Interventions to Improve Breastfeeding Outcomes in Late Preterm and Early Term Infants

Abstract

Background:

Late preterm infants (LPIs; born at 34^0/7 to 36^6/7 gestational weeks) and early term infants (ETIs; 37^0/7 to 38^6/7 gestational weeks) are at higher risk of morbidity and mortality compared with more mature infants. Breastfeeding can reduce these risks, but feeding difficulties are common among these infants and breastfeeding rates are low. We conducted a systematic review to identify the interventions available to improve any breastfeeding, exclusive breastfeeding, or breast milk yield.

Methods:

A literature search was performed up to February 23, 2022, using MEDLINE, CINAHL, Embase, and Google Scholar, and nine articles were included. Only one article was a randomized controlled trial, and only one included ETIs. The remaining articles were quasi-experimental and included only LPIs. Outcomes included breastfeeding duration, breastfeeding exclusivity, and/or breast milk production (volume) before 6 months actual age.

Results:

Professional support significantly improved exclusive breastfeeding rates. A breastfeeding education program delivered at the hospital with weekly telephone follow-up postdischarge significantly increased breastfeeding rates. Neither cup feeding nor early discharge (with in-home lactation support) improved breastfeeding rates, whereas rooming-in (versus direct admission to the neonatal intensive care unit) worsened exclusive breastfeeding rates.

Discussion:

This is the first systematic review to identify interventions available for both LPIs and ETIs. Overall, there are limited studies that investigate interventions promoting breastfeeding in these populations. However, breastfeeding support delivered by health care professionals seems to improve breastfeeding rates. The main limitations are the lack of randomization, blinding, and adjustment for confounding variables. Experimental studies with robust methodological design are needed.

Background

More than 1 in 10 babies are born preterm worldwide,¹ and this rate is rising in almost all countries. Prematurity is the leading cause of death in children under the age of 5 years in nearly all high- and middle-income countries and is associated with a higher risk of health complications, such as chronic respiratory problems, diabetes, and hypertension.²

It was suggested that preterm birth rates have primarily increased overall due to the dramatic rise in late preterm births (34^0/7 and 36^6/7 gestational weeks),³ which constitute 74% of all preterm births.⁴ Previously, not enough attention has been paid to late preterm infants (LPIs) due to the incorrect assumption that these infants may be physiologically and metabolically as mature as term infants, as they may appear of appropriate size at birth. However, LPIs have a significantly higher risk of morbidity and mortality compared with term infants.⁵

Another category of infants that has gained more attention in recent years is early term infants (ETIs), who are infants born between 37^0/7 and 38^6/7 gestational weeks and account for 23% of all live births.⁶ The categorization was developed to highlight the higher risk of morbidity and mortality of ETIs compared with those born at 39–41 weeks.⁷ Despite the complications they experience,⁸ like LPIs, most ETIs are of healthy weight at birth and have normal Apgar scores, which might lead to false reassurance among health professionals.

Both groups are at higher risk of breastfeeding complications compared with infants born later.^9–11 This could be due to infant-related barriers associated with earlier birth, such as rapid fatigue during feeding, lower stamina, fewer awake periods, and reduced effort to stimulate and empty the breast.¹² It could also be due to maternal-related barriers associated with preterm delivery, such as cesarean delivery, obesity, multiple births, smoking status, or maternal psychological distress.^8,9,11–13 Additionally, since ETIs are still within the broad full-term categorization, they might not be receiving special attention to overcome the challenges associated with an earlier birth.

Reduced breastfeeding rates in these two groups is an important issue to tackle as all infants benefit from breastfeeding, but especially those born earlier. Due to their developmental immaturity and increased susceptibility to inflammation, oxidative stress, and infections, breast milk and the constituents of breast milk such as antibodies, growth factors, bacteria, lipids, and enzymes (which are variable with gestational age) are particularly beneficial.¹⁴

Given both the benefits of breast milk for LPIs and ETIs, and the breastfeeding difficulties often experienced by their mothers, it is important to understand what interventions may successfully promote breastfeeding in these populations. Two previous systematic reviews have investigated breastfeeding promotion interventions for LPIs (one also included moderately preterm infants).^15,16 To our knowledge, no review was previously undertaken on available breastfeeding promotion interventions for ETIs.

We conducted a systematic review with the aim of examining the available interventions for LPIs and ETIs that target breastfeeding outcomes. The outcomes included breastfeeding duration and exclusivity, but also breast milk production since many mothers might be dependent (partially or fully) on milk expression in the early postnatal period. In contrast to one of the previous reviews,¹⁶ we only included experimental studies, and specified no language or time restrictions.

The research question that this review addressed follows the PICOS (participants, interventions, comparators, outcomes, study design) model: What is the evidence on the effect of interventions available for LPIs and ETIs on breastfeeding duration/exclusivity or breast milk production?

Methods

Inclusion criteria

Participants

Studies that included LPIs or infants born in a period overlapping 34^0/7 to 36^6/7 gestational weeks by at least 2 weeks (34–35 weeks or 35–36 weeks) were eligible for inclusion. Studies that included ETIs or infants born in a period overlapping 37^0/7 and 38^6/7 gestational weeks by at least 1 week (37 or 38 weeks) were also eligible. If the sample included more than just LPIs or ETIs, data needed to be presented separately for these categories.

Interventions and comparators

Studies investigating any type of intervention or combination of interventions provided with the aim of promoting breastfeeding or breast milk provision, in any setting but starting within 1 month of birth were eligible. Comparators could be other interventions but must also include some type of control, including usual care, placebo, or no treatment.

Outcomes

The outcomes of the interventions included breastfeeding duration, breastfeeding exclusivity, and/or breast milk production (volume) at a time point before 6 months actual age.

Study design

Only randomized controlled trials (RCTs) or quasi-experimental (QE) studies (nonrandomized interventional studies) were eligible for inclusion. Observational studies were excluded due to their inability to establish causation.

