“It gave me so much confidence”: First‐time U.S. mothers' experiences with antenatal milk expression

Abstract

Antenatal milk expression (AME) involves maternal hand‐expression, collection, and storage of breast milk during pregnancy for the purposes of reducing the early formula use in breastfed infants. AME is not widely practiced in the United States, despite its growing popularity elsewhere. In this study, we examined the experiences of first‐time mothers recruited from a U.S. midwife practice who engaged in AME within the context of a pilot randomized controlled trial. The AME intervention involved demonstration and practice of AME with a lactation consultant beginning at 37 weeks of gestation, reinforcement at weekly study visits until delivery, and daily home practice. Nineteen women participated in a semistructured interview at 1–2 weeks postpartum regarding their study experiences. Major themes included (1) perceived benefits and impact of AME, (b) AME implementation, and (c) use of AME milk. Women perceived multiple benefits of AME, most notably that it increased their confidence that they would be able to make milk and breastfeed successfully postpartum. Women expressed some concern that no/little milk expressed could be indicative of postpartum milk production problems. Regarding implementation, women found that the AME protocol fit well into their daily routine. There was mixed feedback regarding comfort with practicing AME in the presence of partners. Reasons for postpartum use of AME milk varied; barriers to provision included inadequate milk storage options at the birth hospital and unsupportive hospital providers/staff. With few caveats, AME appears to be an acceptable breastfeeding support intervention among a sociodemographically homogeneous group of first‐time mothers in the United States.

Key messages.

• Antenatal milk expression enhanced first‐time mothers' commitment to breastfeeding and confidence in their ability to produce and express milk following delivery.

• Antenatal milk expression may be particularly beneficial among women and infants facing postpartum complications.

• Breastfeeding‐supportive infrastructure and culture at the birth hospital is important if women are to derive maximum benefit from AME.

1. INTRODUCTION

Breastfeeding is considered the biologically normative way to feed infants, with significant dose‐dependent health implications for mothers and their offspring (Bartick et al., 2017; Victora et al., 2016). Accordingly, the American Academy of Pediatrics and World Health Organization recommend 6 months of exclusive breastfeeding, with continuation of any breastfeeding to at least 1 or 2 years postpartum, respectively (American Academy of Pediatrics Section on Breastfeeding, 2012; World Health Organization, 2011). However, despite a slow but steady rise in rates of breastfeeding initiation since the early 1990s to 83% in 2015 among U.S.‐born infants (Centers for Disease Control and Prevention, 2018; Coates, 2016; Herrick, Rossen, Kit, Wang, & Ogden, 2016), rates of breastfeeding duration and exclusivity remain below national targets (U.S. Department of Health and Human Services & Office of Disease Prevention and Health Promotion, 2012). In the first 2 days of life, 17% of breastfed U.S. infants receive formula supplementation. By 6 months, just 25% of infants are exclusively breastfeeding, with 58% breastfeeding at all (Centers for Disease Control and Prevention, 2018). In addition, although 85% of American women state a prenatal intention to exclusively breastfeed at least 3 months (with first‐time mothers having higher or equal rates of breastfeeding intention compared with multiparous women [Chertok, Luo, Culp, & Mullett, 2011; Colaizy, Saftlas, & Morriss, 2012]), just 32% meet their prenatal goal (Perrine, Scanlon, Li, Odom, & Grummer‐Strawn, 2012). Although women discontinue breastfeeding for a variety of reasons, among the most prevalently cited barriers to breastfeeding maintenance are perception of insufficient milk, impaired breastfeeding self‐efficacy, and early, in‐hospital formula supplementation (Bahorski et al., 2018; Brownell, Howard, Lawrence, & Dozier, 2012; Chantry, Dewey, Peerson, Wagner, & Nommsen‐Rivers, 2014; Odom, Ruowei, Scanlon, Perrine, & Grummer‐Strawn, 2013).

