners or family are involved. However, varying study quality and lack of standardized assessment of participants’ breastfeeding intentions limited the ability to recommend any single intervention as most effective.

Conclusion:

Future studies should test the strength of maternal breastfeeding intentions, assess the role of family members in influencing breastfeeding outcomes, compare the effectiveness of different health care providers, and include more explicit detail about the time and full cost of different interventions.

Keywords

Baby-Friendly Hospital Initiative breastfeeding breastfeeding support education exclusive breastfeeding intervention lactation education prenatal interventions Ten Steps to Successful Breastfeeding

Background

The Ten Steps to Successful Breastfeeding were originally endorsed as one component of the Innocenti Declaration on the Protection, Promotion, and Support of Breastfeeding, enacted by the World Health Organization (WHO) and United Nations International Children’s Emergency Fund (UNICEF) in 1990 (UNICEF, 2006). The Ten Steps outline maternity practices that protect, promote, and support breastfeeding and serve as the foundation for the Baby-Friendly Hospital Initiative (BFHI). The scientific basis for the Ten Steps was published shortly after the Innocenti Declaration in 1998 (WHO, 1998). Recent reviews of the BFHI have found that implementation of the Ten Steps has a positive, dose-response relationship on breastfeeding outcomes (Howe-Heyman & Lutenbacher, 2016; Munn, Newman, Mueller, Phillips, & Taylor, 2016; Pérez-Escamilla, Martinez, & Segura-Pérez, 2016). However, Munn et al. (2016) noted that the mechanisms contributing to maternal breastfeeding decision-making remain unclear. In addition, after conducting a systematic review of 19 studies, Wong, Tarrant, and Lok (2015) suggested that there is no consensus regarding which prenatal breastfeeding education format is most effective (i.e., individual, group, or combination) to increase breastfeeding duration or exclusivity. Therefore, this article reviews prenatal breastfeeding interventions that target Step 3 of the BFHI.

Step 3 requires BFHI facilities to inform all pregnant women about the benefits of breastfeeding, the importance of exclusive breastfeeding, and the management of breastfeeding, including feeding on demand, ensuring adequate milk, and reviewing proper positioning and attachment. Prenatal education techniques described as effective in a WHO review included the following: building a mother’s confidence to breastfeed using group education sessions; discussing breastfeeding myths and inhibitions; offering practical demonstrations; and involving the mother’s social supporters, such as partners or grandmothers (WHO, 1998).

A review of primary care interventions to extend breastfeeding duration delivered during prenatal and/or infant care (de Oliveira, Camacho, & Tedstone, 2001) found that the only effective strategy during prenatal care was group education; however, the most effective interventions spanned the prenatal and postnatal periods with intensive group sessions, home visits, and individual sessions. These interventions generally provided information on the benefits of breastfeeding; guidance on positioning, attachment, and overcoming problems; and support, encouragement, and reassurance (de Oliveira et al., 2001). A more recent review conducted in 2011 evaluated the effect of breastfeeding support and education on breastfeeding at 4 to 6 weeks and at 6 months postpartum (Imdad, Yakoob, & Bhutta, 2011). Interventions delivered in the prenatal, postnatal, or combined prenatal and postnatal periods were included for review. The authors found that prenatal counseling improved any breastfeeding at 4 to 6 weeks, whereas combined prenatal and postnatal promotion improved any and exclusive breastfeeding at 6 months (Imdad et al., 2011). They noted that lay support and education in addition to professional support were effective at increasing exclusive breastfeeding at 4 to 6 weeks and 6 months, and group counseling was effective only for improving exclusive breastfeeding at 4 to 6 weeks (Imdad et al., 2011). This review included only studies with true randomization of participants and excluded web- or Internet-based interventions, interventions directed primarily toward fathers or family members, and education delivered within the context of a broader package of interventions.

A Cochrane systematic review in 2012 assessed randomized controlled trials (RCTs) of the effect of prenatal education on breastfeeding duration, excluding all interventions that included any intrapartum or postpartum components (Lumbiganon et al., 2012). The authors concluded that educational interventions using multiple methods, such as counseling from a lactation consultant and an educational breastfeeding booklet, were not significantly more effective than a method that included only an educational intervention for increasing breastfeeding initiation or duration (Lumbiganon et al., 2012). However, multiple methods of breastfeeding education were significantly more effective than routine care for increasing exclusive breastfeeding at 3 months. The small effect sizes and methodological limitations of included studies led the authors to avoid making any specific prenatal breastfeeding education recommendations based on these findings. The generalizability of this review is also limited because all included studies, except for one, were drawn from developed countries.

Key Messages

There is no consensus regarding the most effective format to deliver prenatal breastfeeding education, Step 3 of the Ten Steps.

Prenatal interventions are effective where they combine education with interpersonal support and where they involve women’s partners or family.

Heterogeneous study quality and definitions of participant breastfeeding intention limit our ability to recommend any single intervention as most effective.

Future studies should measure strength of breastfeeding intentions, effectiveness of different providers, and time/cost associated with various delivery formats.

Our review aims to update the evidence for Step 3 and to address some of the limitations of previous reviews by including a broader selection of interventions: those delivered during the prenatal period alone or in combination with intrapartum and postpartum components; studies drawn from both developed and developing country contexts; and quasi-experimental studies in addition to randomized trials. The main objective of this review was to determine whether prenatal breastfeeding interventions initiated in any clinic or hospital setting and delivered to the mother and/or her family members or other supportive individuals increase breastfeeding initiation, duration, or exclusivity.

Methods

Design

A systematic review of the literature was conducted to identify all interventions that could provide evidence for the effect of Step 3 on breastfeeding outcomes. In accordance with PRISMA guidelines, the electronic health science databases MEDLINE and CINAHL were searched by two of the authors (K.W. and K.P.T.) for peer-reviewed journal articles published in English between January 1, 2000, and May 5, 2016 (Liberati et al., 2009). These two databases were chosen due to their focus on medical research. Bibliographies of relevant systematic reviews were also screened to identify potential studies. Search methodology combined the terms breast feeding* or breastfeed*; counseling or health education; and prenatal or antenatal in all fields.

Sample

Our inclusion and exclusion criteria are based on the PICOS format: participants, interventions, comparisons, outcomes, and study design (WHO, 2014). For inclusion, we required that articles (1) be published in a peer-reviewed journal, (2) describe a prenatal breastfeeding education intervention initiated in a clinic or hospital setting, (3) include a control or comparison group, and (4) report breastfeeding initiation, duration, or exclusivity as an outcome. We defined breastfeeding initiation as any or exclusive breastfeeding at hospital discharge or within 2 weeks of delivery; breastfeeding duration as any breastfeeding measured after at least 2 weeks postpartum; and exclusive breastfeeding according to study definitions. We excluded pilot, cross-sectional, and retrospective studies as well as studies without comparison groups.

Our search strategy returned 1,369 manuscripts, and 5 additional studies (Gijsbers, Mesters, Knottnerus, Kester, & Schayck, 2006; Graffy, Taylor, Williams, & Eldridge, 2004; MacArthur et al., 2009; Petrova, Ayers, Stechna, Gerling, & Mehta, 2009) were identified in screening of citations of relevant systematic reviews (de Oliveira et al., 2001; Imdad et al., 2011; Lumbiganon et al., 2012). After removing duplicates, 1,329 studies remained and 1,256 studies were excluded based on abstract review. The 73 remaining full-text articles were reviewed for consideration, and 38 studies met inclusion criteria for this review (see Figure 1).

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram.

Risk of Bias Assessment

To assess methodological quality of included studies, we used the Cochrane Collaboration’s tool for examining potential risk of bias: selection bias was assessed through randomization procedures and allocation concealment; performance and detection bias through blinding procedures; detection bias through double-blinding procedures; and attrition bias through completeness of outcome data.

Results

Study Characteristics

Study characteristics are presented in Table 1. Of the 39 studies identified for review, 33 were RCTs and 5 had a quasi-experimental design. Sample sizes ranged from 54 to 2,511 for a total of 15,250 participants or couples across the included studies. Fifteen studies were conducted in the United States, 9 in Asia, 6 in Europe, 2 in Africa, 1 in the Middle East, 4 in Australia, and 1 in Canada. Definitions of usual care ranged from no mention of breastfeeding support to BFHI standard of care. Only 1 study (Wambach et al., 2011) assessed the strength of maternal breastfeeding intentions prior to administering the intervention, although 30 studies measured at least one variable related to breastfeeding plans: intent to initiate breastfeeding (Aidam, Pérez-Escamilla, & Lartey, 2005; Anderson, Damio, Young, Chapman, & Pérez-Escamilla, 2005; Bonuck et al., 2014; Bonuck, Trombley, Freeman, & McKee, 2005; Chapman, Damio, Young, & Pérez-Escamilla, 2004; Forster et al., 2004; Huang et al., 2007; Kamran, Shrifirad, Mirkarimi, & Farahani, 2012; Kellams et al., 2016; Kupratakul, Taneepanichskul, Voramongkol, & Phupong, 2010; Lavender et al., 2005; C. H. Lin, Kuo, Lin, & Chang, 2008; S. S. Lin, Chien, Tai, & Lee, 2007; Mattar et al., 2007; Meedya, Fahy, Yoxall, & Parratt, 2014; Muirhead, Butcher, Rankin, & Munley, 2006; Nichols, Schutte, Brown, Dennis, & Price, 2009; Noel-Weiss, Rupp, Cragg, Bassett, & Woodend, 2006; Otsuka et al., 2014; Petrova et al., 2009; Ryser, 2004; Srinivas, Benson, Worley, & Schulte, 2014; Su et al., 2007; Wambach et al., 2011; Wolfberg et al., 2004; Wong, Tak Fong, Yin Lee, Chu, & Tarrant, 2014); intended breastfeeding duration (Aidam et al., 2005; Anderson et al., 2005; Chapman et al., 2004; Forster et al., 2004; Graffy et al., 2004; Lavender et al., 2005; Meedya et al., 2014; Noel-Weiss et al., 2006; Srinivas et al., 2014); intended breastfeeding exclusivity (Aidam et al., 2005); breastfeeding knowledge (Bonuck et al., 2014; Kronborg, Maimburg, & Vaeth, 2012; S. S. Lin et al., 2007; Ochola, Labadarios, & Nduati, 2013; Petrova et al., 2009); breastfeeding self-efficacy (Chapman et al., 2013; Kronborg et al., 2012; Otsuka et al., 2014; Petrova et al., 2009); breastfeeding attitudes (Huang et al., 2007; C. H. Lin et al., 2008; S. S. Lin et al., 2007; Petrova et al., 2009); previous breastfeeding experience (Bonuck et al., 2014; Mattar et al., 2007; Srinivas et al., 2014); confidence in breastfeeding (Kronborg et al., 2012; Meedya et al., 2014; Petrova et al., 2009); decided to breastfeed prior to pregnancy (Noel-Weiss et al., 2006); or perceived their partner to be supportive of breastfeeding (Meedya et al., 2014). Eight publications did not describe any participant breastfeeding intentions (Bich, Hoa, & Målqvist, 2014; Gijsbers et al., 2006; Gill, Reifsnider, & Lucke, 2007; Gross, Mendelsohn, Gross, Scheinmann, & Messito, 2016; Ickovics et al., 2007; MacArthur et al., 2009; Maycock et al., 2013; Sandy, Anisfeld, & Ramirez, 2009). None of the 38 articles specified the full costs of the interventions, although 2 provided partial-cost information (Ickovics et al., 2007; Muirhead et al., 2006) and Bonuck et al. (2005) estimated the intervention cost per participant.

Table 1.