Searches

The following databases were searched in February 2022: MEDLINE, CINAHL, Embase, and Google Scholar, without time or language restrictions. The search terms included: breastfeeding, breast milk, premature infants, ETIs, 37–38 weeks, 37–38 gestation, term birth and similar words (Table 1). The full search and screening protocol was registered in PROSPERO (CRD42020187000).

RCTs and non-RCTs (QE studies) were extracted from the search.

Other relevant articles were sought by backward reference searching of the included articles.

Table 1.

Keywords and Map Term to Subject Heading Used in Literature Search and Search Strategy Used

No.	Search strategy	MeSH	Keywords
1	MeSH OR Keywords for Breastfeeding	Breast feeding	Breastfeeding or Breast Feeding or Human Milk or Breast milk or breast milk or lactation
2	MeSH OR Keywords for Late Preterm	Infant, premature	Late preterm infant^* or late preterm newborn^* or Late premature infant^* or late premature newborn^* or near term
3	MeSH OR Keywords for Early Term	–	Early term infant^* or early term newborn^*
4	2 OR 3	(Infant, Premature/or Late preterm infant^* or late preterm newborn^* or Late premature infant^* or late premature newborn^* or near term) OR (Early term infant^* or early term newborn^*)
5	1 AND 4	(Breastfeeding or Breast Feeding or Human Milk or Breast milk or Breast milk or Lactation) AND ([Infant, Premature/or Late preterm infant^* or late preterm newborn^* or Late premature infant^* or late premature newborn^* or near term] OR [Early term infant^* or early term newborn^*])

MeSH, map term to subject heading.

Study selection

The search results were imported to EndNote X9 where duplicates were removed. An initial screening of the titles and abstracts against inclusion criteria was conducted by two reviewers independently (S.D. and K.K.). This was followed by a screening of the full texts of relevant articles. Discrepancies were resolved by discussion and by consulting a third reviewer (M.F.).

Data extraction

The guidelines from the Center for Reviews and Dissemination¹⁷ were followed to generate the data extraction forms. For this review, guidelines of the Preferred Reporting Items of Systematic reviews Meta-Analysis (PRISMA) were followed. Details of methodological quality, study design, sample, and intervention were abstracted. For each outcome, the time point, the numeric results, the statistic used, and the p-value were abstracted.

Quality assessment

Two reviewers (S.D. and K.K.) independently assessed the risk of bias based on the Cochrane group methods for systematic reviews (details provided in Supplementary Data). Discrepancies were resolved by discussion with involvement of a third review author (M.F.) where necessary. The level of risk of bias in each of these domains was presented separately for each study.

Data synthesis

Included studies were too diverse (various interventions, targeting different age groups, outcomes at different time points) for a quantitative synthesis. Therefore, a narrative synthesis was undertaken. Results were classified according to the outcome: breastfeeding duration, breastfeeding exclusivity, or breast milk production. They were further grouped according to the target population (LPIs versus ETIs).

Results

Based on the search strategy, 2,408 records were identified from the four databases. As shown in Figure 1, after duplicates were removed, 1,556 titles were screened after which 372 abstracts were assessed for eligibility. After 63 full texts were screened, nine articles were included for this review (Table 2). Two studies reported outcomes related to “exclusive breastfeeding” only,^18,19 two related to “any breastfeeding” only,^20,21 and the other five related to both “any breastfeeding” and “exclusive breastfeeding.”^22–26 None of the studies reported on breast milk volume. Only one study was an RCT,¹⁸ the rest were QE (nonrandomized). Only one study included infants of 37 weeks' gestation,¹⁸ the rest included LPIs exclusively. The studies included 1,325 infants of which 20 were of 37 weeks' gestation (ETIs) and the rest were late preterm.

FIG. 1.

PRISMA flow diagram.

Table 2.