Antenatal milk expression (AME) has emerged as a potential intervention to address these issues. AME involves rhythmic self‐massage of breasts, usually commencing between 36 and 37 weeks of gestation, at which point early milk (colostrum) is available and may be collected and frozen for later use (Forster et al., 2017; Singh, Chouhan, & Sidhu, 2011; Soltani & Scott, 2012). Hand expression, rather than an electric breast pump, is the preferred milk removal strategy when milk volumes are small (as in pregnancy), yielding higher milk output (Ohyama, Watabe, & Hayasaka, 2009). Theory and limited research outside the United States suggest that AME may buoy women's breastfeeding confidence, reduce early formula use through “banking” of a back‐up supply of breast milk during pregnancy, and hasten the onset of a more copious milk supply post birth through critical‐period endocrine modulation (Chapman, Pincombe, & Harris, 2013; Cox, 2006; East, Dolan, & Forster, 2014; Fair, Watson, Gardner, & Soltani, 2018; Forster et al., 2011; Parker, Sullivan, Krueger, & Mueller, 2015). However, women's receptivity to AME, integration of AME into daily life during the third trimester, and perceived benefits and barriers of AME have not previously been evaluated within a sample of U.S. women. The purpose of this study was to examine the experiences of first‐time mothers in the United States who participated in a pilot randomized controlled trial (RCT) of AME.

2. METHODS

2.1. Sample, setting, and initial data collection

Between December 2016 and April 2018, women who were assigned to an AME intervention as part of a pilot RCT were invited to participate in a postpartum interview about their study experiences. The trial investigated the impact of antenatal hand expression of colostrum versus an education control on breastfeeding outcomes among a convenience sample of 45 women. Participants were recruited between 34 and 36^6/7 weeks of pregnancy via study flyers and at prenatal visits within a hospital‐based midwife practice at UPMC Magee‐Womens Hospital (Pittsburgh, Pennsylvania). Women were at least 18 years old, nulliparous, pregnant with a single, healthy foetus, planned to exclusively breastfeed/provide breast milk for the first 4 months postpartum, and did not have known major risk factors for insufficient milk supply (e.g., breast hypoplasia, polycystic ovarian syndrome, diabetes, breast reduction surgery) or preterm labour (e.g., vaginal bleeding after the first trimester, congenital anomalies, polyhydramnios, currently smoking). Background data on maternal demographics, medical and reproductive history, and infant/maternal postpartum hospital course were collected via maternal self‐report and the electronic medical record (EMR) review. The study was approved by the University of Pittsburgh Institutional Review Board. All women provided written informed consent for study participation.

2.2. AME intervention

The AME intervention protocol was modelled after an existing randomized controlled trial of AME in women with diabetes (Forster et al., 2014). At an initial AME study visit at 37 weeks of gestation, participants were shown a video modelling hand‐expression of milk (Breast Time of Life, available on YouTube: https://www.youtube.com/watch?v=9gQ8119RNeI); received written and verbal instructions for safe expression, collection, and storage of milk; and engaged in hands‐on practice of AME utilizing the Marmet technique (Mohrbacher & Stock, 2003) with an International Board Certified Lactation Consultant (IBCLC; author J. R. D.). Thereafter, participants met on a weekly basis until delivery or 40^6/7 weeks (whichever occurred first) with the same lactation consultant to reinforce technique and collect a milk sample if possible. Participants were also instructed to engage in at‐home, independent milk expression and collection of one to two times per day for up to 10 min until delivery and record this in a written diary. Women were provided containers in which to collect and freeze milk and instructions on how to transport and store frozen milk at the birth hospital. Women in the education control group received a commercially produced handout from Lactation Education Resources (https://www.lactationtraining.com/resources/educational-materials/handouts-parents) weekly from gestational weeks 37 to 40. Handouts addressed a weekly theme pertaining to breastfeeding preparation or management of common breastfeeding problems (e.g., Week 37: “Sore Nipples”). Control group participants did not receive any education on AME or breastfeeding from the study IBCLC.