Study Characteristics

First author, year	Study design	No. randomized (N)	Sample characteristics	Intervention type	Person responsible for intervention	Control	Breastfeeding outcome measures	Results
Aidam, 2005	RCT	132	Third trimester women planning to deliver in two hospitals in Tema Township, Ghana	Group 1 (IG1): EBF support provided pre-, peri-, and postnatally. Group 2 (IG2): EBF support provided only peri- and postnatally. Two 20-min educational sessions prenatally; nine home follow-up visits across 6 months postpartum. Education in groups of two to four lasting 20 min per session. BF materials, adapted from WHO/UNICEF BF counseling training manual and a book, Helping Mothers to Breastfeed, provided to both groups.	Two local nurses and one nutritionist with BF experience	Non-BF health education support provided to ensure equal amount of contact. Two education sessions provided prenatally, and nine home follow-up visits across 6 months postpartum. Mothers directed to postpartum child welfare clinics BF-related questions. One study site was BFHI.	Main outcome was EBF (yes/no) based on whether child received no food or liquid in 3 time periods: (1) over previous month, (2) over preceding day, and (3) over 6-month period.	Initiation: not assessed. Duration: not assessed. Exclusivity: at 3 months, 92.1% in IG1 and 88.1% in IG2 EBF the previous day vs. 65.9% in control (p = .004). At 6 months, 90.0% in IG1 and 74.4% in IG2 EBF the previous month vs. 47.7% in control (p = .008). At 6 months, EBF since birth was significantly higher (p = .02) among IG1 and IG2 (39.5%) vs. control (19.6%).
Anderson, 2005	RCT	182	Low-income patients ≥ 18 years old ≤ 32 weeks GA from prenatal clinic of a BFHI hospital in Hartford, CT, USA	Antepartum, in-hospital, and postpartum clinic-based BF support and education plus three prenatal and nine postnatal home visits and daily visits during hospitalization from PC. Entire family encouraged to participate in education, especially principal support person; included option to watch BF video. Average (SD) duration of prenatal home visits was 2.6 (1.9) hours and in-hospital visits was 2.2 (2.0) hours.	PC group plus maternity ward nurses for clinical support	Routine BF education prenatally from clinic staff; BF support and education from maternity ward nurses of BFHI hospital at delivery and access to IBCLC	Interviewed in hospital at delivery, weekly during first month, biweekly during second and third months. EBF defined using “24-hour,” “previous week,” and “ever given” recalls.	Initiation: at discharge, 24% of controls vs. 9% of intervention mothers had not initiated (RR = 2.48; 95% CI [1.04, 5.90]), with 56% and 41%, respectively, non-EBF. Duration: at 3 months, 64% of controls vs. 50.8% of intervention mothers stopped ABF (RR = 1.26; 95% CI [0.93, 1.70]). Exclusivity: non-EBF significantly higher for controls at months 1, 2, and 3 using the “previous 24 hours” or the “previous week” recall definitions. At 3 months, 98.6% of controls vs. 79.4% of intervention mothers non-EBF (RR = 1.24; 95% CI [1.09, 1.41]).
Bich, 2014	RCT (randomized sites in nonadjacent districts)	492	Fathers with pregnant wives between 7 and 30 weeks gestation by August 1, 2010, in Vietnam	Both 30- to 45-min group counseling session at health center during antenatal visit and individual counseling in four home visits (third trimester, first week, 42 days, and 3.5 months postpartum) to discuss BF benefits and activities to support wives in pregnancy/postpartum. Mass media campaign on BF benefits and how fathers can provide support (posters, local speaker systems, a public event, pamphlets, and giveaways).	Group counseling from community health workers and individual counseling from village health workers	All pregnant women in control districts visited health centers ≥ 3 times during pregnancy; received general BF and nutrition consultation	Kindergarten teachers interviewed mothers after birth; surveyed BF at 4 and 6 months. EBF measured over past 24 hours, past week, and since birth as infant not given any water, other liquids, solid, semisolid, or soft foods since birth	Initiation: not assessed. Duration: at 4 months, 99.2% intervention mothers vs. 98.7% controls ABF; at 6 months, 99.2% intervention mothers vs. 97.8% controls ABF (not significant). Exclusivity: at 4 months: based on past week and since birth recall, intervention mothers about twice as likely to EBF as controls: OR = 1.85, 95% CI [1.10, 3.11] and OR = 2.36, 95% CI [1.35, 4.14], respectively. At 6 months: based on past week and since birth recall, intervention mothers about 6 times as likely as controls to EBF: OR = 5.8, 95% CI [1.60, 21.7] and OR = 6.3, 95% CI [1.35, 29.29], respectively.
Bonuck, 2005	RCT	382	Low-income patients < 24 weeks GA drawn from 2 hospital-affiliated health centers in the Bronx, NY, USA	Two individual educational prenatal meetings: discussed BF benefits and initiation in hospital. Educational materials: flip-chart and color pamphlets for BF physiology; culturally appropriate lactation doll and nipple. Postpartum hospital/home visits gave hands-on instruction to avoid BF problems and use a pump. IBCLCs helped mother seek support from families, schools, workplaces, and health care providers. IBCLCs available by phone 12 months postpartum. IBCLCs spent a mean of 111 min in prenatal contact.	IBCLC	Routine hospital/WIC care; sites differed among control mothers	Initiation, short-term, and long-term duration through maternal self-report at weekly interview using 7-level scale to measure percentage of human milk infant received over past week or since hospital discharge, compared with total amount of feedings.	Initiation and duration: intervention mothers significantly more likely to BF every week postpartum up to and including week 20 than controls (53.0% vs. 39.3%; p < .028), with the exception of week 18. Significantly more intervention mothers gave ≥ 50% human milk feedings in the first week after birth (69% vs. 47%; p < .001). Exclusivity: EBF rates not significantly different at any of the 52 weeks observed, but significantly more intervention mothers gave ≥ 50% human milk feedings through week 9.
Bonuck, 2014	RCT	666 (BINGO trial) and 275 (PAIRINGS trial)	English- or Spanish-speaking women ≥ 18 years old; first or second trimester of singleton pregnancy, drawn from a medical center-affiliated prenatal care clinic in the Bronx, NY, USA	BINGO trial: 3 intervention arms: (1) EP anticipatory guidance from prenatal providers at five prenatal visits, (2) IBCLC support along with EP, and (3) IBCLC support alone. PAIRINGS trial (EP + IBCLC): EP guidance of two to three brief open-ended questions for providers to ask that normalized BF (“What are your plans for BF?”), sought to clarify knowledge about how long or how much to BF, and discussed social support. IBCLC support: two prenatal sessions, a hospital visit, and regular phone calls postpartum through 3 months or until BF ceased. Optional postpartum home visits. Prenatal contacts averaged 1 hour.	BINGO: prenatal providers (resident, attending obstetrician, and certified nurse-midwives) and IBCLCs. PAIRINGS: prenatal providers (obstetrician or gynecologist faculty) and IBCLCs.	Routine care: no explicit BF promotion or support; one IBCLC available at the hospital	BF intensity and exclusivity at 1, 3, and 6 months postpartum assessed through follow-up phone interviews using items adapted from the Infant Feeding Practices Survey	Initiation: BINGO and PAIRINGS: IBCLC + EP group had greater odds of initiating (OR = 3.29, 95% CI [1.03, 10.48]). EP group didn’t differ from usual care. Duration: BINGO: ABF differed by treatment group at 3 months and was highest for the IBCLC + EP and IBCLC groups. BF intensity didn’t differ significantly among the four groups at 1, 3, or 6 months. EP group did not differ from usual care. PAIRINGS: intervention group had significantly higher rates of ABF at 1, 3, and 6 months. BF intensity differed by treatment group at 1 and 3 months. Exclusivity: BINGO: IBCLC + EP group had greater odds of 3 months EBF (OR = 4.24, 95% CI [1.01, 37.94]). EP group did not differ from usual care. PAIRINGS: intervention group had greater odds of EBF at 1 month (OR = 4.29, 95% CI [1.94, 9.47]) and at 3 months (OR = 2.86, 95% CI [1.21, 6.76]).
Chapman, 2004	RCT	219	Low-income patients ≤ 26 weeks GA served by BFHI Hartford Hospital in Connecticut, USA	Clinic-based BF support and education plus ≥ 1 prenatal and ≥ 3 postpartum home visits and daily visits from a PC during postpartum hospitalization. Prenatal visit lasted a mean (SD) of 69.0 (57.6) min.	PC and maternity ward nurses for clinical support	Routine education from maternity ward nurses and access to an IBCLC for problems; BFHI standard of care	Participants interviewed monthly via phone until they stopped BF and for maximum 6 months postpartum to obtain data on ABF and EBF	Initiation: 61% lower risk of not initiating in intervention group (9% vs. 23%, RR = 0.39, 95% CI [0.18, 0.86]) vs. controls. Duration: intervention mothers had reduced risk of not BF at 1 and 3 months: 1 month, 36% vs. 49%, RR = 0.72, 95% CI [0.50, 1.05]; at 3 months, 55.6% vs. 71%, RR = 0.78, 95% CI [0.61, 1.00]. At 6 months, there was no significant difference between groups (RR = 0.94, 95% CI [0.79, 1.11]). Exclusivity: no significant difference in EBF at 1 month (RR = 1.07, 95% CI [0.90, 1.27]).
Chapman, 2013	RCT	206	Low-income patients ≥ 18 years and ≤ 36 weeks GA; prepregnancy BMI > 27.0, singleton pregnancy served by BFHI Hartford Hospital in Connecticut, USA	Routine care from hospital prenatal clinic, postpartum staff nurses, and IBCLCs. Access to 3 prenatal visits, daily in-hospital visits after birth, and ≤ 11 postpartum home visits from a PC across 6 months postpartum. Prenatal visits assessed BF experiences/knowledge and presented personalized education on BF logistics, risks of formula feeding, and anticipatory guidance.	Two bilingual PCs from Hartford who BF ≥ 1 child for > 6 months, with 30 hours of training. New PCs shadowed experienced PCs 3−6 months, moving to direct care with IBCLC supervision; 20 hours of specialized training on obesity and EBF.	Prenatal BF education: brief BF discussions during routine clinic appointments and written educational materials. Staff nurses gave routine perinatal BF assistance at BFHI hospital with IBCLCs available. After discharge, participants could call hospital’s “warm line” with BF questions. Optional BF support available from PCs, who provided ≤ 3 prenatal visits, daily in-hospital visits, ≤ 7 personalized home visits during postpartum year, and phone support.	Initiation, ABF, and EBF reported as dichotomous variables at birth, 2 weeks, and monthly through 6 months postpartum	Initiation: at 2 weeks postpartum, intervention group had marginally higher rates of ABF (93% vs. 84%, p = .09) and provided ≥ 50% of feedings as human milk (81% vs. 67%, p = .08) vs. controls. Duration: no significant effect on ABF beyond 2 weeks. Exclusivity: no significant effect on EBF at any time, regardless of EBF definition evaluated.
Forster, 2004	RCT	981: 327 in each of two intervention arms and one control arm	Primiparas 16−24 weeks GA who attended a public, tertiary BFHI-certified women’s hospital in Melbourne, Australia. Women choosing a birth center or private obstetric practice excluded due to high BF in those populations.	Practical Skills Group: one 1.5-hour session on practical BF skills using teaching aids. Latch-on technique demonstrated with dolls and knitted breasts; discussed BF complications and management. Access to standard care available. Attitudes Group: two 1-hour sessions focused on changing attitudes to BF. Women encouraged to bring partners or significant others. Access to standard care available.	Midwives and a community educator with specific training in childbirth education conducted classes. IBCLC qualifications not required, since the interventions were intended to be implemented in normal pregnancy care.	Controls could access standard care available at BFHI hospital, which included formal BF education sessions; BF information as a component of childbirth education courses; IBCLC support as necessary (inpatient and outpatient); community peer support; optional attendance at a BF information evening; videos or education on BF presented in the postnatal ward; 24-hour phone counseling support; and a postnatal home visit by a midwife.	Primary outcomes were initiation and duration; duration included ABF and EBF at 6 months	Initiation: no significant differences among groups as measured by interview at 2−4 days. Duration: no significant differences among groups at 6-month interview: proportions of women ABF were 55% (practical skills), 50% (attitudes), and 54% (standard care). Exclusivity: no significant differences among groups at 6-month interview: proportions of women EBF were 36% (practical skills), 34% (attitudes), and 35% (standard care).
Gijsbers, 2006	Cluster RCT of family practices	113 Families at 108 practices	Families eligible when mothers were < 7 months GA and ≥ 1 first-degree relative had asthma diagnosed by a doctor; family practices drawn from southeastern Netherlands	Renewed education program based on principles of the Attitude–Social Influences–Self-Efficacy model. Trained assistants visited participants twice during pregnancy and once postpartum to provide basic information about BF and expressing milk; prepare women to anticipate future BF problems for ≥ 6 months. During the first visit, women received a booklet that reinforced information given verbally.	Researcher	Families received usual care according to the guidelines of the Dutch College of General Practitioners in which BF is recommended for all babies in the Netherlands.	Main outcome was prevalence of EBF at 6 months. EBF defined according to WHO guidelines where no other milk or solids given except vitamins and minerals. Duration of EBF was counted in weeks for the analysis.	Initiation: not assessed. Duration: not assessed. Exclusivity: at 6 months, 48% intervention mothers EBF compared with 27% controls (p = .03, OR = 2.91, 95% CI [1.10, 7.71]) corrected for maternal age, education level, and BF experience. Significant difference in the duration of EBF between groups (for discontinuation of BF, HR = 0.58, 95% CI [0.33, 0.99], p = .05).
Gill, 2007	RCT with convenience sample	200	Low-income Mexican American patients in second trimester from waiting rooms at two health department maternity clinics in large city in southwestern United States	≥ 1 Individual prenatal education session at clinic; postpartum phone support; home visits at the mothers’ requests or if the research team deemed it necessary	IBCLC	Standard BF education available at WIC clinics; no IBCLC or lactation educators employed at either the maternity or WIC clinic	BF initiation and ABF duration	Initiation: intervention group was 2.31 times as likely to BF, 95% CI [1.10, 4.96], vs. controls. Duration: intervention group was 3.15 times as likely to BF for 6 months and half as likely to quit at any one time vs. controls. Exclusivity: not assessed.
Graffy, 2004	RCT	720	32 General practices in London and south Essex selected on proximity to PCs, having low-income population, providing prenatal and postnatal care, and not undertaking specific BF promotion. Women 28−36 weeks GA.	Women visited once before birth and offered postnatal support by phone or further home visits if requested. At the prenatal visit, the PCs gave the women a contact card and two leaflets published by the National Childbirth Trust and Health Education Authority.	PC	General practices selected on the criterion that they were not undertaking specific initiatives to promote BF	Primary outcome was prevalence of ABF at 6 weeks. Secondary outcomes were proportion of women giving any breast or bottle feeds at 4 months, duration of ABF, time to introduction of bottle feeds, and satisfaction with BF.	Initiation: 95% women in the intervention group BF vs. 96% controls (RR = 0.99, 95% CI [0.84, 1.16], p = .44). Duration: not significantly different between the women in intervention and control groups (median = 110 days vs. 96 days, log rank statistic = 0.58, p = .445).
Gross, 2016	RCT	533	Pregnant women ≥ 18 years; self-identified as Hispanic/Latina; fluent in English or Spanish; singleton uncomplicated pregnancy; able to provide phone number; intending to receive care at study sites in New York City, USA	The “Starting Early” intervention involved individual nutrition counseling in the prenatal and postpartum periods (once during third trimester and once on postnatal ward); 45- to 60-min NPSGs coordinated with well-child visits; plain language handouts; and nutrition education DVDs. Curriculum and materials developed in English and Spanish. NPSGs of four to eight mother–infant pairs formed into cohorts by infant’s birth date. NPSGs met at 1 and 2 months postpartum, relevant to these BF outcomes. Intervention continued through 3 years of age.	Bilingual RDs/IBCLCs	Control groups received same routine prenatal, postpartum, and pediatric primary care: included prenatal visits with attending or resident obstetrician or nurse midwife; individual consultation with a nutritionist (RD) and group childbirth and BF classes offered. Women with poor weight gain, obesity, or diabetes offered additional RD visits. On the postpartum unit, nurses trained in lactation support and a part-time certified LC available to assist mothers with BF difficulties. An RD/IBCLC available in the pediatric clinic for mother–infant pairs with BF or other feeding difficulties. About 90% study participants were WIC clients: offered BF and nutrition counseling at WIC sites.	Infant-feeding practices assessed using questions adapted from the Infant Feeding Practices Study II. BF assessed using survey questions and a 24-hour diet recall. BF levels classified according to WHO definitions. BF intensity is the percentage of human milk feedings per day, and combination feedings are giving both human milk and formula at the same feed.	Initiation: no difference in ever BF in the hospital: control 95.3% vs. intervention 95.9% (p = .82, OR = 1.16, 95% CI [0.47, 2.85]). Duration: no difference in 3-month ABF: control 80.4% vs. intervention 83.3% (p = .47, OR = 1.21, 95% CI [0.75, 1.95]). Exclusivity: at 3 months, more intervention mothers (33.0%) EBF vs. controls (23.4%) (p = .03, OR = 1.61, 95% CI [1.07, 2.44]). Intervention mothers reported higher BF intensity (67.7% vs. 59.7%, p = .03) and lower percentage of combination feedings per day vs. controls (2.0% vs. 4.3%, p = .03) and were less likely to introduce complementary food or liquids (6.3% vs. 16.7%, p = .001).