Data Extraction of Included Studies

Author, year, location	Study design	Population	Intervention & comparators	Outcomes	Time point	Main result^a
McKeever et al (2002),¹⁸ Canada	RCT	101 Term infants	Standard hospital care with early discharge and home support from nurses who were lactation consultants versus standard hospital care and standard length of hospitalization	EBF (feeding by breast only)	At discharge	The intervention had no effect on EBF (OR = 1.4; 95% CI = 0.3–7.2)
McKeever et al (2002),¹⁸ Canada	RCT	37 LPIs (35–37)		Exclusive breast milk feeds (feeding only breast milk by breast or bottle)	5–12 Days postpartum
Maastrup et al (2021),¹⁹ Denmark	QE	110/421 Control LPIs (35^0/7 to 36^6/7)	Standard care (preintervention) versus neonatal nurse training program (with focus on early breast milk expression, skin-to-skin contact, rooming-in, use of test-weighing and minimizing use of pacifiers)	EBF (feeding by breast only)	At discharge	The intervention had 1.3 times the odds of EBF at discharge (95% CI = 0.8–2.3) in infants born at 35–36 weeks (study is not powered to detect subgestational group differences)
Maastrup et al (2021),¹⁹ Denmark	QE	142/494 Intervention LPIs (35^0/7 to 36^6/7)		EBF (feeding by breast only)	At discharge
Jang and Ju (2020),²⁰ Korea	QE	27 Control LPIs (34^0/7 to 36^6/7)	Standard care versus four-session LPI education program (characteristics of LPIs, breastfeeding for LPIs, postdischarge management of LPIs, emotional support)	Breastfeeding (newborn received any breast milk)	At discharge	OR of 3.9 (95% CI = 1.2–12.6) of breastfeeding at 1 month postdischarge in the experimental group
Jang and Ju (2020),²⁰ Korea	QE	26 Intervention LPIs (34^0/7 to 36^6/7)		Breastfeeding (newborn received any breast milk)	1 Month postdischarge
Jang and Ko (2021),²¹ Korea	QE	19 Control LPIs (34^0/7 to 36^6/7)	Control group (four home visits with counseling related to nurturing LPIs) versus breastfeeding coaching program (web-based breastfeeding education program involving four home visits, practical breastfeeding support based on infant's and mother's needs, and encouragement and advice to express milk regularly)	Breastfeeding (feeding by breast, or bottle-feeding pumped milk and/or small amounts of formula once or twice a day)	At discharge	The interaction effect between treatment and time had an OR of 1.2 (95% CI = 0.6–2.5), 1.3 (0.5–3.12), 2.4 (0.6–9.3), and 3.7 (1.0–14.2) for the first, second, third, and fourth week, respectively
					First week
					Second week
		21 Intervention LPIs (34^0/7 to 36^6/7)			Third week
		21 Intervention LPIs (34^0/7 to 36^6/7)			Fourth week
Jang and Hong (2020),²² Korea	QE	20 Control LPIs (34^0/7 to 36^6/7)	Control group (four home visits with counseling related to nurturing LPIs) versus breastfeeding support program (web-based breastfeeding education program involving four home visits, practical breastfeeding support based on infant's and mother's needs, and encouragement and advice to express milk regularly)	EBF (feeding by breast only)	At discharge	OR of 5.2 (95% CI = 1.1–16.7) of EBF overall and 7.1 (95% CI = 1.7–29.9), 12.0 (95% CI = 2.7–52.9), and 15.2 (95% CI = 3.3–69.2) the odds of EBF at 2, 3, and 4 weeks, respectively
					First week
					Second week
		20 Intervention LPIs (34^0/7 to 36^6/7)		Mixed feeding (feeding breast milk by breast or bottle, and supplemented with formula milk)	Third week
		20 Intervention LPIs (34^0/7 to 36^6/7)			Fourth week
Estalella et al (2020),²³ Spain	QE	212 Control LPIs (34^0/7 to 36^6/7)	Standard care (recommendations provided on a sheet, breastfeeding evaluated verbally, visits conducted separately by pediatrician and nurse, phototherapy at NICU) versus intervention (postnatal booklet provided, breastfeeding evaluated by chart, visits conducted together, phototherapy at bedside)	EBF (newborn received breast milk)	At discharge	The intervention group had twice the odds of EBF (OR = 2.1; 95% CI = 1.4–3.2)
Estalella et al (2020),²³ Spain	QE	161 Intervention LPIs (34^0/7 to 36^6/7)		Breastfeeding (newborn received breast milk and human milk substitute)	At discharge
Abouelfettoh et al (2008),²⁴ Egypt	QE	30 Control LPIs (34^0/7 to 36^6/7)	Control group (bottle feeding all oral feeds at the NICU) versus intervention group (cup feeding all oral feeds at the NICU)	EBF (feeding breast milk only)	Weeks 1, 2, 3, 4, 5, and 6 postdischarge	The intervention did not increase exclusive breastfeeding at 1 week postdischarge (OR = 1.8; 95% CI = 0.6–5.0)
Abouelfettoh et al (2008),²⁴ Egypt	QE	30 Intervention LPIs (34^0/7 to 36^6/7)		EBF (feeding breast milk only)	Weeks 1, 2, 3, 4, 5, and 6 postdischarge
Jang (2020),²⁵ Korea	QE	19 Control LPIs (34^0/7 to 36^6/7)	Control group (four home visits with counselling related to nurturing LPIs) versus breastfeeding coaching program (web-based breastfeeding education program involving four home visits, practical breastfeeding support based on infant's and mother's needs, and encouragement and advice to express milk regularly)	Breastfeeding (feeding by breast or bottle-feeding pumped milk)	At discharge	OR for breastfeeding (adjusted for length of hospital stay and newborn disease) was 13.7 (95% CI = 1.2–157.0) and 20.6 (95% CI = 2.0–214.4) at the third and fourth weeks, respectively
					First week
					Second week
		21 Intervention LPIs (34^0/7 to 36^6/7)		Mixed feeding (feeding breast milk by breast or bottle, and supplemented with formula milk)	Third week
		21 Intervention LPIs (34^0/7 to 36^6/7)			Fourth week
Dani et al (2022),²⁶ Italy	QE	190 Control LPIs (35^0/7 to 36^6/7)	Standard care at hospital 1 (LPIs directly admitted to SCU/NICU) versus rooming-in assistance at hospital 2 (followed by admission to SCU/NICU if needed)	EBF (newborn fed only breast milk, including human donor milk and/or expressed breast milk)	At discharge	The intervention reduced the odds of EBF (OR = 0.2; 95% CI = 0.1–0.4) in infants born at hospital 2 (rooming-in), but the odds were nil when adjusting for differences in infants' characteristics
Dani et al (2022),²⁶ Italy	QE	240 Intervention LPIs (35^0/7 to 36^6/7)		Mixed (any breastfeeding mixed with formula feeding)	At discharge

OR and 95% CI were calculated if not available in articles.

CI, confidence interval; EBF, exclusive breastfeeding; LPI, late preterm infant; NICU, neonatal intensive care unit; OD, odds ratio; QE, quasi-experimental; RCT, randomized controlled trial; SCU, special care unit.

Interventions

All the interventions in the included studies involved breastfeeding support delivered by health professionals. McKeever et al¹⁸ investigated early hospital discharge coupled with breastfeeding support delivered at home by lactation consultants. Maastrup et al¹⁹ investigated a neonatal nurse training program that focused on improving certain hospital practices such as early breast milk expression, skin-to-skin contact, and rooming-in. Estalella et al²³ targeted hospital practices that could improve breastfeeding such as bedside phototherapy and more detailed evaluation of breastfeeding. Similarly, Dani et al²⁶ explored rooming-in assistance in comparison to direct admission to neonatal intensive care unit (NICU). Abouelfettoh et al²⁴ studied the influence of cup feeding for LPIs admitted to a NICU, compared with bottle feeding. The remaining studies were education-based interventions designed specifically for LPIs. The first involved a four-session/intervention education program, which covered topics such as the characteristics of LPIs, breastfeeding the LPIs, and postdischarge management delivered face-to-face at the hospital before discharge.²⁰

Mothers were also followed up on a weekly basis, for a month postdischarge, through telephone to offer them emotional support and to allow them to ask questions. The other three studies covered education on late preterm topics over four home visits but also offered individualized practical breastfeeding support and advice to express breast milk regularly.^21,22,25