2.3. Interviews

Interviews were conducted at a study visit at 1–2 weeks postpartum by author M. G. Interviews followed a semistructured script and included questions about AME problems/concerns, perceived utility/benefits, protocol acceptability and integration into daily routines, and breastfeeding/milk expression practices since birth. The interview guide was modified with progressive interviews to establish consensus and divergence in emerging themes. Interviews were audio recorded and transcribed verbatim.

2.4. Analysis

Two authors (J. F. and E. C.) independently coded 11 of 19 interview transcripts for major themes and subthemes. Author J. R. D. adjudicated any discrepancies between coders and developed a codebook based on this preliminary coding, which was used for third‐pass line‐by‐line coding of all transcripts. During coding, J. R. D. further refined codes in terms of their properties and dimensions and grouped them into categories reflective of themes and subthemes using analytic techniques described by Corbin and Strauss, including constant comparison, questioning, and review of theoretical memos (Corbin & Strauss, 2008).

3. RESULTS

Of 22 women assigned to AME in the RCT, 19 completed an interview, including 18 women who received AME instruction. The participant who did not receive the AME intervention delivered early due to medical complications prior to completing any study visits; she was retained in the analysis to provide additional perspective regarding acceptability of AME unaffected by her perceived “success” with the intervention (e.g., whether or not she was able to express/collect milk). One interviewee was an IBCLC who had prior knowledge of AME. The majority of women were married, college educated, and White, non‐Hispanic. Almost half of the sample developed obstetrical medical complications during the study, and a third of study infants were admitted to the neonatal intensive care unit (NICU) following delivery. Most women were able to visualize and/or collect antenatal milk during the study, with 11 women freezing milk at home and seven feeding antenatal milk to their infant at some point in the postpartum period. Most women were also breastfeeding exclusively and had used an electric breast pump at the time of the 1‐ to 2‐week interview. Interview data did not support that any women had attempted to use an electric breast pump prior to delivery (Table 1).

Table 1.

Sample characteristics (N = 19)

Characteristic	N (%)	Mean (SD); range
Age (years)		31.2 (3.0); 26–36
Married	15 (79)
Education
High school diploma	0
Some college or vocational program	1 (5)
Bachelor's degree	9 (47)
Post‐graduate degree	9 (47)
Race
White/Caucasian	18 (95)
Black/African American	0 (0)
Other	1 (5)
Hispanic ethnicity	1 (5)
WIC recipient	0 (0)
Employed (at enrolment)	19 (100)
Delivery method
Spontaneous vaginal	11 (58)
Assisted vaginal	1 (5)
Caesarean section	7 (37)
Medical complications during pregnancy, delivery, and postpartuma	9 (47)
Endorsement of breast growth during pregnancy	16 (84)
Infant birthweight (g)		3349 (440); 2475–3970
Infant gestational age (weeks)		401/7 (10/7); 375/7–415/7
Infant NICU admission	6 (32)
NICU length of stay (days; n = 6)b		2.5 (2.9); 0–7
Number of AME study visitsc		2.7 (1.1); 1–4
Total number of home AME sessionsc		17.8 (12.0); 0–45
Visualization and/or collection of milk during AMEc	15 (83)
Total milk volume collected via AME (mLs)c		22.5 (29.4); 0–93.3
Formula use during birth hospitalization	9 (47)
Used electric breast pump in first 1–2 weeks of postpartum	12 (63)
Breastfeeding status at time of 1‐ to 2‐week interview
Breast milk only	16 (84)
Breast milk and formula	3 (16)

Abbreviations: AME, Antenatal milk expression; NICU, neonatal intensive care unit.