Huang, 2007	Quasi-experimental study	130	Primigravidas ≥ 18 years; 29−36 weeks GA, without major pregnancy complications; singleton pregnancy; planning to BF at time of enrollment from a hospital in Taipei, Taiwan	Web-based prenatal BF education program: topics included composition of human milk, BF benefits, mechanism of lactation, BF techniques, BF problems and solutions, contraindications to BF and human milk storage. Teaching media included text, graphics, animations, interactive flash animations, audio, and video. A real-time communication system and message board enabled learners to interact asynchronously, helping them exchange ideas and communicate with each other.	Web-based	Historic controls who did not receive such a program. Women recruited from the same prenatal clinic as intervention mothers, which had a certified Lactation Intervention Program offering BF promotion in the prenatal checkup and postnatal ward. Hospital BF education: pamphlet to women during prenatal checkup. Nurse available to answer individual questions.	Percentage of BF for the cases studied, including EBF and mixed BF. EBF represented infants who received only human milk, with no supplemental formula used. Mixed BF consisted of the subset of infants who were ABF.	Initiation: significantly different at 2 weeks: 45% vs. 20% at 2 weeks (p < .05). Duration: significant effect on mixed feeding at 6 weeks: 53.3% vs. 38.3% (p < .05); after controlling for the time trend and infant birth weight (OR = 2.69, p < .05). Exclusivity: significantly different at 6 weeks: 26.7% vs. 20% (p < .05). After adjusting for the time trend and infant birth weight (OR = 1.75, z = 1.97, p = .048).
Ickovics, 2007	Multisite RCT	1,047	Young women, 14−25 years, entering prenatal care at two publicly funded clinics affiliated with Yale in Connecticut, USA	Group prenatal care, which occurred over approximately 20 hours across pregnancy. Integrated care combined assessment, education and skills building, and support. Education as 10 sessions with trained providers and structured materials.	Credentialed provider	Individual prenatal care, which occurred over the course of 2 hours across pregnancy	BF initiation was based on self-report at the first postpartum interview conducted 6 months postpartum	Initiation: significantly improved for women in group care vs. those in individual care: 66.5% vs. 54.6% (p < .001). Duration: not assessed. Exclusivity: not assessed.
Kamran, 2012	RCT; randomized health centers	88	Eight health centers in Khorramabad city, Lorestan province, Iran. Four assigned as controls and four as experimental centers; whole center assigned as group to prevent contamination. Primiparas selected before delivery.	Group education: Health Belief Model: used to formulate the education and behavioral skill-building interventions. Instruction given in three sessions (30–40 min each). After the intervention, educational materials summarized in a pamphlet. Pamphlets distributed to experimental groups for self-study. Mothers encouraged to contact researchers at any time if they were worried about the baby’s condition or BF.	Researchers	Health centers not offering the intervention	Levels of BF classified according to WHO definitions	Initiation: not assessed. Duration: at 1 month, no significant difference in BF between control and experimental groups (p = .293). Exclusivity: at 4 months, EBF in experimental group (77.3%) significantly higher vs. controls (47.7%) (p = .007).
Kellams, 2016	RCT	522	WIC-eligible women 24–41 weeks GA from University of Virginia Health System and the Virginia Commonwealth University Health System in Virginia, USA	Intervention women shown 25-min educational BF video (Better BF, InJoy Productions, 2008) while waiting for prenatal clinic appointment: information about BF, including importance, latch, hunger cues, positioning, sore nipples, engorgement, how human milk is made, and lifestyle issues	Researcher	Control women shown a 20-min educational video about nutrition during pregnancy (Healthy Pregnancy Nutrition, InJoy Productions, 2007) with information about healthy diet and the importance of exercise during pregnancy	BF initiation and EBF during the newborn hospital stay	Initiation: no significant difference (unadjusted OR = 1.03, 95% CI [0.70, 1.50]). Among women who initiated, no differences in EBF during hospital stay or in hospital practices thought to be associated with BF success. Duration: not assessed. Exclusivity: not assessed.
Kronborg, 2012	RCT	1,193	Prenatal patients < 21 weeks + 6 days GA at Aarhus Midwifery Clinic, a large clinic connected to university hospital in urban area of Denmark. Hospital adopted most BFHI standards, but not certified at time of study.	“Ready for Child programme” comprised three modules, each lasting 3 hours with 2 hours dedicated to BF. Training sessions attended between weeks 30 and 35 GA and woman’s partner invited to participate. Maximum number of couples in each class was eight. The three modules included lectures and discussions about the delivery process, pain relief, and coping strategies; infant care and BF; parental role; and relationship between woman and partner.	Four midwives of varying seniority chosen as instructors; attended 3-day preparation course, 1 day for each module	Standard care offered by prenatal clinic, which did not include any prenatal training program; no effort made to prevent reference group from seeking additional support elsewhere. Different prenatal training programs provided by other stakeholders, mainly by relaxation therapists.	Duration of full and ABF collected 6 weeks postpartum and 1 year postpartum	Initiation: not assessed. Duration: no differences in BF duration, self-efficacy score, or BF problems, but in week 36 GA, intervention mothers reported higher confidence (p = .05), and 6 weeks after birth they reported having obtained sufficient knowledge about BF (p < .02). Among intervention women, those with sufficient knowledge BF significantly longer vs. women without sufficient knowledge (HR = 0.74, 95% CI [0.58, 0.97]). This association not found in controls (HR = 1.12, 95% CI [0.89, 1.41]). Exclusivity: not assessed.
Kupratakul, 2010	RCT	80	Pregnant women > 32 GA from clinics at the Department of Obstetrics and Gynecology, King Chulalongkorn Memorial Hospital, Faculty of Medicine, Chulalongkorn University and Theptarin Hospital, Bangkok, Thailand	In addition to standard BF education, mothers received a knowledge sharing practices program with 3-hour empowerment strategies (KSPES): explored women’s knowledge through storytelling, sharing best practices, and discussing each other’s knowledge, according to the steps of the Knowledge Sharing Practice applied with an empowerment perspective based on Gibson’s theory, which includes four steps: discovering reality, critical reflection, taking charge, and holding on, used to build potential for successful EBF. Women shared BF success and learned from others’ experiences. Postnatal support included phone interview between 7 days and 6 months postpartum to motivate mother and improve EBF. Intervention mothers received home visits if they had problems with EBF.	Researcher	Routine BF education ≥ 2 times. First at the prenatal check-up, where their breasts and nipples were checked, and any abnormality corrected such as by using Hoffman’s maneuver or using breast cups to cover the nipples. At second and subsequent check-ups, pregnant women received advice on BF benefits, significance of EBF, techniques for successful EBF, importance of immediate BF, disadvantages of formula milk and water, and ways to BF after return to work outside the house. Provided a document about BF.	Primary outcome: EBF at 6 months. Secondary outcomes: EBF at 7 and 14 days and 1, 2, 3, 4, and 5 months. EBF: only mothers’ milk without any other oral rehydration solution or water. Predominant BF: mothers’ milk and water, sweetened water, and juices without formula. Partial BF: mothers’ milk and complementary food. No BF: only formula/other liquids or food.	Initiation: intervention mothers had significantly higher rates of EBF at 14 days vs. controls (82.5% vs. 52.6%, p = .005). Duration: predominant BF significantly higher in study group vs. controls at 3, 5, and 6 months. However, rates of partial BF were no different between groups. Rates of “no BF” were significantly higher for controls at 2, 3, 4, 5, and 6 months. Exclusivity: EBF significantly higher in study group vs. controls at 1, 2, 4, 5, and 6 months: 77.5% vs. 52.6%, p = .021 at 1 month; 62.5% vs. 36.8%, p = .023 at 2 months; 35.0% vs. 7.9%, p = .008 at 4 months; 25.0% vs. 2.6%, p = .012 at 5 months; and 20.0% vs. 0%, p = .005 at 6 months.
Lavender, 2005	Cluster RCT; eight electoral wards in one county randomized; pairs matched according to Jarman Underprivileged Area score	1,312	Women in third trimester; wanted to BF, registered with a practice site/general practitioner in one of eight wards in Northwest UK	Normal prenatal care plus attendance in third trimester of one 2.5-hour prenatal BF education session. Each session involved ≤ 8 women, facilitated by a qualified infant-feeding coordinator. Reviewed BF importance and how BF works; watched a video; reviewed positioning and attachment; discussed avoiding problems, expressing milk, and use of aids.	Midwife	Standard prenatal care with BF advice from attending clinic midwives	ABF at hospital discharge and EBF at 4 months	Initiation: no significant difference in BF at discharge (OR = 1.2, 95% CI [0.8, 1.7], p = .3). Exclusivity: At 4 months, no significant difference in EBF (OR = 1.1, 95% CI [0.6, 1.8], p = .8).
S. S. Lin, 2007	Quasi-experimental with matching	92	Primigravidas 20–36 weeks GA from prenatal clinic of a medical center in Taipei City, Taiwan; patients > 20 years, currently married, no pregnancy complications, intending to BF	90-min educational program on BF plus usual prenatal care; 50 min for lecture and skill training, 20 min for group discussion and skill practicum, and 20 min for a tour of the postpartum ward. BF practices at the postpartum ward (e.g., rooming-in, early feeding, etc.) were introduced and several women who successfully BF shared their experience. Three aspects of BF education considered: knowledge, affective component, and skills.	Experienced obstetric nurse, with 12 hours of training on BF for health professionals provided by the Bureau of Health of Taipei City	Each intervention woman matched by a control based on their age group, education level, work status, and pregnancy GA. Controls selected from those women who did not attend the educational program; received the usual prenatal care.	No definition of EBF or how obtained	Initiation: EBF not significantly higher for intervention group vs. controls at 3 days (56.5% vs. 41.3%, p = .14). Duration: not assessed. Exclusivity: EBF almost significantly higher for intervention group vs. controls at 1 month (52.2% vs. 32.6%, p = .06).
C. H. Lin, 2008	Quasi-experimental	100	Primigravidas 36–39 weeks GA who had decided on a cesarean delivery; from prenatal clinic in a large urban hospital in Taipei, Taiwan	In addition to standard care, education program included a 22-page booklet and a 22-min videotape with two follow-up phone calls; provided information on BF benefits, discussed difficulties and solutions with rooming-in and EBF, demonstrated how the woman’s partner could assist with BF. Literature review and interviews with five couples choosing cesarean childbirth; 10 nurses supported the development of the program. 10–15 min spent highlighting primary points of the booklet contents during phone calls; responded to any questions.	Researcher	Standard hospital care: (1) Activities to promote BF in prenatal check-up and postnatal ward. (2) BF teaching pamphlets. (3) Epidural anesthetics during cesarean section. (4) Within 30 min after delivery, mother and infant placed together for physical contact and BF. (5) 4 hours after delivery, baby moved to the postnatal ward while mother receives BF education and rooming-in and BF benefits explained. (6) Women discharged on 4th day after cesarean.	BF data collected 4 days after delivery. BF divided into two categories: EBF (excluding any formula and water) and non-EBF (a combination of human milk and formula or water).	Initiation: EBF significantly higher in the intervention vs. control group during the hospital stay: 79.6% vs. 52.2% (p < .05). Duration: not assessed. Exclusivity: EBF significantly higher in the intervention vs. control group at 1 month: 75.9% vs. 34.7% (p < .001).
MacArthur, 2009	Cluster RCT	66 Prenatal clinics with 2,511 pregnant women	33 Clinics in Birmingham, UK: 1,140 women > 24 weeks GA randomized to receive peer support and 33 clinics with 1,371 women randomized to standard care.	In addition to usual prenatal care from midwives, provided prenatal peer support worker service: peer support worker made initial introduction in the antenatal clinic followed by a minimum of two contacts—one at 24–28 weeks GA and one around 36 weeks GA. At least one contact made in the home. Duration of each support session was based on need. Peer support worker followed up women who initiated BF to give postnatal support. The purpose of the antenatal consultations was to provide advice and information on BF benefits and to support women with particular cultural barriers or concerns.	Trained peer support workers of similar ethnic and sociodemographic backgrounds to their clinic population	Control clusters received standard prenatal care: usual information and advice from midwives on BF without input from community peer support workers	Initiation of BF obtained from computerized maternity records of the hospitals where women delivered	Initiation: no significant difference: 69.0% of intervention group and 68.1% of control; cluster adjusted OR = 1.11, 95% CI [0.87, 1.43]. Ethnicity, parity, and mode of delivery independently predicted initiation, but randomization to the peer support worker service did not. Duration: not assessed. Exclusivity: not assessed.
Mattar, 2007	RCT	401 (123 Group A; 132 Group B; 146 Group C)	Random sample of low-risk prenatal patients ≥ 36 weeks GA from clinics in the National University Hospital, Singapore	Group A received an information booklet describing BF techniques and benefits, with practical advice on feeding techniques, expressing milk, and management of common BF problems. Patients watched 16-min educational video introducing benefits of BF, correct positioning, latch-on, breast care, and common concerns. Each woman had one 15-min session with an IBCLC who examined the woman’s nipples to assess adequacy for BF and answered questions on BF. Group B received the same booklet and watched the video but did not have an individual session with the IBCLC.	IBCLC	Group C did not receive the BF booklet, watch the video, or have counseling. No attempt made to prevent the women in any group from seeking additional information or BF assistance from the regular clinic and hospital resources. Standard care for pregnancy and delivery, including access to postnatal BF support.	Primary outcomes: EBF (no formula or water) and predominant BF (water allowed) at 2 weeks, 6 weeks, 3 months, and 6 months. Secondary outcomes: ABF at same intervals. BF initiation was defined as ABF within the first 2 weeks of delivery. Questionnaires administered 1 day after delivery and 6 weeks later, by phone or in person during clinic visit. Follow-up at 3 and 6 months by phone.	Initiation: not assessed. Duration: not assessed. Exclusivity: at 3 months, 36.0% Group A vs. 17.9% Group C EBF or predominant BF (p < .01, OR = 2.6, 95% CI [1.2, 5.4]). At 6 months, 20.0% Group A vs. 9.5% Group C (p < .047, OR = 2.4, 95% CI [1.0, 5.7]). After correction for multiple comparisons, the difference in EBF and predominant BF rates between Groups A and C remained significant only at 3 months (p < .03). No difference between Groups B and C for BF at any time point.
Maycock, 2013	RCT	699 Couples	Fathers whose partners were prenatal patients at eight public maternity hospitals in Perth, Western Australia	Fathers Infant Feeding Initiative (FIFI Study): 2-hour prenatal education session provided to fathers led by a male facilitator, followed by a social support package for 6 weeks postnatal. Additional support included printed materials and promotional materials at weekly intervals. At birth, a congratulatory card sent to fathers, and at 4 weeks, a beer can holder bearing the study logo. Average of 6 participants attended each session with range of 2–14 attendees.	Male facilitator	Four weekly prenatal classes offered at all participating hospitals and attended by almost all mothers. Both intervention and controls received usual care available at participating hospitals, including routine hospital and postnatal care.	The outcome variables were ABF, full BF, and full formula feeding assessed at 6 weeks	Initiation: not assessed. Duration: significantly greater ABF at 6 weeks: 81.6% intervention group vs. 75.2% controls, OR = 1.46, 95% CI [1.01, 2.13]. After adjustment for age and hospital, OR for intervention vs. control ABF was 1.58, 95% CI [1.06, 2.35]. Infants of older vs. younger fathers more likely to be BF at 6 weeks (p < .01), and infants of fathers of high vs. low socioeconomic status more likely to ABF (p = .013). Exclusivity: not assessed.
Meedya, 2014	Quasi-experimental (nonrandomized)	366	Nulliparous women ≥ 19 years early in second trimester from tertiary, metropolitan hospital in Sydney, Australia	Milky Way intervention: weekly 90-min sessions comprising 3 elements: (1) group sessions, (2) take-home learning activities, and (3) postnatal BF support by phone. Grounded in midwifery theories and self-efficacy, class content based on Ten Steps. Women and partner/support person participated in groups of 10–20. During sessions, IBCLC attempted to strengthen BF intention and self-efficacy using verbal persuasion, visualization of BF success, observation of BF photos, hands-on activities, and role playing. Encouraged women to connect with social support including family and friends who had experienced prolonged BF. Learning strategies included giving the women a BF calendar, DVDs, and booklet of motivational postcards.	IBCLC	Standard maternity care: prenatal visits; intrapartum and postnatal care by midwives and/or obstetricians. Midwives and doctors provided BF information during prenatal care. Women with major BF issues referred to lactation resource midwife in the hospital. Offered attendance at parenting education classes via five 2-hour evening sessions or one full-day session. Early skin-to-skin contact and initiating BF in the first hour provided as standard care; included postnatal BF education support by lactation resource midwife.	Primary outcome was ABF at 6 months postpartum. Secondary outcomes were ABF at 1 and 4 months. EBF at 1, 4, and 6 months also presented.	Initiation: not assessed. Duration: intervention mothers had higher rates of ABF at 1 (83.7% vs. 61.3%, p < .001), 4 (64.5% vs. 37.1%, p < .001), and 6 months (54.3% vs. 31.4%, p < .001) vs. controls. Exclusivity: more intervention mothers EBF at 1 (63.9% vs. 49.9%), 4 (50.5% vs. 44.4%), and 6 months (19.1% vs. 6.6%) vs. controls; difference only significant at 6 months, p = .045.
Muirhead, 2006	RCT	225	Prenatal patients at 28 weeks GA from a general practice in Ayrshire, Scotland	In addition to usual BF support, peer support group was visited ≥ 1 time prenatally by peer supporter and further support provided to women on request. Peer support available to intervention women by phone or in person if they were BF after hospital discharge ≥ every 2 days or as often as required until day 28. If requested, the same peer supporters provided further support after 28 days until 16 weeks.	12 Peer supporters, experienced BF mothers recruited and trained by trial team.	Usual BF support: a community midwife for the first 10 days, health visitor after 10 days, BF support groups and BF workshops. Hospital midwives helped initiate BF.	Main outcomes were the differences in breastfeeding initiation and duration	Initiation: 53% of controls and 54% of peer support groups initiated BF. Duration: peer support group higher ABF at 6, 8, and 16 weeks vs. controls, but not significant at any time. Median duration among primigravidae was 7 days for peer support group vs. 3 days for the controls. Among women who initiated, the median was 72 days for peer support group vs. 56 days for controls; differences not significant. Exclusivity: not assessed.