Exclusive breastfeeding

McKeever et al,¹⁸ who conducted the only RCT included, showed that home lactation support compared with hospital support did not improve breastfeeding exclusivity at 5–12 days postpartum in LPIs and EPIs (37 weeks). Dani et al²⁶ found that rooming-in assistance rather than direct admission to NICU resulted in lower exclusive breastfeeding rates. Conversely, one QE study reported twice the odds of exclusive breastfeeding at discharge in LPIs whose mothers received an intervention designed to promote breastfeeding in this population.²³ Two other studies showed that a breastfeeding support intervention delivered over 4 weeks increased exclusive breastfeeding at the second, third, and fourth week postdelivery (odds ratio [OR] = 7.1, 95% confidence interval [CI] = 1.7–29.9; OR = 12.0, 95% CI = 2.7–52.9; and OR = 15.2, 95% CI = 3.3–69.2)²² or at 4 weeks postdelivery (OR = 4.0, 95% CI = 1.2–12.6),²⁵ compared with the control group.

Maastrup et al¹⁹ also found that the intervention (nurse training program) resulted in higher odds of exclusive breastfeeding at discharge (OR = 1.3; 95% CI = 0.8–2.3) in infants born at 35–36 weeks. Lastly, Abouelfettoh et al²⁴ showed that the proportion of feedings that were breast milk (direct or expressed) at 1 week postdischarge was significantly higher in the cup feeding (80.2%; 95% CI = 70.6–89.8) group than in the bottle feeding group (64.4%; 95% CI = 53.4–75.4), although there were no significant differences in the proportion who were exclusive breastfeeding between the two groups.

Any breastfeeding

Estalella et al²³ also investigated breastfeeding at discharge as an outcome. Breastfeeding rate was lower in the intervention group (25.9%) compared with the control group (37.8%), a reflection of the significantly higher exclusive breastfeeding rate in the intervention group rather than more formula feeding. Jang and Hong²² showed that there was a small increase in breastfeeding rate in the experimental group compared with the control group, but the lack of the large increase is again a reflection of the noticeable increase in exclusive breastfeeding. Similarly, another QE study revealed that the intervention group who received a four-session education program had 3.9 times the odds (95% CI = 1.2–12.6) of breastfeeding at 1 month postdischarge compared with the control group.²⁰ Conversely, three studies^21,25,26 did not find any significant differences in breastfeeding rates between the intervention and control groups.

Breast milk volume

None of the included studies reported breast milk volume as an outcome.

Quality of the studies

The quality of each study was assessed according to six domains. Only one RCT was included in this review, in which the method of randomization sequence generation was not mentioned.¹⁸ While the researchers collecting the data were originally blinded to the group status, the participants were not blinded to the intervention due to its nature (early discharge, home support) and later revealed their status during the interviews.

The other eight studies were of QE design, where the researchers and the participants were aware of the group allocation. Six of the studies collected data at different periods for the control and experimental groups (∼1-year gap),^{19–21,23,25,26} one of which was also carried out at different hospitals for each group,²⁶ which might have introduced a bias in the characteristics of the sample. However, Estalella et al²³ and Maastrup et al¹⁹ reported no significant differences in the baseline characteristics between the control and experimental group (there were fewer extremely preterm infants in the Maastrup et al¹⁹ but this is not relevant to this review). Similarly, Jang and Hong,²² Jang and Ko,²¹ and Jang and Ju²⁰ reported no significant differences at baseline, but there were some notable differences in infant characteristics, which might have contributed to the findings. On the other hand, the remaining three QE studies found significant differences in baseline characteristics.^24–26 Both Dani et al²⁶ and Jang²⁵ provided an adjustment for the differences in infant characteristics, whereas Abouelfettoh et al²⁴ did not account for the significantly higher birthweight in the cup feeding group.

Other study quality concerns are evident. For example, there was a high number of attrition in Abouelfettoh's study,²⁴ where only 13/30 participants in the intervention group but 25/30 in the control group maintained participation by 6 weeks. The differences in characteristics between participants included and lost to follow-up were not investigated, and the results of the first week only were reported despite the intention to investigate breastfeeding practices at 1–6 weeks postdischarge. In three other studies, it is unclear how the decision to allocate infants to each group was made, which might produce a high risk of bias.^21,22,25 Additionally, it is uncertain how the intervention investigated by Jang²⁵ and Jang and Ko²¹ was modified from the one developed by Jang and Hong.²² Lastly, it also unclear why the breastfeeding/mixed-feeding rates were reported differently in Jang²⁵ and Jang and Ko²¹ despite involving the same sample Table 3.

Table 3.

Data Quality Assessment of Included Studies Based on the Cochrane Group Methods for Systematic Reviews