^aPre‐eclampsia (n = 5), oligohydramnios (n = 1), suspected/confirmed chorioamnionitis (n = 3), and infant resuscitation (assisted ventilation, cardiopulmonary resuscitation; n = 2).

^bNICU hospitalization counted as 0 days when <24‐hr stay.

^cThose who received AME intervention only (n = 18).

3.1. Qualitative findings

Major themes encompassing women's experiences with AME included (a) perceived benefits and impact of AME, (b) AME implementation, and (c) use of AME milk. The overarching theme, which permeated a substantial portion of interview content and tone, was impact of AME on breastfeeding confidence.

3.2. Perceived benefits and impact of AME on postpartum breastfeeding trajectory

3.2.1. Confidence in ability to breastfeed and produce milk postpartum

Most participants were successful in expressing some milk in the antepartum period, and those who were, were effusive in their enthusiasm for AME and its impact on their breastfeeding confidence before and after birth. AME increased these women's commitment to breastfeeding and demystified the mechanics of breastfeeding and milk production. Visualizing milk during AME, particularly when volume increases were observed over the study course, evoked a sense of appreciation for one's body/breasts and provided reassurance that milk would be available postpartum. The participant who did not receive AME echoed this sentiment—anticipating that expressing milk in pregnancy would have provided a head start on ensuring adequate milk production postpartum. For participants unable to express any milk or only drops, reactions varied from indifference to frustration, disappointment, or anxiety regarding potential implications for their postpartum milk supply. These feelings were tempered by counselling regarding the normalcy of varying levels of antepartum milk production from the study IBCLC and perceived benefit of learning hand‐expression skills ahead of birth.

I feel like [AME] really actually helped me with breastfeeding—learning how to express ahead of time … and it helped me build confidence, because I knew that I had milk there and knew how to express it if I needed to.

I just thought it was literally the coolest thing. Because it was like my body is working the way that it's supposed to. I have vials of proof that it's working, and it gave me so much confidence.

Well, the first time that I saw any milk come out, I was like, shocked, in a good way. I was like, “Whoa!” Kind of didn't think it could happen, so it was kind of cool to see that it actually worked. So I think that made me feel good. That made me feel more—much more confident, that I would actually be able to make breast milk. You know that you're supposed to, but it just seems very unlikely if it never happened to you before.

… The more I practiced [AME], the more colostrum I was able to produce, which was really cool, and just kind of reassuring. Just to know that I'm already producing it before the baby gets here, so I feel like there wouldn't be a reason that I wouldn't be producing it once he did get here …

I was only a little anxious [about not seeing any milk during AME], but I know that [the study IBCLC] said that, you know, a lot of women don't get any, so I tried to remember that.

3.2.2. Development of breastfeeding skills and mindset

AME taught women how to hand express milk, which they perceived as useful in the postpartum period. Introduction of hand expression during pregnancy was perceived as an ideal, low‐stakes opportunity to develop embodied knowledge regarding breastfeeding techniques.

I think you guys should teach every single woman how to hand‐express. I think it's a key skill, and they don't teach you that in the hospital.

I think it was really nice for me to learn part of this new, like, motor process, but on my own, and before putting [baby] on [my breast], too. 'Cause it's, like, tougher to [learn to breastfeed] with two people, so it was nice to start one person and then add a second person.

The first time I used the pump, I'm like, “what is this, this is not doing anything.” And I'm like, “I'm just gonna hand express,” and so I was actually hand expressing the colostrum and bringing it down to the NICU for her. Yeah, she was in the incubator for the first little while, and she had an IV and things, too. So it wasn't that easy for us to just zip down there and breastfeed her … I was able to get quite a bit [of milk hand‐expressing postpartum], and I was able to get a lot more hand expressing than using the breast pump.

Women found that AME provided the motivation necessary to prepare oneself ahead of birth for the physical and psychological challenges of breastfeeding. AME also provided a sense of liberation that if breastfeeding did not work out as planned, it was not for lack of effort (“at least I'm doing all the right things”).