Nichols, 2009	RCT with convenience sample	117	Women ≥ 36 weeks intending to BF recruited at prenatal clinics of three South East Queensland, Australia, hospitals	Self-efficacy–based education intervention: nine-page interactive workbook to enhance BF self-efficacy based on four sources of self-efficacy information specified by Bandura; workbook information was solution-oriented to increase likelihood of positive cognitive appraisal. Participants asked to record their reactions and relevant experiences. The workbook was self-contained and did not link other potential support or training received by participants. Participants received written and verbal instructions on how to complete the workbook and to return it via an addressed stamped envelope within 2 weeks or before the birth of their baby. Prompts to complete and return the workbooks made by phone at 10 days and 2 weeks after receiving the workbook.	In addition to the research team, a panel of consultants at the Queensland College of Lactation Consultants reviewed the workbook for accuracy and suitability for the BF population.	Five-page exploratory interactive workbook on parenting issues with no reference to BF. Participants received written and verbal instructions on how to complete the workbook and to return it via an addressed stamped envelope within 2 weeks or before the birth of their baby, with same prompts as intervention.	BF defined as ABF within the past 24 hours, assessed at 4 weeks. BF further classified into following categories: (a) EBF (human milk only); (b) almost EBF (human milk and other fluids but not formula); (c) high BF (< 1 bottle of formula per day); (d) partial BF (≥ 1 bottle of formula per day); (e) token BF (breast given to comfort baby but not for nutrition); and (f) bottle feeding (no human milk).	Initiation: not assessed. Duration: when prenatal BF self-efficacy used as a covariate, trend toward a significant difference between intervention and control groups for number of days of BF (p = .05). Significant relationship between change in BF self-efficacy and number of days ABF for both groups (p = .0001). Exclusivity: intervention mothers showed a trend toward significant difference in EBF (p = .07).
Noel-Weiss, 2006	RCT	109	Nulliparous women intending to BF, planning an uncomplicated birth at a large tertiary hospital in Ontario, Canada	2.5-hour prenatal BF workshop designed using Bandura’s theory of self-efficacy and adult learning principles; four sources influence self-efficacy: performance accomplishment, vicarious learning, social/verbal persuasion, and emotional/physiological arousal. Four sources provided using dolls, videos, and discussions in a comfortable atmosphere. Adult learning principles assume adults are self-directed, self-motivated, and come to learning with past experience; various strategies used in accordance with these principles.	Not described	No workshop; no other standard of care described	BF duration measured at 4 and 8 weeks by asking whether and how much mother BF. Duration in number of days and amount of BF. EBM means only fluid received is human milk—with or without vitamins, whether by breast or by bottle as expressed human milk.	Initiation: not assessed. Duration: mean difference in number of days BF at 8 weeks not significant based on intention to treat (p = .948), whereas a significant difference detected when actual workshop attendance was analyzed (p = .020). Exclusivity: based on intention to treat, OR of EBF at week 8 not significantly higher for intervention mothers (OR = 1.7, 95% CI [0.73, 4.07]); based on actual attendance, attenders significantly more likely to be EBF than nonattenders (OR = 3.2, 95% CI [1.26, 7.94]).
Ochola, 2013	Cluster RCT: nine villages were assigned on a 1:1:1 ratio	360	Pregnant HIV-negative women 34–36 weeks GA reporting at health center for prenatal services in Kibera slum, Nairobi, Kenya	Home-based intensive counseling group (HBICG) received seven counseling sessions at home by trained peers—one prenatally and six postnatally, each lasting 30–40 min. Facility-based semi-intensive counseling group (FBSICG) received only one counseling session prenatally, conducted by investigator and BF counselors.	Three women with a minimum of secondary level education, residing in study area, recruited as BF counselors.	Standard health and nutrition education offered by health center staff; offered to women in group sessions depending on time constraints and staff availability. Nutrition education allocated about a third of 1 hour/week allowed for health education. BF promotion: 1 out of 10 topics covered. No additional counseling on BF, by the research team, provided to the study participants in this group.	The primary outcome was EBF at 1, 3, and 6 months determined by 24-hour recall. Secondary outcome was EBF since birth, asked at 6 months. Infants who received only human milk and/or medicines or oral rehydration solutions categorized as EBF.	Initiation: not assessed. Duration: not assessed. Exclusivity: at 1 month, significantly higher EBF among FBSICG (84.3%) and HBICG (87.0%) mothers vs. controls (72.0%). At 3 months, HBICG mothers were 2 times as likely to EBF vs. controls; no significant difference for FBSICG vs. controls. At 6 months, HBICG mothers 4 times as likely to EBF vs. controls; no significant difference for FBSICG vs. controls. At 6 months, 3.2% controls EBF since birth vs. 6.9% FBSICG and 15.6% HBICG. HBICG significantly more likely (RR = 3.40, 95% CI [1.34, 8.80]) to EBF since birth vs. controls. FBSICG not significantly more likely to EBF (RR = 1.46, 95% CI [0.49, 4.33]) vs. controls. Mean EBF duration: controls = 1.4 months; FBSICG = 1.8 months; HBICG = 2.4 months.
Otsuka, 2014	Quasi-experimental intervention (convenience sample) with control	781	Prenatal patients in third trimester of singleton pregnancy drawn from prenatal wards of four Japanese hospitals (two BFHI and two non-BFHI)	Conventional support services and self-efficacy intervention: BF self-efficacy workbook for women in third trimester with encouragement to complete it before delivery	Panel of three BF experts, two IBCLCs, and a university professor of nursing evaluated workbook content and judged it to be relevant to mothers in Japan	Access to the conventional in-hospital and community support services	Current infant feeding status by 24-hour recall at 4 and 12 weeks and classified into six categories based on Labbok’s BF classification	Initiation: not assessed. Duration: in BFHIs, intervention nonsignificantly improved full BF at 4 weeks (adjusted OR = 2.12, 95% CI [0.75, 6.01]). In non-BFHIs, no intervention effect on feeding status at 4 weeks. No effect at 12 weeks in either BFHI or non-BFHI hospitals. Exclusivity: intervention improved EBF at 4 weeks in BFHIs (adjusted OR = 2.32, 95% CI [1.01, 5.33]). In non-BFHIs, no intervention effect at 4 weeks; no effect at 12 weeks in either BFHI or non-BFHI hospitals.
Petrova, 2009	RCT	104	Convenience sample of WIC-qualified third trimester mostly Hispanic women; from maternal/pediatric ambulatory care center for low-income inner-city population, New Brunswick, NJ, USA	Two prenatal one-on-one BF education sessions and postdelivery support at the hospital or by phone soon after discharge, at the end of the first or second week, and in the first and second month	IBCLC	Routine BF education and support during pregnancy and postpartum. IBCLC services available for any postpartum women with BF problems during hospital stay.	Infant feeding during the first 7 days and 1, 2, and 3 months. BF categorized as follows: (1) EBF (human milk only or human milk and vitamins); (2) partial BF (human milk and any amount of formula per day); (3) bottle feeding (no human milk).	Initiation: program participation: 16.7% increase (adjusted OR = 1.66, 95% CI [0.55, 4.99]) in EBF at 7 days. Duration: not assessed. Exclusivity: program participation: 6.6% increase (adjusted OR = 1.34, 95% CI [0.41, 4.42]), 6.6% increase (adjusted OR = 1.70, 95% CI [0.47, 6.10]), and 3.4% increase (adjusted OR = 2.21, 95% CI [0.42, 11.7]) in EBF at 1, 2, and 3 months, respectively.
Ryser, 2004	RCT	54	Low-income pregnant women ≥ 18 years unsure about feeding method or planning to bottle feed, receiving care at a private, urban physician’s office in Houston, TX, USA	Education program (Best Start) included: (1) counseling session, (2) viewing video, and (3) written materials addressing common BF barriers perceived by low-income women using a social marketing approach. Presented across four prenatal visits.	Researcher	No exposure to the program	BF at 7 days postdelivery	Initiation: at 1 week, mothers who received Best Start program were BF more than those who had not received the program (p < .01). Duration: not assessed. Exclusivity: not assessed.
Sandy, 2009	RCT	281	Mostly Hispanic pregnant women selected from 588 families participating in Best Beginnings, a primary prevention home visitation program in Washington Heights, New York City, USA	Prenatal and postpartum BF support within a larger home visitation program: explored mother’s previous experience with infant feeding; explained BF mechanics using charts and written materials; provided pamphlets, manuals, and handouts about BF; and discussed benefits and challenges of BF. PC visited mother in the hospital to assist with any problems initiating BF. After discharge, PCs offered information and support weekly in the home. If a mother encountered difficulties after hospital discharge, PC referred to local WIC lactation clinic. A pediatric resident visited families at home after birth.	PC: bilingual “family support workers” from similar ethnic backgrounds and all born outside the United States; pediatric resident	Provided information about community services and referrals to community agencies during home visits; gave educational booklets and pamphlets covering topics related to birth, child rearing, and infant feeding	Infant-feeding method between birth and 7 days postpartum: current exclusive bottle feeding, EBF, or breast and bottle feeding	Initiation: no significant difference in ABF at 7 days, with 86% exposed mothers and 78% nonexposed mothers ABF (OR = 1.73, 95% CI [0.88, 3.40]). EBF at 7 days significantly associated with intervention exposure, with 32% exposed mothers EBF compared to 20% nonexposed mothers (OR = 1.92, 95% CI [1.05, 3.52]). Duration: not assessed. Exclusivity: not assessed.
Srinivas, 2014	RCT	120	Prenatal patients ≥ 28 weeks GA drawn from Westown Physician Center, a hospital-affiliated urban clinic in Cleveland, OH, USA	PC contacted women between 28 weeks GA and 1 week before delivery, with additional contacts at mother’s request. PC contacted the intervention group 3–5 days after delivery, weekly to 1 month, every 2 weeks up to 3 months, and once at 4 months, in person during clinic visits or via phone. Although not part of protocol, at maternal request, PC visited a few mothers in hospital.	WIC definition of a BF PC: woman who BF her own infant(s) to 1 year with EBF for 6 months or is currently BF following recommended practice	Standard of care included access to IBCLCs in hospital and outpatient lactation support from clinic pediatricians and the WIC nutritionist. During the study, the WIC site had a peer helper less than once a month.	BF initiation was defined as ABF attempted after birth. EBF defined as only BF since birth. BF status obtained by maternal self-report.	Initiation: no difference in ABF (PC Group = 86% vs. control = 77%, p = .34) or EBF before discharge (44% vs. 40%, p = .95). Duration: median duration of ABF: PC group = 7.0 weeks vs. control = 6.0 weeks. No difference in ABF at 1 month (68% vs. 53%, p = .14) or 6 months (4% in each group). PC group tended to achieve personal BF goal, but difference did not achieve significance (43% vs. 22%, p = .073). Adjusting for postnatal maternal self-efficacy, PC (vs. control) was significantly associated with higher ABF at 1 month (OR = 3.2, 95% CI [1.02, 9.8]). Exclusivity: not assessed.
Su, 2007	RCT	450	Women with uncomplicated pregnancies, > 34 weeks GA at time of delivery, and intending to BF; from a tertiary hospital in Singapore	One prenatal BF education session: shown a 16-min educational video entitled “14 Steps to Better BF,” which introduced benefits of BF, demonstrated correct positioning, latch, and breast care, and discussed common concerns; printed guides on BF and an opportunity to talk to an IBCLC for about 15 min. Additional routine intrapartum and postnatal obstetric care available.	IBCLC	Routine prenatal, intrapartum, and postnatal obstetric care with no special intervention applied: included optional prenatal classes addressing infant feeding, and postnatal visits by an IBCLC should any BF problems arise	Primary outcomes: EBF at discharge, 2 weeks, 6 weeks, 3 months, and 6 months. Secondary outcomes: ABF at those time points.	Initiation: no significant differences in ABF or EBF at hospital discharge. Duration: no statistically different duration of ABF at 2 or 6 weeks, or 3 or 6 months. Exclusivity: significant differences in EBF at 6 weeks and 3 and 6 months (no significant differences at 2 weeks).
Wambach, 2011	Prospective, nonblinded, three-group RCT	390	Girls 15–18 years in second trimester; mostly Black, single, and primiparous; from bistate metropolitan area in midwestern United States. Enrolled at four prenatal clinics affiliated with three urban teaching hospitals. Three prenatal clinics and four high schools added during the study, with two affiliated hospitals.	Intervention based on Theory of Planned Behavior: two prenatal classes provided content from the BF Educated and Supported Teen Club (BEST) curriculum: encouraged to bring support person to class to enhance social network support. PC phone calls occurred before and after Class 1 and following Class 2 to provide ongoing decision-making support and information. In-hospital support included a PC visit for encouragement and support for early BF. Teens choosing to BF received an IBCLC visit. Postpartum phone calls with IBCLC and/or PC occurred at 4, 7, 11, and 18 days and 4 weeks for those who initiated BF, unless they ceased BF before 4 weeks; provided support and advice to address barriers. Prenatal classes were 1.5 and 2 hours in length.	IBCLC (also a registered nurse) and PC	Two groups: one received standard prenatal and postpartum care at respective clinics with varying provider types and birth settings; one was an attention control group: advanced-practice nurse and PC team taught two prenatal education classes with no focus on BF and same amount of content and time as intervention; plus PC prenatal phone support and in-hospital PC visit focused on postpartum maternal physical recovery and maternal role adjustment. Postdischarge, those BF received postpartum phone calls from PCs.	BF initiation defined as initiating in hospital with intention to provide ≥ 50% feeds at breast or with pumped human milk; measured by self-report. Posthospital discharge data collected by phone at 3 weeks, 6 weeks, 3 months, and 6 months from all participants who initiated BF and continued until BF ceased if that occurred before 6 months. Initiation and duration to 6 weeks validated with postpartum clinic medical records.	Initiation: proportion ABF in experimental (EX), attention control (AC), and usual care (UC) groups was 79%, 66%, and 63%, respectively; significant association between EX and BF initiation (p < .03); no significant differences between the EX, AC, and UC groups (65%, 68%, and 60%, respectively) in initiating EBF vs. partial BF (p = .60). Duration: median ABF duration was 177, 42, and 61 days for EX, AC, and UC groups, respectively; significantly longer among EX compared to control groups (p < .001); effect remained significant (p = .015) after adjusting for positive BF sentiment and social and professional support. Exclusivity: not assessed.
Wolfberg, 2004	RCT	Not clear: 567 expectant mothers approached; 59 couples completed the study	Mostly black expectant fathers; partners of women seeking prenatal care in resident and faculty practices at Johns Hopkins Hospital; approached during the first and second trimesters	Groups of 4–12 expectant fathers attended one 2-hour class, facilitated by a man who is a father. Informal space to discuss beliefs and values about BF and learn to advocate for partner and health of the baby by supporting BF. Used a variety of teaching media, including videos, descriptive color slides, and role play.	Black father facilitator	Control class covered infant care and safety with no BF content using the same facilitator and methods of interactive and informal education as the intervention group.	Initiation and continued ABF at the follow-up intervals measured by patient’s self-report	Initiation: significant difference (74% intervention mothers vs. 41% controls) (p = .02). Duration: no significant difference at 4, 6, or 8 weeks.
Wong, 2014	RCT	469	Pregnant nulliparous women intending to BF, attending antenatal clinics of two geographically distributed public hospitals in Hong Kong	Standard prenatal care plus one 20- to 30-min one-to-one BF education and support session based on WHO guidelines for BFHI: colored flipchart used to illustrate content and handouts distributed at the end of intervention. Topics included: benefits of EBF, risks of formula supplementation, rooming-in, normal milk supply, feeding frequency, feeding cues, return to work, etc. Active communication with family/peers encouraged. At the end of the session, mothers had 10–15 min to ask questions and address concerns.	Registered nurse who completed WHO/UNICEF’s BF course and had substantial BF research experience; midwives and obstetricians provided standard care	Standard hospital care includes routine maternal and fetal health checks and health education to promote healthy pregnancy and BF. Childbirth preparation and BF classes available to mothers at no cost. Postdischarge follow-up at clinic.	Primary outcome was EBF at 6 weeks. Secondary outcomes were EBF at 3 and 6 months as well as the overall duration of any and EBF across the first 6 months. Reported over previous 24 hours. Also compared ABF at 6 weeks and 3 and 6 months.	Initiation: not assessed. Duration: ABF at 6 weeks not significantly different. No significant differences in the overall duration of ABF (HR = 1.11, 95% CI [0.88, 1.40]) across 6-month follow-up. Exclusivity: intervention mothers no more likely to EBF at 6 weeks vs. controls (37.8% vs. 36.4%, p = .77), 3 months (26.6% vs. 25.9%, p = .85), and 6 months (14.6% vs. 12.7%, p = .55). No significant differences in the overall duration of EBF (HR = 0.96, 95% CI [0.79, 1.17]) across 6-month follow-up. Adjusted OR of EBF at 6 weeks was 1.06 (95% CI [0.73, 1.56]).