Article	Sequence	Treatment allocation concealment	Blinding	Completeness of outcome data	Selective outcome reporting	Other sources of bias
McKeever et al¹⁸	Not mentioned	Interviewers were originally blinded to group status, but mothers later revealed their status. Baseline measurements were taken before randomization.	Did not mention blinding during analysis but personnel assessing outcomes were not adequately blinded throughout.	No differences between those who declined to participate and those who completed. Exclusions discussed but did not discuss if there were differences in those who were lost to follow-up compared with included.	No	Restricted inclusion criteria make the results less generalizable to LPIs:
						Not experienced c-sections or postpartum complications
						Babies at least 35 weeks gestation, no morbidities, including hyperbilirubinemia
						Included sample are mainly White, middle class
Maastrup et al¹⁹	Not applicable	No concealment. Control group data were collected between October 2016 and July 2017 (preintervention) and intervention data between February 2018 and December 2018 (postintervention).	Did not mention blinding during analysis but staff assessing outcomes were not blinded due to study design.	High percentage of missing data. Consent was obtained for 72% and 65% of eligible participants and information on breastfeeding outcome was collected for 53% and 48% of participants in the control and intervention groups, respectively.	No	QE design that was conducted in different periods.
Jang and Ju²⁰	Not applicable	No concealment. Control group data were collected between 2014 and 2015 and intervention data between 2015 and 2016.	Did not mention blinding during analysis but researchers assessing outcomes were not blinded due to study design.	Only three participants were lost to follow-up in the control group, unclear if there were differences in those who were not followed up with.	No	QE design that was conducted in different periods (1 year gap). Control group had a higher proportion of 34 weekers (48.1% versus 23.1%), whereas the experimental group had a higher proportion of 36 weeks' gestation (53.8% versus 29.6%).
Jang and Ko²¹	Not applicable	No concealment. Control group data were collected from June to October 2017 and intervention data from November 2017 to May 2018. Did not mention how the mothers and infants were allocated to each group.	Did not mention blinding during analysis but researchers assessing outcomes were not blinded due to study design.	Data seem to be complete. Not clear if any participants were excluded.	No	QE design that was conducted in different periods. The rate of ventilator used in the experimental group was higher than in the control, whereas the mean 1-minute and 5-minute Apgar scores were lower in the experimental group.
Jang and Hong²²	Not applicable	No concealment. Did not mention how the mothers and infants were allocated to each group, or why it was not a randomized controlled trial.	Did not mention blinding during analysis but researchers assessing outcomes were not blinded due to study design.	Data seem to be complete. Not clear if any participants were excluded or if any data were not collected.	No	Control group had a higher proportion of infants with feeding intolerance (25% versus 10%), or an illness/complication (65% versus 50%), and who initiated feeding on a day after the day of birth (25% versus 10%).
Estalella et al²³	Not applicable	No concealment. Control group data were collected between 2012 and 2013 and intervention data between 2014 and 2015 (after implementation of intervention).	Did not mention blinding during analysis but personnel assessing outcomes were not blinded due to study design.	Only six participants were excluded for refusing to participate. No attrition but data collected from medical records so missing values are possible. It was not clear which values might be missing.	No	QE design that was conducted in different periods. However, baseline pregnancy and delivery characteristics were not significantly different.
Abouelfettoh et al²⁴	Not applicable	No concealment. Did not mention how the mothers and infants were allocated to each group.	Did not mention blinding during analysis but researchers assessing outcomes were not blinded due to study design.	High risk of bias as only the results from the first week were reported and the plan was to investigate breastfeeding practices at 1–6 weeks postdischarge. High number of attrition but the differences in characteristics of participants who remained and lost to follow-up were not investigated.	Yes	Birthweight was significantly higher in the cup feeding group compared with the bottle feeding group at baseline, which was not accounted for. Unclear if mothers were expressing breast milk, and if there were differences in milk expression between the two groups.
Jang²⁵	Not applicable	No concealment. Control group data were collected from June to October 2017 and intervention data from November 2017 to May 2018. Did not mention how the mothers and infants were allocated to each group.	Did not mention blinding during analysis but researchers assessing outcomes were not blinded due to study design.	Data seem to be complete. Not clear if any participants were excluded.	No	QE design that was conducted in different periods. The length of hospital stay was significantly longer in the experimental group, and more LPIs in the experimental group had diseases.
Dani et al²⁶	Not applicable	No concealment. Control group data were collected from January 2018 to December 2020 at hospital 1 and intervention data from July 2019 to December 2020 at hospital 2.	Did not mention blinding during analysis but researchers assessing outcomes were not blinded due to study design.	Data seem to be complete.	No	Study was conducted in different periods and settings. Gestational age was lower and the need for respiratory support was higher in hospital 2 (intervention), whereas in hospital 1 (control) there was higher need of peripheral vascular catheters and higher incidence of gestational diabetes.

LPI, late preterm infant; QE, quasi-experimental.

Discussion

The findings from this review are inconclusive but might indicate that breastfeeding support interventions delivered by health professionals in the early postnatal period (birth-4 weeks) can be beneficial at improving exclusive breastfeeding duration in LPIs.

Professional breastfeeding support and hospital practices

McKeever et al¹⁸ found that in-home lactation support did not increase exclusive breastfeeding. While the authors aimed to investigate early discharge of infants of 35–37 weeks' gestation with the additional in-home lactation support, there were no significant differences in the length of hospital stay between groups (45 versus 48 hours). This was because they did not meet the early discharge criteria and because of the already existing practice of early discharge for LPIs. Several studies have shown that LPIs who were discharged early were more likely to be readmitted.^27–29 Similarly, a recent study showed that early discharge (<48 hours) of healthy LPIs was not associated with cost savings, probably due to the higher risk of rehospitalization after early discharge.³⁰ In all these studies, the most common reason for rehospitalization was jaundice. Therefore, it is possible that early discharge for LPIs does not allow for enough time to support breastfeeding practices and establish adequate breast milk supply. Nevertheless, other similar studies have also shown no significant differences in exclusive breastfeeding rates between the two groups.^31,32

Estalella et al²³ delivered an intervention where hospital practices were changed with the aim of promoting parents' education and involvement, avoiding separation from the infant (when phototherapy is provided), and creating a multidisciplinary approach to support breastfeeding. The results showed that the intervention group had higher exclusive breastfeeding rate at discharge. This could be partially explained by the significantly higher proportion of mothers in the intervention group expressing breast milk after some or all of the feeds. The results should be interpreted with caution due to the QE design and the 1-year gap in data collection, however, no significant differences were found in the baseline characteristics and the sample size was relatively large with only a few exclusions. Additionally, Maastrup et al¹⁹ investigated another intervention that aimed at improving hospital practices supportive of breastfeeding such as skin-to-skin contact, rooming-in, and early breast milk expression. The findings also revealed an increase in exclusive breastfeeding rates in the intervention group. As with Estallela et al,²³ the study involved a large sample with low risk of bias.