I think just having a check‐in person [provides] accountability … it's my homework … I think the other part of the study is like kind of a test of yourself. It's like learning to run a mile, because it's like “gosh, this is really hurting my hand,” … and this is – I would not say hurting my breast, but it's like my breast is not used to this kind of attention. So I think learning how to do that, and like, I would say build a callus … because [now] I'm used to kind of like manhandling my breast. Once [baby] got to it, the pain that women say they feel [with early breastfeeding], like I didn't really have that …

3.2.3. General breastfeeding education

Participants found it helpful to converse with the study IBCLC during hand‐expression about breastfeeding questions or concerns. The AME sessions provided dedicated time and space to consider breastfeeding and postpartum breast changes in concrete terms and discuss how breastfeeding would tangibly work.

… When we were in the hospital, when the milk did come in, it was just like, “OK, this is normal, my breasts are gonna get hard.” So massage and doing warm compresses kinda help[ed] alleviate some of the [breast] tenderness, but having all that knowledge beforehand [was helpful]. Even though like people tell you this is gonna happen, you don't really think about it. But having the hand expression exercises [prepared me]. And the lactation consultant came into our room and just like gave us some tips while we were there, and she was like, “oh, you already know how to do this [hand expression],” so that was also kind of a good thing.

3.2.4. Early availability of milk

Women who collected and froze antenatal milk expressed a sense of satisfaction that they were building a stockpile of milk ahead of birth and that milk would be available immediately if their infant required it (“… so if she needs it, it's there”). Several women who fed antenatally expressed milk to their infants credited AME with helping to avoid or minimize early formula use. Several women thought AME contributed to more abundant colostrum prior to lactogenesis II and their milk “coming in” sooner.

Yeah, it was kind of really difficult the first couple nights, separated from each other [with infant in NICU] … I was so glad we had [the AME milk], 'cause then we were able to give it to [the NICU staff to feed to infant], and at least for that first night, she wasn't fed formula …

Because of the study, I was able to [use AME milk as a supplement to at‐breast feeds], and she never got any formula, even in that high risk type of a situation [mother with preeclampsia, infant in NICU], which I think for sure she would have otherwise … I [also] pumped that night in the ICU and was able to get 8ccs out and was able to give that to her so I could rest. And I thoroughly believe that that would never have happened had I not been doing AME, because just of what I've seen in the hospital, the first time you pump, you get like drops … I don't think it made my mature milk come in any faster, but I felt like I had enough colostrum to satisfy her without having to supplement …

3.3. AME implementation

3.3.1. Format of AME education

Participants found the study's introductory video featuring a real woman (postpartum) expressing and links provided to additional milk expression videos helpful. Women also thought that it was important to have the study IBCLC present at each study visit to offer real‐time support, answer questions, and confirm that their technique was “correct.”

[AME] is not something I've ever done, so I just wanted – any support I could get, I was very happy to get, and just kind of having somebody to check in with. And you guys offered a video too, but then asking for help when I needed it [from the IBCLC at each visit], so like that made me realize that it's not just me that has questions about it, it's just kind of the general population.

3.3.2. General problems/concerns

Women reported few problems with AME. Hand fatigue and breast soreness were mentioned as minor issues that limited AME frequency at times. Participants perceived antenatal milk to be essentially equivalent in composition to postpartum colostrum, although some expressed initial concerns about whether colostrum expressed antenatally would be replenished in the postpartum period.

I think that was a concern of mine, actually, that whenever I expressed prenatally, that I was getting rid of good stuff. But I think I actually asked [study IBCLC] about that, and she said that's not true, so that eased my fears about that.

Women were specifically queried on concerns regarding the theoretical link between nipple stimulation in pregnancy and contractions/early labour via endogenous oxytocin release. Although some women endorsed this as an initial concern and experienced transient, mild uterine contractility during AME, their worries tended to dissipate as they neared their due dates and/or they continued AME without a clear correlation with prodromal labour.