Note. ABF = any breastfeeding; BF = breastfeeding; BFHI = Baby-Friendly Hospital Initiative; BMI = body mass index; CI = confidence interval; EBF = exclusive breastfeeding; EP = electronically prompted; GA = gestational age; HR = hazard ratio; IBCLC = International Board Certified Lactation Consultant; LC = lactation counselor; NPSGs = nutrition and parenting support groups; OR = odds ratio; PC = peer counsel; RCT = randomized controlled trial; RD = registered dietician; RR = risk ratio; UNICEF = United Nations International Children’s Emergency Fund; WHO = World Health Organization; WIC = Special Supplemental Nutrition Program for Women, Infants, and Children.

Prenatal Interventions

Seventeen studies (Forster et al., 2004; Huang et al., 2007; Ickovics et al., 2007; Kamran et al., 2012; Kellams et al., 2016; Kronborg et al., 2012; Lavender et al., 2005; C. H. Lin et al., 2008; S. S. Lin et al., 2007; Mattar et al., 2007; Nichols et al., 2009; Noel-Weiss et al., 2006; Otsuka et al., 2014; Ryser, 2004; Su et al., 2007; Wolfberg et al., 2004; Wong et al., 2014) evaluated interventions delivered in the prenatal period without additional intrapartum or postpartum components. Of these, 10 assessed breastfeeding initiation and 5 (Huang et al., 2007; Ickovics et al., 2007; C. H. Lin et al., 2008; Ryser, 2004; Wolfberg et al., 2004) found a statistically significant increase in breastfeeding initiation. Of the 10 studies that assessed breastfeeding duration, 2 (Huang et al., 2007; Noel-Weiss et al., 2006) found a statistically significant increase. Finally, of the 12 prenatal studies that assessed breastfeeding exclusivity, 6 (Huang et al., 2007; Kamran et al., 2012; C. H. Lin et al., 2008; Mattar et al., 2007; Noel-Weiss et al., 2006; Otsuka et al., 2014) showed a statistically significant increase.

Six of the 17 prenatal interventions found a significant difference for all breastfeeding outcomes measured (Huang et al., 2007; Ickovics et al., 2007; C. H. Lin et al., 2008; Mattar et al., 2007; Noel-Weiss et al., 2006; Ryser, 2004). Huang et al. (2007) used a web-based breastfeeding program with text, graphics, animations, interactive flash animations, audio, and video to provide prenatal education to nulliparous women planning to breastfeed at the time of enrollment. This intervention included a real-time communication system and message board to allow program participants to interact with one another for interpersonal support, leading to significant increases in breastfeeding initiation, duration, and exclusivity through 6 weeks (Huang et al., 2007). Another prenatal intervention improved both breastfeeding duration and exclusivity at 8 weeks postpartum using self-efficacy and adult learning principles delivered in a 2.5-hour prenatal breastfeeding workshop, with additional tools such as dolls, videos, and group discussions (Noel-Weiss et al., 2006). C. H. Lin et al. (2008) significantly improved breastfeeding initiation and exclusivity at 1 month among women with a scheduled cesarean childbirth in Taiwan by providing educational booklets in combination with a 22-minute videotape and two follow-up phone calls shortly before delivery. Mattar et al. (2007) significantly increased exclusive breastfeeding at 3 and 6 months by providing a 15-minute session with a lactation counselor alongside an informational booklet and a 16-minute educational video. In a separate arm of this study, education provided without the lactation counselor component yielded no significant increase in exclusive breastfeeding (Mattar et al., 2007). Two interventions that significantly improved breastfeeding initiation incorporated interpersonal support into prenatal education in the form of group prenatal care with health care providers (Ickovics et al., 2007) or four prenatal counseling sessions to address breastfeeding barriers perceived by low-income women (Ryser, 2004).