Jang and Ju²⁰ designed a within-hospital breastfeeding education intervention for mothers of LPIs admitted to the NICU, but this was followed by weekly phone call checkups (for 1 month) after discharge. The results showed that at 1-month postdischarge breastfeeding rates and parenting confidence were significantly higher in the experimental group compared with the control group. However, there are several limitations to the study, such as the QE design, the 1-year gap in data collection, and higher proportion of infants of 34 weeks' gestation and lower proportion of infants of 36 weeks' gestation in the control group versus intervention group at baseline. Other observational studies have also analyzed the association between hospital practices and support on breastfeeding in LPIs. A study, including 579 LPIs from the UK 2010 Infant Feeding Survey, showed that mothers who reported that they did not receive enough support with breastfeeding at the hospital were less likely to be breastfeeding at 10 days.³³ Another showed that high levels of professional support at the hospital was associated with an increased likelihood of any breastfeeding in LPIs, ETIs, and term infants.¹⁰

Jang and Hong,²² Jang,²⁵ and Jang and Ko²¹ evaluated a breastfeeding education/coaching/support program delivered at discharge and over four home visits after (once per week). In the initial period, the intervention encouraged mothers to express breast milk to increase production, then later provided practical instruction on latching and positioning. The results suggested an improvement in exclusive breastfeeding rates, however, there were some differences in infant characteristics at baseline as well as other study quality concerns. Similarly, an RCT has shown that a program involving 11 one-hour sessions pre- and postdischarge, aiming to educate parents on their infant's characteristics and improve responsiveness to cues and interaction, increased breastfeeding duration in mothers of moderately preterm infants and LPIs at 6, 9, and 12 months postdischarge, although the effects were only significant at 9 months.³⁴

Another RCT, which included mainly moderately preterm infants and LPIs (n = 414; 84%) from six NICUs in Sweden found no significant differences in exclusive breastfeeding at 8 weeks between the proactive telephone support group (received daily telephone calls from day 1 to day 14 postdischarge from a member of the breastfeeding support team) and the reactive telephone support group (given the option to call a member from the team if they face any breastfeeding problems).³⁵ However, mothers in the proactive telephone support group reported less parental stress at 8 weeks postdischarge. This might indicate the importance of face-to-face support for practical breastfeeding difficulties, whereas telephone support might be beneficial for emotional support.

Dani et al²⁶ compared direct admission to the NICU in one hospital with providing rooming-in assistance and only admitting to NICU if necessary, in another hospital. The odds of exclusive breastfeeding were lower in the hospital that provided rooming-in assistance (OR = 0.17; 95% CI = 0.07–0.4). This could be explained by the differences in infant characteristics between the two hospitals. It might also be possible that infants admitted to the NICU are recognized to be more vulnerable and thus their parents might be provided with more support to breastfeed and with more encouragement to express breast milk.

Early breast milk expression and supplementation

The abovementioned studies provide further evidence that breast milk expression might be a crucial factor that increases the likelihood of exclusive/any breastfeeding in LPIs.^19–22,25 A prospective survey in Denmark, including 1,488 preterm infants with a gestational age of 24–36 weeks, 483 of which were 35–36 weeks, found that delayed initiation of breast milk expression beyond 6 hours postdelivery had a dose–response association with failure to exclusively breastfeed (directly at the breast) at discharge.³⁶ Multiple factors in the early postnatal period after a late preterm/early term birth might necessitate expressing breast milk. For example, LPIs often have lower stamina and fewer awake periods leading to reduced effort to stimulate the breast, which might lead to decreased breast milk production and ejection. As a result, LPIs are commonly supplemented with infant formula, which might further interfere with establishment of breast milk. Therefore, breast milk expression could be beneficial in maintaining or increasing breast milk supply while simultaneously providing nutrients to less mature infants who might be unable to feed effectively.

A comparative study that investigated the influence of formula supplementation in the hospital on breastfeeding rates in LPIs found that 87% of infants who were exclusively breastfed from birth were exclusively breastfed at discharge compared with 24% who were supplemented with formula regularly from birth.³⁷ The study also showed that 65% of mothers whose infants were prescribed breast milk substitutes on a regular basis never used a breast pump, while the rest took an average of 42 hours before using one. Therefore, the low exclusive breastfeeding rates may be partially explained by the inadequate milk expression to establish breastfeeding, and also due to formula volume exceeding the amount of milk the infant would receive at the breast, although this was not investigated in this study.

The method by which supplementation, whether breast milk or formula milk, is delivered to LPIs has also been studied. Abouelfettoh et al²⁴ showed that infants fed by cup had a higher proportion of feedings that were breast milk at 1 week postdischarge compared with infants fed by bottle. Likewise, in another study, including LPIs that was defined in the study as 32–35 weeks gestation, exclusive breastfeeding was significantly higher at discharge, 3, and 6 months, and breastfeeding was significantly higher at discharge and 6 months in the cup feeding group.³⁸ However, both studies had several methodological limitations, including the lack of information on breast milk expression practices and the high level of attrition. Moreover, the analysis was not intention to treat,³⁸ and 85 participants were excluded for noncompliance, which introduced bias to the study results.

Kangaroo mother care

Mörelius³⁹ and Hake-Brooks and Anderson⁴⁰ conducted two RCTs that studied kangaroo mother care, but which were not included in this review because the data were not analyzed separately for LPIs versus more preterm infants. Mörelius³⁹ found that in infants (32–36 gestational age) admitted to the NICU, breastfeeding rates were higher in the continuous skin-to-skin group compared with infants in the standard care group, at discharge (100% versus 84%) and at 1 month (94% versus 74%) and 4 months (77% versus 53%) corrected age, although the results were not statistically significant. In the other study, in infants of 32–36 gestational age, kangaroo care significantly increased breastfeeding and breastfeeding exclusivity at 6 months, compared with control.⁴⁰ The discrepancy in the significance might be due to the difference in the duration skin-to-skin was practiced. For example, Hake-Brooks and Anderson⁴⁰ reported that participants in the intervention group practiced skin-to-skin for an average of 4.47 hours per day, whereas Mörelius³⁹ reported 7 hours of skin-to-skin per day on average in the control group.