3.3.3. Time/energy burden

In terms of daily AME home practice, women did not find the protocol burdensome or time consuming (“didn't really interfere with my work schedule or life”), but some reported difficulty in remembering to do AME. Women reported engaging in AME once or twice per day for about 10 min, typically at home before and/or after work. One participant noted that she regularly practiced AME during her workday and that doing so empowered her to lay early claim to both the right and space to express milk at work.

I got to commandeer an open office at my work with a door that locks, and so I was like, “hey, you guys, I'm going to be using this office. It is going to be the breastfeeding office.” They were like, “you're still pregnant.” I was like, “yeah, I don't care.” 'Cause I wanted to be able to do it—I wanted to be able to do it twice a day, as many times as possible … It just felt really good, I was like, “yeah, we are doing this at the office. I can do this anywhere.”

3.3.4. Privacy

Discussions about privacy for AME were circumscribed by the fact that all participants were first‐time mothers, the majority living in nuclear households with the father of the baby. In general, participants had few reservations about practicing AME at home around their partners (“wasn't an issue”); others endorsed some degree of initial or continued awkwardness in having their partner present during AME, which variably influenced when and where they chose to practice.

I scheduled [AME] around my boyfriend's schedule, so that he wasn't there sitting, watching me while I was trying to do it … He was usually at work when I would do it. But he was supportive. I just didn't want to freak him out or anything.

… he's still a little bit uncomfortable with the whole, “oh, there's milk in there.” He thinks it's good. He absolutely wants the baby to breastfeed. But [expressing milk in pregnancy] kind of grosses him out, personally. The milk itself coming from my breasts … there was something in my stomach [when I was pregnant], he thinks, but [it wasn't] quite real yet … So I think that was part of why I didn't want to try to express in front of him. And after [birth], I mean, he was [present during] labor, and he saw all that stuff. So I have no shame now. So that's kind of what changed.

In some cases, participants found that AME contributed to their partners' increased enthusiasm, interest, and investment in breastfeeding.

I was like [to my partner], “I'm gonna squeeze my boobs now, OK?” He′s like, “OK.” And he would just do what he was doing. But he was actually paying way more attention than I thought, 'cause he was like, “whoa, you made more [milk] this day.” And then he was like so excited … he took a picture [of all of my antenatal milk] on his phone when I came into the hospital … He just thought it was cool.

3.3.5. Milk collection and storage

Participants did not describe any issues with regard to storage of antenatal milk in their home freezers and found that the provided milk containers (11 ml Snappies® colostrum collectors) established realistic expectations regarding small volumes likely to be obtained. Conversely, when milk volumes were very low (e.g., <1 ml), the comparably larger container also served as barrier to saving/freezing milk versus discarding it (“I only ever had this much [milk] in a thing and I was just like, eh, it's kind of pointless.”). As a workaround, some participants combined several days' worth of refrigerated AME milk into a single container (as permitted by study protocol).

3.4. Use of antenatally expressed milk

3.4.1. Reasons for hospital and home use

Reasons for provision of antenatal milk among five women in the hospital setting and four women within the 2 weeks after hospital discharge included the following: temporary satiation of the infant to enable latching, NICU nurses' recommendations for at‐breast supplementation, participant concern about inadequate infant intake at breast, participant request that the infant be temporarily bottle fed after a prolonged or difficult delivery following initial successful at‐breast feeding, and desire to have a partner share feeding responsibilities (“That was the point of that [AME]—so he could get involved with feeding early on”).

There were a couple times I did use it, whenever she was really, really fussy and trying to feed a lot, just sort of when she got herself really worked up, and I would give her a couple drops [of milk from a syringe] to sort of pacify her, to get her calm enough to breastfeed. So that was very helpful.