Seven studies found no significant differences between intervention and control groups for all breastfeeding outcomes assessed (Forster et al., 2004; Kellams et al., 2016; Kronborg et al., 2012; Lavender et al., 2005; S. S. Lin et al., 2007; Nichols et al., 2009; Wong et al., 2014). These interventions provided prenatal education using a variety of tools, such as an interactive workbook (Nichols et al., 2009) or an informational video (Kellams et al., 2016; Lavender et al., 2005); however, the lack of interpersonal support in all but one (Wong et al., 2014) of these seven interventions may have contributed to the nonsignificant effect on breastfeeding outcomes. Furthermore, three of these interventions used comparison groups that received BFHI standard of care (Forster et al., 2004; Kronborg et al., 2012; Wong et al., 2014), which may have led to a nonsignificant effect of additional prenatal breastfeeding education.

Prenatal Interventions Combined With Intrapartum or Postpartum Support

Twenty-one studies described prenatal interventions delivered to women with additional intrapartum or postpartum support or counseling components. Of the 15 studies reporting breastfeeding initiation outcomes, 8 found a statistically significant effect on initiation (Anderson et al., 2005; Bonuck et al., 2014; Bonuck et al., 2005; Chapman et al., 2004; Gill et al., 2007; Kupratakul et al., 2010; Sandy et al., 2009; Wambach et al., 2011). Of the 15 that measured breastfeeding duration outcomes, 7 reported a significant intervention effect (Bonuck et al., 2014; Bonuck et al., 2005; Gill et al., 2007; Kupratakul et al., 2010; Maycock et al., 2013; Meedya et al., 2014; Wambach et al., 2011). Of the 13 studies assessing breastfeeding exclusivity, 9 showed a significant increase on exclusivity outcomes (Aidam et al., 2005; Anderson et al., 2005; Bich et al., 2014; Bonuck et al., 2014; Gijsbers et al., 2006; Gross et al., 2016; Kupratakul et al., 2010; Meedya et al., 2014; Ochola et al., 2013).

Nine of the prenatal interventions delivered alongside intrapartum or postpartum support found a significant difference for all breastfeeding outcomes measured. Two significantly increased breastfeeding initiation, duration, and exclusivity (Bonuck et al., 2014; Kupratakul et al., 2010). Bonuck et al. (2014) provided IBCLC support alone or in combination with electronically prompted guidance from a health care provider at five prenatal visits for predominantly low-income mothers in New York. Electronically prompted guidance included several open-ended questions where providers normalized breastfeeding, clarified knowledge about how long and how much to breastfeed, and discussed the importance of social support, and IBCLC support included two prenatal sessions, a hospital visit, and postpartum phone calls (Bonuck et al., 2014). The second intervention, conducted in Thailand, provided a 3-hour knowledge-sharing and empowerment-building program added to routine prenatal breastfeeding education coupled with postnatal support by phone and home visits for women who reported breastfeeding problems (Kupratakul et al., 2010). Three interventions delivered postpartum telephone support alongside one-on-one (Gill et al., 2007) or group prenatal education (Meedya et al., 2014; Wambach et al., 2011) to significantly increase breastfeeding initiation (Gill et al., 2007; Wambach et al., 2011), duration (Gill et al., 2007; Meedya et al., 2014; Wambach et al., 2011), and exclusivity (Meedya et al., 2014). All three interventions were delivered by an IBCLC who in Wambach et al. (2011) also worked collaboratively with a peer counselor to support young women ages 15 to 18 years. The two interventions with group prenatal education sessions used theories of midwifery and self-efficacy (Meedya et al., 2014) or the Theory of Planned Behavior (Wambach et al., 2011) to promote breastfeeding among women and their primary support partners. Three interventions, conducted in Ghana, the Netherlands, and Kenya, provided prenatal education alongside home-based support by a variety of breastfeeding support personnel to significantly increase breastfeeding exclusivity (Aidam et al., 2005; Gijsbers et al., 2006; Ochola et al., 2013).

Seven studies found no significant differences between intervention and control groups for all breastfeeding outcomes measured (Bonuck et al., 2014; Chapman et al., 2013; Graffy et al., 2004; MacArthur et al., 2009; Muirhead et al., 2006; Petrova et al., 2009; Srinivas et al., 2014). Four of these interventions used peer counselors to offer prenatal and postpartum breastfeeding education and support and found no significant increase in breastfeeding initiation (Graffy et al., 2004; MacArthur et al., 2009; Muirhead et al., 2006; Srinivas et al., 2014) or duration (Graffy et al., 2004; Muirhead et al., 2006; Srinivas et al., 2014). In one of the intervention arms of their study, Bonuck et al. (2014) found no significant improvement in breastfeeding initiation, duration, or exclusivity for mothers who received electronically prompted anticipatory guidance from their prenatal providers at five prenatal visits when no additional lactation support was provided. In a BFHI setting, Chapman et al. (2013) reported no significant additional improvement in breastfeeding initiation, duration, or exclusivity where mothers with a high pre-pregnancy body mass index were offered 3 prenatal visits, in-hospital visits after birth, and up to 11 postpartum home visits by a peer counselor. Finally, Petrova et al. (2009) found no significant increase in initiation or exclusivity from two prenatal one-on-one education sessions with an International Board Certified Lactation Consultant (IBCLC), possibly due to a ceiling effect of high levels of breastfeeding in the mostly Hispanic population across both the intervention and control groups.

Prenatal Interventions Involving Partners or Other Supportive Individuals

Evidence for the positive effect of incorporating partners into prenatal breastfeeding education comes from three interventions directly targeting expectant fathers (Bich et al., 2014; Maycock et al., 2013; Wolfberg et al., 2004) and five studies that encouraged participation of the woman’s primary support person (Anderson et al., 2005; Bonuck et al., 2014; C. H. Lin et al., 2008; Meedya et al., 2014; Wambach et al., 2011). All of these studies found significant differences in at least one breastfeeding outcome measure (initiation, duration, or exclusivity). Group sessions with a facilitator to discuss benefits of breastfeeding, breastfeeding myths, and overcoming difficulties were effective in increasing breastfeeding initiation (Maycock et al., 2013; Wolfberg et al., 2004) and exclusivity (Bich et al., 2014; Maycock et al., 2013). In one of these interventions, an additional mass media campaign used posters, public events, and giveaways to promote fathers’ involvement in breastfeeding support (Bich et al., 2014). Among studies that encouraged partner participation in education and support offered to women, two studies involved partners in prenatal clinic visits as well as postpartum home visits by an IBCLC (Bonuck et al., 2014) or a peer counselor (Anderson et al., 2005) with significant increases in breastfeeding initiation and exclusivity. Two interventions invited partners to participate in group prenatal education targeting breastfeeding intention and self-efficacy (Meedya et al., 2014) and enhancing social network support (Wambach et al., 2011) with significant increases in breastfeeding duration. Finally, C. H. Lin et al. (2008) significantly increased breastfeeding initiation and exclusivity by including partners in a prenatal education program to review their role in assisting with breastfeeding before elected cesarean childbirth.

Risk of Bias Assessment

The risk of bias assessment is presented in Table 2. Seven studies demonstrated a high risk of bias due to the lack of adequate randomization procedures or allocation concealment. Prenatal education interventions that significantly increased breastfeeding initiation, duration, and exclusivity were drawn from studies assessed to have a generally low risk of bias. Common sources of bias across studies in this review included the lack of blinding of participants and intervention personnel. Eleven studies showed a high risk of bias due to incomplete outcome data. Selective reporting was not an issue for the selected studies because they specified breastfeeding outcome variables of interest prior to intervention administration and data collection.

Table 2.

Risk of Bias Assessment

First author, year	Random sequence generation	Allocation concealment	Blinding of participants/personnel	Blinding of outcome assessment	Incomplete outcome data
Aidam, 2005	Low risk; randomization achieved by writing numbers 1–3 on folded pieces of paper. Blinded researcher shuffled numbers and gave one to each mother to determine group assignment.	Low risk; randomization numbers not viewed by study staff or mothers, and the pieces of paper all looked identical from the outside	High risk; counselors aware of the study design and group allocation of mothers	High risk; two research assistants who were not the intervention counselors collected outcome data, but they were aware of mother’s allocation	Low risk; 10% LTFU
Anderson, 2005	Low risk; participants randomly assigned to either the intervention or control group by the study field coordinator. SPSS used for random assignment of women drawn from one BFHI-affiliated prenatal center.	Unclear risk; not described	Unclear risk; appears to have been no blinding	High risk; interviewer knew the study hypotheses	High risk; 26% LTFU at 3 months
Bich, 2014	Low risk; nonadjacent but demographically similar districts randomly assigned to intervention and control groups	High risk; impossible to conceal allocation	High risk; impossible to blind participants to participation	Low risk; independent team of data collectors for outcome assessment	Low risk; < 10% LTFU
Bonuck, 2005	Low risk; used undisclosed blocking factor and stratification according to community health center of women	Low risk; research assistant shared demographic and contact information with study LC but did not discuss baseline BF intentions, knowledge, or attitudes	High risk; participants and study LCs not blinded to treatment assignment but prohibited from discussing or sharing feeding data	High risk; research assistants collecting BF outcomes not blinded to treatment assignment but prohibited from discussing or sharing feeding data	Low risk; LTFU: 17% from intervention and 16% controls
Bonuck, 2014	Low risk; randomization by sequentially numbered opaque sealed envelopes, generated by study biostatistician; incorporated an undisclosed blocking factor and nativity status (U.S.-born vs. foreign-born)	Low risk; restricted access to allocation assignment, stripped group assignment from study databases to which research staff had access, and omitted group identifiers from participant interview forms	High risk; participants and clinical staff not blinded	Low risk; blinded outcomes assessors	Low risk; 11% LTFU
Chapman, 2004	High risk; randomization to control or intervention group using SPSS but excluded many at delivery who were in the originally randomized sample	Unclear risk; not described	High risk; participants and clinical staff not blinded	Low risk; not double blinded but interviewers unaware of group assignment at beginning of each outcome assessment, and questions pertaining to PC contact asked at the end of the interview	High risk; of 219 randomized, 165 received intervention/control (25% excluded) and 144 followed through 6 months (13% LTFU)
Chapman, 2013	Low risk; each week, the study coordinator used SPSS to randomly assign 50% of newly recruited participants to the intervention group	Low risk; study coordinator did not know participant assignment	High risk; participants and clinical staff not blinded	Low risk; interviewer was not informed of participants’ group assignment but was not completely blinded because she collected PC contact data. To minimize potential bias, PC contact questions were asked at the end of each interview.	High risk; 30% of those who received the intervention LTFU (28% intervention vs. 32% control); of those initially allocated: 48% LTFU (47% intervention vs. 49% controls)
Forster, 2004	Low risk; using computerized system of biased urn randomization. Data collected about ineligibility and why women declined participation.	Low risk; the research midwife ascertained allocation by phone	High risk; participants and clinical staff not blinded	Unclear; not described	Low risk; follow-up rates in practical skill, attitudes, and standard care were 91%, 90%, and 91%, respectively
Gijsbers, 2006	Low risk; a family practice was assigned to the control or intervention group using a randomization list prepared by the statistical adviser when the first potential participant of a particular practice received information about the study	High risk; impossible to maintain allocation concealment: once randomized, families assigned to the same group as their practice	High risk; participants and clinical staff not blinded	Low risk; cluster randomized since intervention could not be performed in a blinded fashion	High risk; 21% LTFU
Gill, 2007	High risk; convenience sample: recruitment ceased when 100 mothers meeting inclusion criteria were enrolled in each group	Unclear risk; not described	High risk; participants and clinical staff not blinded	Unclear; not described	Low risk; 9% LTFU
Graffy, 2004	Low risk; used numbered, sealed envelopes prepared by statistical adviser from random permuted blocks held in the study office; sample was stratified by practice and birth order using separate sets of envelopes for mothers of first and subsequent babies	Low risk; the identity of participants was held separately from the data records prepared when questionnaires were returned	Low risk; one-sided blinding	Low risk; responses were coded blind to treatment allocation. A second researcher checked the coding of responses to open questions, and they checked data for consistency before analysis. PCs played no part in assessing feeding outcome.	Low risk; 14% LTFU
Gross, 2016	Low risk; women were randomized using a random number generator, stratified by site	Low risk; enrolled women were randomized after completing the baseline assessment and attending a prenatal visit ≥ 32 weeks gestational age	Unclear risk; women randomized by a nutritionist who conducted the intervention; no mention if participants were aware of their group randomization. However, control group appears to have received only standard prenatal and postnatal care, so participants do not seem to have been blinded.	Low risk; group assignments were concealed from the research assistants, who conducted the follow-up assessments	Low risk; LTFU = 12% control and 17% intervention
Huang, 2007	Low risk; although women were not allocated to intervention or control, historic controls used had generally similar demographic characteristics as mothers who participated in the intervention	Low risk; intervention and control mothers drawn from different time periods, so concealment not required	Low-risk; one-sided blinding, as mothers did not know about change in prenatal BF program	Unclear; not described	Low risk; 8% LTFU
Ickovics, 2007	Low risk; a computer-generated randomization sequence, password protected to recruitment staff and participants, was used to assign participants	Low risk; allocation concealed from participant and research staff until eligibility screening was completed and study condition was assigned	High risk; participants and clinical staff not blinded	Low risk; all measurement and data collection were conducted in a blinded manner independently of the clinical care setting	Low risk; 11% LTFU
Kamran, 2012	Low risk; whole health centers randomized to intervention or control to prevent cross-contamination	Unclear risk; not described	High risk; participants and clinical staff not blinded	Unclear risk; not described	Unclear/low risk; appears that no women were LTFU
Kellams, 2016	Low risk; computer-generated block randomization sequence using random block sizes, stratified by prenatal clinic	Low risk; one member of the study team with no direct contact with participants prepared all numbered, sealed, opaque envelopes, which the research assistant opened just prior to loading the video for the participant to view	Low risk; women randomly assigned to view one of two educational videos but did not know that the intervention was education on BF or what the outcomes were	Low risk; research assistants abstracting outcome data were blinded to the group to which the participant was assigned	Low risk; < 5% LTFU
Kronborg, 2012	Low risk; computer generated	Unclear risk; not described	Unclear risk; not described	Unclear risk; not described	Low risk: 16 of 603 (3%) and 15 of 590 (3%) women in the intervention and reference groups were LTFU
Kupratakul, 2010	Low risk; randomization scheme generated using a random number table	Low risk; co-investigator generated allocation sequence while principal investigator enrolled participants and assigned them to groups. When a woman met the inclusion criteria, the principal investigator picked up a sequentially numbered opaque envelope that contained a ticket identifying the group.	High risk; participants and clinical staff not blinded	Unclear risk; not described	Low risk; < 12% LTFU
Lavender, 2005	Unclear; not described	Low risk; opaque sealed envelopes were used	High risk; participants and clinical staff not blinded	Low risk; outcome assessors were blinded	Low risk; < 10% LTFU
S. S. Lin, 2007	High risk; not randomized but matched on observable covariates	High risk; women not allocated to intervention or control	High risk; participants and clinical staff not blinded	Low risk; maternal self-report of outcomes through questionnaires	Unclear risk; not described
C. H. Lin, 2008	Low risk; although randomization not possible, the study compared women seeking a cesarean at two different time points over the course of 1 year	Low risk; intervention and control mothers drawn from different time periods, so concealment not required	Unclear risk; not described	Unclear risk; not described	Low risk; 14% LTFU
MacArthur, 2009	Low risk; computer generated	Unclear risk; not described	High risk; not feasible to blind participants to participation	Low risk; data on outcome were supplied to the research team in an anonymous format	Low risk; 113 of 2,511 women (4.5%) who participated in the trial were not available for primary outcome assessment
Mattar, 2007	Low risk; computer generated	Low risk; each woman allocated to the intervention group next on the list after written informed consent had been obtained	High risk; participants and clinical staff not blinded	Low risk; blinded outcomes assessors	Low risk; 10% LTFU
Maycock, 2013	Low risk; using a random number generator	Unclear risk; not described	High risk; participants and clinical staff not blinded	Low risk; self-completion questionnaires were administered by printed questionnaire or by phone (by the research officer) according to their individual preferences	Low risk; 14% LTFU
Meedya, 2014	High risk; not randomized	High risk; women were not allocated to intervention or control	High risk; participants and researchers not blinded	High risk; the study was conducted by the primary researcher who was not blind to the study group assignments	Low risk; 6.6% LTFU
Muirhead, 2006	Low risk; computer-generated block randomization	Low risk; postrecruitment concealed allocation	High risk; participants and peer supporters not blinded	Low risk; questionnaires used to obtain outcome data, but counselors were present during their completion	Low risk; < 3% LTFU
Nichols, 2009	High risk; randomization of a convenience sample using consecutively numbered sealed opaque envelopes	Unclear risk; not described	High risk; participants and researchers not blinded	Low risk; participants phoned by a researcher blinded to group allocation at 4 weeks postpartum to collect outcome data	High risk; 23% LTFU
Noel-Weiss, 2006	Low risk; participants returned their registration package in a sealed envelope, and randomization was completed by matching the envelope with a sealed, sequentially numbered, opaque envelope containing a slip of paper stating either Control or Workshop.	Unclear risk; not described	High risk; participants and clinical staff not blinded	Low risk; research assistant blinded to the mother’s group assignment and workshop attendance	Low risk; 16% LTFU
Ochola, 2013	Low risk; 9 out of 10 villages in Kibera were randomly assigned to the three study groups on a 1:1:1 ratio. Villages rather than women were randomized to prevent cross- contamination.	Unclear risk; not described	High risk; participants and researchers not blinded	Low risk; enumerators conducting the interviews to determine BF practices were blinded	High risk; 26% LTFU
Otsuka, 2014	High risk; recruited from a convenience sample: In two of the hospitals, for the first 2 months of recruitment, all eligible women were assigned to the intervention group when they reached the third trimester. After a month, all eligible women who reached third trimester and were not in the intervention group were assigned to the control group for the following 2 months of recruitment. In the other two hospitals, the order of assignment to intervention and control groups was reversed.	Unclear risk; not described	High risk; participants and researchers not blinded	Low risk; self-report questionnaires were dropped in a collection box or mailed	High risk; 25%–34% LTFU
Petrova, 2009	High risk; recruited from convenience sample of women at prenatal visits in third trimester	Low risk; computer-generated random numbers to assign women to the control and intervention groups. Each random number was related to an ordinal number assigned to the woman once she signed the informed consent.	High risk; participants and clinical staff not blinded	Unclear; blinding not mentioned	High risk; 30% LTFU
Ryser, 2004	Unclear risk; not described	Unclear risk; women selected a sealed envelope	Unclear risk; appears to have been no blinding	Unclear risk; not described	Low risk; no LTFU
Sandy, 2009	Unclear risk; not described: appear to have analyzed a subgroup of a larger randomized controlled trial	Unclear risk; not described	High risk; participants and family support workers not blinded	High risk; family support workers collected outcome data	Low risk; 15% LTFU
Srinivas, 2014	Low risk; the Iowa Infant Feeding Attitude Scale (IIFAS) measuring mother’s attitude to BF was used to stratify to positive (IIFAS score ≥ 58) or negative (IIFAS score ≤ 57) BF attitude; study participants randomized within these strata in blocks of four participants in a 1:1 ratio to intervention or control	Unclear risk; not described	High risk; participants and peer supporters not blinded	High risk; not blinded	Low risk; 15% LTFU
Su, 2007	Low risk; computer generated	Low risk; research assistant managed allocation concealment	High risk; participants and clinical staff not blinded	Unclear risk; data appear to have been collected by study staff but no data analyst appears to have been blinded	Low risk; 18% LTFU
Wambach, 2011	Low risk; codes generated by a biostatistician	Unclear risk; not described	High risk; participants and clinical staff not blinded	Unclear risk; data appear to have been collected by study staff	Low risk; < 4% LTFU
Wolfberg, 2004	Unclear risk; not described	Unclear risk; not described	High risk; not feasible to blind participants to participation	Unclear risk; not described	High risk; 36%LTFU
Wong, 2014	Low risk; randomization by sequentially numbered opaque sealed envelopes; used block randomization procedures	Low risk; an independent researcher who did not participate in recruitment or data collection generated the allocation sequence using Stata software	High risk; participants and researchers not blinded	Low risk; research assistant blinded to the mother’s group allocation	Low risk; < 6% LTFU