Similar evidence was reported in a few observational studies. For example, a cohort study conducted in the United Kingdom also found that kangaroo mother care increased breastfeeding rates at discharge and reduced average length of hospital stay in LPIs, small-for-gestational-age infants, and infants of diabetic mothers.⁴¹ Likewise, in a QE study involving a large sample of LPIs and their mothers, mothers who chose to provide kangaroo mother care were twice as likely to exclusively breastfeed at discharge and at 42 days postdelivery compared with mothers who opted not to.⁴²

Alternative and relaxation therapies

The use of herbal therapies and meditation audio for mothers of LPIs and ETIs with perceived insufficient milk supply was investigated.⁴³ The study did not report significant differences in breast milk volume or breastfeeding status, however, this is probably due to the small sample size (n = 11), which was underpowered to detect differences. Additionally, the study did not include a control comparator, which is why it was not included in this review. Nevertheless, the study showed that these two complementary and alternative therapies are safe and acceptable in this population. Other relaxation interventions such as music and meditation were shown to be beneficial at increasing breast milk volume in mothers of preterm infants^44,45 and in full-term infants.⁴⁶ However, their efficacy was not studied in LPIs or ETIs specifically, and further investigation is warranted.

Conclusion

In summary, professional breastfeeding support and education programs tailored to LPIs might improve breastfeeding and/or exclusive breastfeeding rates. Other interventions, such as early breast milk expression, kangaroo mother care, and relaxation therapies, are promising and warrant more investigation in this population using robust study design. Overall, there are limited experimental studies that exclusively include LPIs or ETIs, or present the data for these groups separately. Additionally, since the experiments are mainly conducted in single hospitals, which makes randomizing infants and avoiding cross-contamination difficult, 8/9 studies in this review are QE. This introduces confounding bias and limits their ability to conclude a causal association, which is especially true in 6/8 QE studies in this review that were found to have differences at baseline. Therefore, RCTs that target LPIs and ETIs are needed to improve breastfeeding and health outcomes for these infants. More interventions that are delivered postdischarge are also needed, as many of these infants do not have long hospital stays and thus their mothers might require support at home.

Footnotes

Authors' Contributions

S.D. and M.F. proposed the idea and concept of this review. S.D. conducted the search and screened the titles. S.D. and K.K. screened the abstracts and full-text articles and assessed the quality of the studies. S.D. drafted the article, and J.W., M.F., and K.K. edited the article and contributed critical intellectual input. All authors approved the final article for submission.

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

Supplementary Material

References

Blencowe

, Cousens

, Oestergaard

, et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: A systematic analysis and implications. Lancet, 2012; 379(9832):2162–2172; doi: 10.1016/S0140-6736(12)60820-4

Crump

An overview of adult health outcomes after preterm birth. Early Hum Dev, 2020; 150:105187; doi: 10.1016/j.earlhumdev.2020.105187

Committee on Obstetric Practice. ACOG committee opinion no. 404 April 2008. Late-preterm infants. Obstet Gynecol, 2008; 111(4):1029–1032; doi: 10.1097/AOG.0b013e31817327d0

Davidoff

, Dias

, Damus

, et al. Changes in the gestational age distribution among US singleton births: Impact on rates of late preterm birth, 1992 to 2002. Semin Perinatol, 2006; 30(1):8–15; doi: 10.1053/j.semperi.2006.01.009

Wang

, Dorer

, Fleming

, et al. Clinical outcomes of near-term infants. Pediatrics, 2004; 114(2):372–376; doi: 10.1542/peds.114.2.372

Richards

, Kramar

, Deb-Rinker

, et al. Temporal trends in late preterm and early term birth rates in 6 high-income countries in North America and Europe and association with clinician-initiated obstetric interventions. JAMA, 2016; 316(4):410–419; doi: 10.1001/jama.2016.9635

Engle

WA.

Morbidity and mortality in late preterm and early term newborns: A continuum. Clin Perinatol, 2011; 38(3):493–516; doi: 10.1016/j.clp.2011.06.009

Brown

, Speechley

, Macnab

, et al. Neonatal morbidity associated with late preterm and early term birth: The roles of gestational age and biological determinants of preterm birth. Int J Epidemiol, 2014; 43(3):802–814; doi: 10.1093/ije/dyt251

Hackman

, Alligood-Percoco

, Martin

, et al. Reduced breastfeeding rates in firstborn late preterm and early term infants. Breastfeed Med, 2016; 11(3):119–125; doi: 10.1089/bfm.2015.0122

10.

Goyal

, Attanasio

, Kozhimannil

. Hospital care and early breastfeeding outcomes among late preterm, early-term, and term infants. Birth, 2014; 41(4):330–338; doi: 10.1111/birt.12135

11.

Boyle

, Johnson

, Manktelow

, et al. Neonatal outcomes and delivery of care for infants born late preterm or moderately preterm: A prospective population-based study. Arch Dis Child Fetal Neonatal Ed, 2015; 100(6):F479–F485; doi: 10.1136/archdischild-2014-307347

12.

Kair

, Colaizy

. Breastfeeding continuation among late preterm infants: Barriers, facilitators, and any association with NICU admission?. Hosp Pediatr, 2016; 6(5):261–268; doi: 10.1542/hpeds.2015-0172

13.

Crippa

, Colombo

, Morniroli

, et al. Do a few weeks matter? Late preterm infants and breastfeeding issues. Nutrients, 2019; 11(2):312; doi: 10.3390/nu11020312

14.

Andreas

, Kampmann

, Le-Doare

. Human breast milk: A review on its composition and bioactivity. Early Hum Dev, 2015; 91(11):629–635; doi: 10.1016/j.earlhumdev.2015.08.013

15.

Carpay

, Kakaroukas

, Embleton

, et al. Barriers and facilitators to breastfeeding in moderate and late preterm infants: A systematic review. Breastfeed Med, 2021; 16(5):370–384; doi: 10.1089/bfm.2020.0379

16.

Cartwright

, Atz

, Newman

, et al. Integrative review of interventions to promote breastfeeding in the late preterm infant. J Obstet Gynecol Neonatal Nurs, 2017; 46(3):347–356; doi: 10.1016/j.jogn.2017.01.006

17.