3.4.2. Barriers to use

Two participants noted that they would prefer to defer any feeding of antenatal milk via bottle as the infant was learning the mechanics of breastfeeding. Other barriers to use of antenatal milk in the hospital setting included lack of storage options during labour and delivery (particularly when labour was prolonged) and staff discomfort/lack of comfort with provision of antenatally expressed milk, particularly within the NICU. Women suggested providing freezer packs and storage bags as part of study participation, obtaining freezers/refrigerators for antenatal milk on all hospital units, and providing more education to hospital staff about the study protocol and safety of AME milk.

It was the pediatrician who was going to take her to the NICU, and I said, “I have my colostrum in the car. I have frozen colostrum, can you use that,” and he just said “No.” … [He] didn't [give any reason], and at that time, that was when I was going to go in and have the C‐section, so didn't not really continue the conversation … I'm not sure if it had anything to do with me and [my baby] possibly having an infection …

Relatedly, women suggested more guidance for feeding infants small volumes of antenatal milk.

… It would be useful would be if you guys could give some information on, like, eyedropper or syringe feeding for the [antenatal] colostrum. So we ended up—[my husband] looked that up [online]. But we had to figure out, “oh, what are the devices we need in order to feed her this.”

4. DISCUSSION

This is the first study to report on perceptions of and experiences with antenatal hand expression of milk among an exclusive sample of first‐time mothers in the United States. We found that women who engaged in AME once or twice per day from 37 weeks of gestation had overwhelmingly positive views of AME. AME was thought to contribute to breastfeeding success on multiple fronts, most notably in increasing confidence in one's capability to produce milk and express it in the event of postpartum breastfeeding difficulties. Our findings are in line with qualitative and survey research conducted among women in the United Kingdom and Australia, which found that AME was acceptable to women, increased breastfeeding confidence, and contributed to breastfeeding preparation (Brisbane & Giglia, 2015; Fair et al., 2018).

These studies also corroborate some of the same concerns about AME that we documented in our sample, including issues of breast tenderness, privacy/breast exposure, uncertainty about “using up” colostrum, and association of AME with early labour. Our study indicates that most of these issues may be tempered by continued observation and feedback on technique by a lactation expert, trial‐and‐error during home practice, and frank discussions about lactation physiology and evidence‐based research with a knowledgeable healthcare provider. We found that the primary modifiable barrier to use of antenatal milk was inadequate milk storage options and provider reservations about provision of antenatal milk at the delivery hospital. Thus, implementation of similar AME protocols should carefully consider hospital infrastructure and involvement of clinical “champions” from multiple hospital service lines (e.g., labour and delivery, mother/baby, and neonatal intensive care) who are familiar with and supportive of AME.

A novel finding in our study was the impact of AME on women's partners. Recent research, including a meta‐analysis addressing the efficacy of targeting fathers in breastfeeding interventions, supports an independent, positive association between partners' involvement in breastfeeding and higher rates of breastfeeding initiation, continuation, and exclusivity (Hunter & Cattelona, 2014; Mahesh et al., 2018). Feedback from women in our study suggest that AME could be a promising intervention to pre‐emptively leverage partners' investment in and support of early breastfeeding efforts. Relatedly, we also found that AME increased women's commitment to breastfeeding and contributed to a simultaneous sense of attunement with her body/breasts and newfound awareness of her capabilities to produce milk. Thus, inclusion of hand‐expression education and demonstrations in prenatal care may benefit not only women who intend to breastfeed but may also encourage women who are ambivalent about breastfeeding to consider it further.