Note. BF = breastfeeding; BFHI = Baby-Friendly Hospital Initiative; EBF = exclusive breastfeeding; LC = lactation counselor; LTFU = loss to follow up; SPSS = Statistical Package for the Social Sciences.

Discussion

There is mounting evidence that prenatal interventions, alone or in combination with intrapartum and/or postpartum support, can increase breastfeeding initiation, duration, and exclusivity. Although we recognize that some of the effect of interventions combining prenatal education with intrapartum or postpartum support may not be attributable to the prenatal component alone, we included evidence from combination interventions to better understand broad delivery methods for realistic application within the Ten Steps.

Prenatal education that includes both informational materials and interpersonal support increases breastfeeding initiation, duration, and exclusivity. Effective interventions included both provision of information along with support from an IBCLC, peer counselor, medical provider, or researcher. These findings align with previous systematic reviews that showed that health education and peer support interventions increase breastfeeding initiation and that peer counseling, IBCLCs, and formal prenatal breastfeeding education increase breastfeeding duration (Dyson, McCormick, & Renfrew, 2005; Lumbiganon et al., 2012). Interpersonal support included home visits, telephone support, and novel web-based formats that allowed mothers to communicate in real time to provide peer support.

Breastfeeding education was delivered through home visits and individual or group sessions at prenatal clinics. When combined with interpersonal support, a variety of education technologies were effective in increasing breastfeeding initiation, duration, or exclusivity, such as web-based tools with interactive flash animations and a message board (Huang et al., 2007) and informational videos (Anderson et al., 2005; C. H. Lin et al., 2008; Mattar et al., 2007; Ryser, 2004; Su et al., 2007); however, without additional support, the use of education technology was not adequate to increase breastfeeding outcomes (Kellams et al., 2016; Lavender et al., 2005; Mattar et al., 2007; Noel-Weiss et al., 2006). Although we did not include pilot studies in this review, a recent videoconferencing pilot intervention highlights the promise of incorporating new technologies to connect hospital-based health care providers with difficult-to-reach communities to provide both breastfeeding education and support (Friesen, Hormuth, Petersen, & Babbitt, 2015). Text-messaging technologies have also been shown to increase exclusive breastfeeding duration when delivered during the postpartum period (Gallegos, Russell-Bennett, Previte, & Parkinson, 2014) and should be considered for use in prenatal education by future studies.

Involving a woman’s family in prenatal education can improve breastfeeding initiation, duration, and exclusivity. Health care providers should consider how the attitudes of the infants’ father or grandmother toward breastfeeding may influence women’s breastfeeding intentions and success (Arora, McJunkin, Wehrer, & Kuhn, 2000; Grassley & Eschiti, 2008). In addition, Odom, Li, Scanlon, Perrine, and Grummer-Strawn (2013) found that maternal perceptions of their health care providers’ infant-feeding recommendations interact with the opinions of family members to affect breastfeeding initiation. Therefore, shared decision making regarding infant feeding should be integrated within family dynamics, with clear messaging, support, and information from providers.

Our review has several limitations. Studies were identified using only two databases and manuscripts published in languages other than English were excluded, limiting generalizability. Although the diversity of contexts, breastfeeding definitions, and comparison groups of included studies improve the generalizability of results to various clinical settings, the heterogeneity of interventions and differences in study quality limit our ability to recommend any single intervention as most effective. Furthermore, because we were unable to quantitatively compare the strength of findings across studies, clinicians should consider results from those interventions that have similar patient populations and available resources. In addition, only one study (Wambach et al., 2011) assessed the strength of participant breastfeeding plans at baseline, which was found to be the greatest predictor of breastfeeding initiation in their sample. Most of the other studies included some measure of breastfeeding plans or previous experience; however, baseline assessments would benefit from inclusion of the Infant Feeding Intentions Scale to assess the strength of maternal plans for breastfeeding exclusivity and duration, which is correlated with breastfeeding duration (Nommsen-Rivers & Dewey, 2009). Without a validated pre-intervention measure of the strength of breastfeeding intention, room for improved outcomes associated with the intervention is unclear. Such ceiling effects mask the effectiveness of the interventions themselves. Finally, information about the cost of delivering prenatal breastfeeding interventions was not available for most studies; changes in breastfeeding intentions and realization of infant-feeding goals must be considered alongside cost-effectiveness when determining the most appropriate prenatal breastfeeding education interventions for particular settings.

Conclusion

Prenatal education that includes both informational materials and interpersonal support is positively associated with breastfeeding initiation, duration, and exclusivity. Breastfeeding support delivered in a variety of formats, including home visits and individual or group sessions at prenatal clinics, is associated with improved breastfeeding outcomes. However, lack of standardized assessment of participants’ breastfeeding intentions limits the comparability of the interventions assessed in this review. If women have strong plans for breastfeeding initiation, duration, and exclusivity, the effectiveness of prenatal education and support may be masked by ceiling effects. Additional research could improve understanding of the mechanisms by which Step 3 increases breastfeeding outcomes by including pretesting and posttesting of the strength of maternal breastfeeding intentions, further assessment of the interactive role of family members, comparison of the effectiveness of different providers involved in prenatal and postnatal breastfeeding support, and explicit detail about the time and cost associated with different intervention delivery formats.

Footnotes

Authors’ Note

Kathryn Wouk is a doctoral student in the Gillings School of Global Public Health. She is a reproductive, perinatal, and pediatric epidemiology trainee in the Department of Maternal and Child Health.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a NICHD predoctoral training grant (5T32HD052468-08) to K.W. and by the Carolina Global Breastfeeding Institute (K.W., M.H.L., and K.P.T.).

References

Aidam

B. A.

Pérez-Escamilla

Lartey

(2005). Lactation counseling increases exclusive breast-feeding rates in Ghana. The Journal of Nutrition, 135(7), 1691–1695. Retrieved from http://jn.nutrition.org/content/135/7/1691.full.pdf

Anderson

Damio

Young

Chapman

Pérez-Escamilla

(2005). A randomized trial assessing the efficacy of peer counseling on exclusive breastfeeding in a predominantly Latina low-income community. Archives of Pediatric & Adolescent Medicine, 159(9), 836–841. doi:10.1001/archpedi.159.9.836

Arora

McJunkin

Wehrer

Kuhn

(2000). Major factors influencing breastfeeding rates: Mother’s perception of father’s attitude and milk supply. Pediatrics, 106(5), E67.

Bich

T. H.

Hoa

D.T.P.

Målqvist

(2014). Fathers as supporters for improved exclusive breastfeeding in Viet Nam. Maternal and Child Health Journal, 18, 1444–1453. doi:10.1007/s10995-013-1384-9

Bonuck

Stuebe

Barnett

Labbok

M. H.

Fletcher

Bernstein

P. S.

(2014). Effect of primary care intervention on breastfeeding duration and intensity. American Journal of Public Health, 104(Suppl. 1), S119–S127. doi:10.2105/AJPH.2013.301360

Bonuck

Trombley

Freeman

McKee

(2005). Randomized, controlled trial of a prenatal and postnatal lactation consultant intervention on duration and intensity of breastfeeding up to 12 months. Pediatrics, 116(6), 1413–1426. doi:10.1542/peds.2005-0435

Chapman

D. J.

Damio

Young

Pérez-Escamilla

(2004). Effectiveness of breastfeeding peer counseling in a low-income, predominantly Latina population: A randomized controlled trial. Archives of Pediatrics & Adolescent Medicine, 158(9), 897–902. doi:10.1001/archpedi.158.9.897

Chapman

D. J.

Morel

Bermúdez-Millán

Young

Damio

Pérez-Escamilla

(2013). Breastfeeding education and support trial for overweight and obese women: A randomized trial. Pediatrics, 131(1), e162–e170. doi:10.1542/peds.2012-0688

de Oliveira

M. I.

Camacho

L. A.

Tedstone

A. E

. (2001). Extending breastfeeding duration through primary care: A systematic review of prenatal and postnatal interventions. Journal of Human Lactation, 17(4), 326–343. doi:10.1177/089033440101700407

10.

Dyson

McCormick

F. M.

Renfrew

M. J.

(2005). Interventions for promoting the initiation of breastfeeding. Cochrane Database of Systematic Reviews, (2), 1–42. doi:10.1590/1516-3180.20141321T1

11.

Forster

McLachlan

Lumley

Beanland

Waldenstrom

Amir

(2004). Two mid-pregnancy interventions to increase the initiation and duration of breastfeeding: A randomized controlled trial. Birth, 31(2), 176–182.

12.

Friesen

C. A.

Hormuth

L. J.

Petersen

Babbitt

(2015). Using videoconferencing technology to provide breastfeeding support to low-income women: Connecting hospital-based lactation consultants with clients receiving care at a community health center. Journal of Human Lactation, 31(4), 595–599. doi:10.1177/0890334415601088

13.

Gallegos

Russell-Bennett

Previte

Parkinson

(2014). Can a text message a week improve breastfeeding? BMC Pregnancy and Childbirth, 14, 374. doi:10.1186/s12884-014-0374-2

14.

Gijsbers

Mesters

Knottnerus

J. A.

Kester

A.D.M.

van Schayck

C. P.

(2006). The success of an educational program to promote exclusive breastfeeding for 6 months in families with a history of asthma: A randomized controlled trial. Pediatric Asthma, Allergy & Immunology, 19(4), 214–222. doi:10.1089/pai.2006.19.214

15.

Gill

S. L.

Reifsnider

Lucke

J. F.

(2007). Effects of support on the initiation and duration of breastfeeding. Western Journal of Nursing Research, 29, 708–723. doi:10.1177/0193945906297376

16.

Graffy

Taylor

Williams

Eldridge

(2004). Primary care for mothers considering breast feeding. Primary Care, 328, 1–6.

17.

Grassley

Eschiti

(2008). Grandmother breastfeeding support: What do mothers need and want? Birth, 35(4), 329–335. doi:10.1111/j.1523-536X.2008.00260.x

18.

Gross

Mendelsohn

Gross

Scheinmann

Messito

(2016). Randomized controlled trial of a primary-care based child obesity prevention intervention on infant feeding practices. Journal of Pediatrics. Advance online publication. Retrieved from http://dx.doi.org/10.1016/j.jpeds.2016.03.060

19.

Howe-Heyman

Lutenbacher

(2016). The Baby-Friendly Hospital Initiative as an intervention to improve breastfeeding rates: A review of the literature. Journal of Midwifery & Women’s Health, 61(1), 77–102. doi:10.1111/jmwh.12376

20.

Huang

M. Z.

Kuo

S. C.

Avery

M. D.

Chen

Lin

K. C.

Gau

M. L.

(2007). Evaluating effects of a prenatal web-based breastfeeding education programme in Taiwan. Journal of Clinical Nursing, 16(8), 1571–1579. doi:10.1111/j.1365-2702.2006.01843.x

21.

Ickovics

J. R.

Kershaw

T. S.

Westdahl

Magriples

Massey

Reynolds

Rising

S. S.

(2007). Group prenatal care and perinatal outcomes: A randomized controlled trial. Obstetrics & Gynecology, 110(2, Pt. 1), 330–339. doi:10.1097/01.aog.0000275284.24298.23

22.

Imdad

Yakoob

M. Y.

Bhutta

Z. A.

(2011). Effect of breastfeeding promotion interventions on breastfeeding rates, with special focus on developing countries. BMC Public Health, 11(Suppl. 3), S24. doi:10.1186/1471-2458-11-s3-s24

23.

Kamran

Shrifirad

Mirkarimi

S. K.

Farahani

(2012). Effectiveness of breastfeeding education on the weight of child and self-efficacy of mothers. Journal of Education & Health Promotion, 1, 11. doi:10.4103/2277-9531.98569

24.

Kellams

A. L.

Gurka

K. K.

Hornsby

P. P.

Drake

Riffon

Gellerson

. . . Coleman

(2016). The impact of a prenatal education video on rates of breastfeeding initiation and exclusivity during the newborn hospital stay in a low-income population. Journal of Human Lactation, 32(1), 152-159. doi:10.1177/0890334415599402

25.

Kronborg

Maimburg

R. D.

Vaeth

(2012). Antenatal training to improve breast feeding: A randomised trial. Midwifery, 28(6), 784–790. doi:10.1016/j.midw.2011.08.016

26.

Kupratakul

Taneepanichskul

Voramongkol

Phupong

(2010). A randomized controlled trial of knowledge sharing practice with empowerment strategies in pregnant women to improve exclusive breastfeeding during the first six months postpartum. Journal of the Medical Association of Thailand, 93(9), 1009–1018.

27.

Lavender

Baker

Smyth

Collins

Spofforth

Dey

(2005). Breastfeeding expectations versus reality: A cluster randomised controlled trial. BJOG: An International Journal of Obstetrics and Gynaecology, 112(8), 1047–1053. doi:10.1111/j.1471-0528.2005.00644.x

28.

Liberati

Altman

D. G.

Tetzlaff

Mulrow

Gotzsche

P. C.

Ioannidis

J. P.

. . . Moher

(2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. Journal of Clinical Epidemiology, 62(10), e1–e34. doi:10.1016/j.jclinepi.2009.06.006

29.

Lin

C. H.

Kuo

S. C.

Lin

K. C.

Chang

T. Y.

(2008). Evaluating effects of a prenatal breastfeeding education programme on women with caesarean delivery in Taiwan. Journal of Clinical Nursing, 17(21), 2838–2845. doi:10.1111/j.1365-2702.2008.02289.x

30.

Lin

S. S.

Chien

L. Y.

Tai

C. J.

Lee

C. F.

(2007). Effectiveness of a prenatal education programme on breastfeeding outcomes in Taiwan. Journal of Clinical Nursing, 17(3), 296–303. doi:10.1111/j.1365-2702.2006.01927.x

31.

Lumbiganon

Martis

Laopaiboon

Festin

M. R.

J. J.

Hakimi

(2012). Antenatal breastfeeding education for increasing breastfeeding duration. Cochrane Database of Systematic Reviews, (9), CD006425. doi:10.1002/14651858.CD006425.pub3

32.

MacArthur

Jolly

Ingram

Freemantle

Dennis

C. L.

Hamburger

. . . Khan

(2009). Antenatal peer support workers and initiation of breast feeding: Cluster randomised controlled trial. British Medical Journal (Clinical Research Ed.), 338, b131. doi:10.1136/bmj.b131

33.

Mattar

C. N.

Chong

Y. S.

Chan

Y. S.

Chew

Tan

Chan

Y. H.

Rauff

M. H.

(2007). Simple antenatal preparation to improve breastfeeding practice: A randomized controlled trial. Obstetrics & Gynecology, 109(1), 73–80. doi:10.1097/01.AOG.0000249613.15466.26

34.

Maycock

Binns

C. W.

Dhaliwal

Tohotoa

Hauck

Burns

Howat

(2013). Education and support for fathers improves breastfeeding rates: A randomized controlled trial. Journal of Human Lactation, 29(4), 484–490. doi:10.1177/0890334413484387

35.

Meedya

Fahy

Yoxall

Parratt

(2014). Increasing breastfeeding rates to six months among nulliparous women: A quasi-experimental study. Midwifery, 30(3), 137–144. doi:10.1016/j.midw.2013.12.010

36.

Muirhead

P. E.

Butcher

Rankin

Munley

(2006). The effect of a programme of organised and supervised peer support on the initiation and duration of breastfeeding: A randomised trial. British Journal of General Practice, 56(524), 191–197. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1828262/pdf/bjpg56-191.pdf

37.

Munn

Newman

Mueller

Phillips

Taylor

(2016). The impact in the United States of the Baby-Friendly Hospital Initiative on early infant health and breastfeeding outcomes. Breastfeeding Medicine, 11(5), 1–9. doi:10.1089/bfm.2015.0135

38.

Nichols

Schutte

N. S.

Brown

R. F.

Dennis

C. L.

Price

(2009). The impact of a self-efficacy intervention on short-term breast-feeding outcomes. Health Education & Behavior: The Official Publication of the Society for Public Health Education, 36(2), 250–258. doi:10.1177/1090198107303362

39.

Noel-Weiss

Rupp

Cragg

Bassett

Woodend

A. K.

(2006). Randomized controlled trial to determine effects of prenatal breastfeeding workshop on maternal breastfeeding self-efficacy and breastfeeding duration. Journal of Obstetric, Gynecologic & Neonatal Nursing, 35(5), 616–624. doi:10.1111/j.1552-6909.2006.00077.x

40.

Nommsen-Rivers

L. A.

Dewey

K. G.

(2009). Development and validation of the Infant Feeding Intentions scale. Maternal & Child Health Journal, 13(3), 334–342. doi:10.1007/s10995-008-0356-y

41.

Ochola

S. A.

Labadarios

Nduati

R. W.

(2013). Impact of counselling on exclusive breast-feeding practices in a poor urban setting in Kenya: A randomized controlled trial. Public Health Nutrition, 16(10), 1732–1740. doi:10.1017/s1368980012004405

42.

Odom

E. C.

Scanlon

K. S.

Perrine

C. G.

Grummer-Strawn

(2013). Association of family and health care provider opinion on infant feeding with mother’s breastfeeding decision. Journal of the Academy of Nutrition and Dietetics, 114(8), 1203–1207. doi:10.1016/j.jand.2013.08.001

43.

Otsuka

Taguri

Dennis

C. L.

Wakutani

Awano

Yamaguchi

Jimba

(2014). Effectiveness of a breastfeeding self-efficacy intervention: Do hospital practices make a difference? Maternal & Child Health Journal, 18(1), 296–306. doi:10.1007/s10995-013-1265-2

44.

Pérez-Escamilla

Martinez

Segura-Pérez

(2016). Impact of the Baby-Friendly Hospital Initiative on breastfeeding and child health outcomes: A systematic review. Maternal & Child Nutrition, 2(3), 402–417. doi:10.1111/mcn.12294

45.

Petrova

Ayers

Stechna

Gerling

J. A.

Mehta

(2009). Effectiveness of exclusive breastfeeding promotion in low-income mothers: A randomized controlled study. Breastfeeding Medicine, 4(2), 63–69. doi:10.1089/bfm.2008.0126

46.

Ryser

F. G.

(2004). Breastfeeding attitudes, intention, and initiation in low-income women: The effect of the Best Start program. Journal of Human Lactation, 20(3), 300–305. doi:10.1177/0890334404266985

47.

Sandy

J. M.

Anisfeld

Ramirez

(2009). Effects of a prenatal intervention on breastfeeding initiation rates in a Latina immigrant sample. Journal of Human Lactation, 25(4), 404–411. doi:10.1177/0890334409337308

48.

Srinivas

G. L.

Benson

Worley

Schulte

(2014). A clinic-based breastfeeding peer counselor intervention in an urban, low-income population: Interaction with breastfeeding attitude. Journal of Human Lactation, 31(1), 120–128. doi:10.1177/0890334414548860

49.

L. L.

Chong

Y. S.

Chan

Y. H.

Chan

Y. S.

Fok

Tun

K. T.

. . . Rauff

(2007). Antenatal education and postnatal support strategies for improving rates of exclusive breast feeding: Randomised controlled trial. British Medical Journal, 335(7620), 596. doi:10.1136/bmj.39279.656343.55

50.

United Nations International Children’s Emergency Fund. (2006). Celebrating the Innocenti Declaration on the Protection, Promotion, and Support of Breastfeeding: Past achievements, present challenges and priority actions for infant and young child feeding. Retrieved from http://www.unicef.org/nutrition/files/Innocenti_plus15_BreastfeedingReport.pdf

51.

Wambach

K. A.

Aaronson

Breedlove

Domian

E. W.

Rojjanasrirat

Yeh

H. W.

(2011). A randomized controlled trial of breastfeeding support and education for adolescent mothers. Western Journal of Nursing Research, 33(4), 486–505. doi:10.1177/0193945910380408

52.

Wolfberg

A. J.

Michels

K. B.

Shields

O’Campo

Bronner

Bienstock

(2004). Dads as breastfeeding advocates: Results from a randomized controlled trial of an educational intervention. American Journal of Obstetrics and Gynecology, 191(3), 708–712. doi:10.1016/j.ajog.2004.05.019

53.

Wong

K. L.

Tak Fong

D. Y.

Yin Lee

I. L.

Chu

Tarrant

(2014). Antenatal education to increase exclusive breastfeeding. Obstetrics & Gynecology, 124(5), 961–968. doi:10.1097/AOG.0000000000000481

54.

Wong

K. L.

Tarrant

Lok

K.Y.W.

(2015). Group versus individual professional antenatal breastfeeding education for extending breastfeeding duration and exclusivity: A systematic review. Journal of Human Lactation, 31(3), 354–356. doi:10.1177/0890334415583294

55.

World Health Organization. (1998). Evidence for the ten steps to successful breastfeeding. Geneva, Switzerland: World Health Organization.

56.

World Health Organization. (2014). WHO handbook for guideline development (2nd ed.). Geneva, Switzerland: World Health Organization.