Our findings indicate that banked AME milk was used and contributed to decreased formula use in the case of pregnancy and labour complications, including prolonged labour, maternal morbidity (e.g., preeclampsia), and NICU admittance. Thus, AME may be of particular benefit to women who are at higher risk for such events, including first‐time mothers, those with diabetes or hypertension, over 35 years of age, overweight/obese, or carrying multi‐foetal pregnancies. Women at heightened risk for low milk supply may also benefit from AME, including those with insufficient glandular breast tissue, polycystic ovarian syndrome, or those who have undergone breast surgery (Cox, 2006). Women with diabetes are considered prime candidates for AME, as their infants often require formula supplementation due to post‐birth hypoglycemia. Two pilot studies (Forster et al., 2011; Soltani & Scott, 2012) and one large RCT (Forster et al., 2014; Forster et al., 2017) have been conducted on the safety and efficacy of antenatal hand expression of colostrum among diabetic women in the United Kingdom and Australia. In the latter, a multicentre, unblinded RCT including 635 diabetic participants, AME twice per day from 36 weeks of pregnancy was found to be feasible and safe (not associated with increased incidence of preterm births, NICU admissions, or neonatal or maternal morbidity). AME was also found to have a beneficial effect on exclusive breast milk feeding in the first 24 hr post birth (aRR 1.15) and during hospitalization (aRR 1.16; Forster et al., 2017). In addition to these studies, a small RCT conducted among healthy nulliparous and multiparous women in India found that significantly more women randomized to antenatal hand expression of colostrum after 37 weeks established “full lactation” where no “top feed” was required within 6 hr postpartum, as compared with usual care (Lamba, Chopra, & Negi, 2016).

Our sample consisted of mostly White women of high socioeconomic status who were highly committed to breastfeeding. Future research should investigate the acceptability, uptake, and efficacy of AME among U.S. women from more diverse backgrounds with different levels of social and breastfeeding support. Extrapolation of our findings is also limited by the idiosyncrasies related to our specific AME protocol (e.g., structure/timing of study visits and characteristics/style of IBCLC interventionist). Wider scale dissemination of AME in prenatal care will necessarily need to consider costs in relationship to availability and quality of key resources, including interventionists, milk collection/storage supplies, and facilities.

An additional point of consideration and caution in adoption of AME into clinical practice is the potential for conflation of the current data on antenatal hand expression with antenatal use of electric or manual breast pumps. The risks and benefits of the latter have not been systematically evaluated, although research suggests that pumps may differ from hand expression in several critical aspects. These include (a) the stimulation/suction action of a pump may induce oxytocin surges associated with labour onset, which may be undesirable (Kavanagh, Kelly, & Thomas, 2005); (b) small volumes of available antenatal milk may accumulate within the pumping apparatus rather than the collection container, reducing the potential for antenatal milk banking (Ohyama et al., 2009); and (c) breast milk of mothers who primarily pump versus directly breastfeed or hand express exhibits less microbial diversity and increased pathogens; this may be due to the impact of the infant's oral bacteria on milk composition (retrograde inoculation hypothesis) and/or contaminants in the pump itself (Moossavi et al., 2019).

Finally, because our AME protocol was developed based on an initial safety trial of AME in a high‐risk population of diabetic women (Forster et al., 2014), we implemented similar conservative eligibility criteria, timing of AME introduction, and restrictions on at‐home practice. The safety, efficacy, and acceptability of potentially implementing AME earlier in pregnancy (thus increasing the “dose” and likely exposure to AME if delivery <37 weeks), with variant levels of intensity (frequency and duration of expression sessions) and with women at higher risk for postpartum breastfeeding complications, should be further explored.

CONFLICTS OF INTEREST

The authors declare that they have no conflicts of interest.

CONTRIBUTIONS

JRD and KH conceived of the study idea and implementation details. MG recruited participants and conducted interviews. JRD, JF and EC contributed to data analysis. JRD wrote the initial manuscript draft. All authors provided critical feedback on manuscript drafts and approved the final version.

Demirci JR, Glasser M, Fichner J, Caplan E, Himes KP. “It gave me so much confidence”: First‐time U.S. mothers' experiences with antenatal milk expression. Matern Child Nutr. 2019; 15:e12824 10.1111/mcn.12824