Akers

, Aguiar-Ibáñez

, Baba-Akbari

, (eds). Systematic Reviews: CRD's Guidance for Undertaking Reviews in Health Care. Centre for Reviews and Dissemination, University of York: United Kingdom; 2009.

18.

McKeever

, Steven

, Miller

, et al. Home versus hospital breastfeeding support for newborns: A randomized controlled trial. Birth, 2002; 29(4):258–265; doi: 10.1046/j.1523-536x.2002.00200.x

19.

Maastrup

, Rom

, Walloee

, et al. Improved exclusive breastfeeding rates in preterm infants after a neonatal nurse training program focusing on six breastfeeding-supportive clinical practices. PLoS One, 2021; 16(2):e0245273; doi: 10.1371/journal.pone.0245273

20.

Jang

, Ju

. Effects of an infant care education program for mothers of late-preterm infants on parenting confidence, breastfeeding rates, and infants' growth and readmission rates. Child Health Nur Res, 2020; 26(1):11–22; doi: 10.4094/chnr.2020.26.1.11

21.

Jang

, Ko

Effects of a breastfeeding coaching program on growth and neonatal jaundice in late preterm infants in South Korea. Child Health Nurs Res, 2021; 27(4):377–384; doi: 10.4094/chnr.2021.27.4.377

22.

Jang

, Hong

. Effects of a breastfeeding support program on the prevalence of exclusive breastfeeding and growth in late preterm infants. Child Health Nur Res, 2020; 26(1):90–97; doi: 10.4094/chnr.2020.26.1.90

23.

Estalella

, Millan

, Trinacado

, et al. Evaluation of an intervention supporting breastfeeding among late-preterm infants during in-hospital stay. Women Birth, 2020; 33(1):e33–e38; doi: 10.1016/j.wombi.2018.11.003

24.

Abouelfettoh

, Dowling

, Dabash

, et al. Cup versus bottle feeding for hospitalized late preterm infants in Egypt: A quasi-experimental study. Int Breastfeed J, 2008; 3(1):1–11; doi: 10.1186/1746-4358-3-27

25.

Jang

GJ.

Influence of a breastfeeding coaching program on the breastfeeding rates and neonatal morbidity in late preterm infants. Child Health Nurs Res, 2020; 26(3):376–384; doi: 10.4094/chnr.2020.26.3.376

26.

Dani

, Ciarcia

, Miselli

, et al. The management of late preterm infants: Effects of rooming-in assistance versus direct admission to neonatal care units. Eur J Pediatr, 2022; 181(4):1643–1649; doi: 10.1007/s00431-021-04337-z

27.

Tomashek

, Shapiro-Mendoza

, Weiss

, et al. Early discharge among late preterm and term newborns and risk of neonatal morbidity. Semin Perinatol, 2006; 30(2):61–68; doi: 10.1053/j.semperi.2006.02.003

28.

Escobar

, Greene

, Hulac

, et al. Rehospitalisation after birth hospitalisation: Patterns among infants of all gestations. Arch Dis Child, 2005; 90(2):125–131; doi: 10.1136/adc.2003.039974

29.

Escobar

, Joffe

, Gardner

, et al. Rehospitalization in the first two weeks after discharge from the neonatal intensive care unit. Pediatrics, 1999; 104(1):e2; doi: 10.1542/peds.104.1.e2

30.

Isayama

, O’ Reilly

, Beyene

, et al. Hospital care cost and resource use of early discharge of healthy late preterm and term singletons: A population-based cohort study and cost analysis. J Pediatr, 2020; 226(7):96–105; doi: 10.1016/j.jpeds.2020.06.060

31.

Holm

, Clemensen

, Brodsgaard

, et al. Growth and breastfeeding of preterm infants receiving neonatal tele-homecare compared to hospital-based care. J Neonatal Perinatal Med, 2019; 12(3):277–284; doi: 10.3233/NPM-18143

32.

Gunn

, Thompson

, Jackson

, et al. Does early hospital discharge with home support of families with preterm infants affect breastfeeding success? A randomized trial. Acta Paediatr, 2000; 89(11):1358–1363; doi: 10.1080/080352500300002570

33.

Rayfield

, Oakley

, Quigley

. Association between breastfeeding support and breastfeeding rates in the UK: A comparison of late preterm and term infants. BMJ Open, 2015; 5(11):e009144; doi: 10.1136/bmjopen-2015-009144

34.

Ravn

, Smith

, Smeby

, et al. Effects of early mother-infant intervention on outcomes in mothers and moderately and late preterm infants at age 1 year: A randomized controlled trial. Infant Behav Dev, 2012; 35(1):36–47; doi: 10.1016/j.infbeh.2011.09.006

35.

Ericson

, Eriksson

, Hellstrom-Westas

, et al. Proactive telephone support provided to breastfeeding mothers of preterm infants after discharge: A randomised controlled trial. Acta Paediatr, 2018; 107(5):791–798; doi: 10.1111/apa.14257

36.

Maastrup

, Hansen

, Kronborg

, et al. Factors associated with exclusive breastfeeding of preterm infants: Results from a prospective national cohort study. PLoS One, 2014; 9(2):e89077; doi: 10.1371/journal.pone.0089077

37.

Mattsson

, Funkquist

, Wickstrom

, et al. Healthy late preterm infants and supplementary artificial milk feeds: Effects on breast feeding and associated clinical parameters. Midwifery, 2015; 31(4):426–431; doi: 10.1016/j.midw.2014.12.004

38.

Yilmaz

, Caylan

, Karacan

, et al. Effect of cup feeding and bottle feeding on breastfeeding in late preterm infants: A randomized controlled study. J Hum Lact, 2014; 30(2):174–179; doi: 10.1177/0890334413517940

39.

Mörelius

, Ortenstrand

, Theodorsson

, et al. A randomised trial of continuous skin-to-skin contact after preterm birth and the effects on salivary cortisol, parental stress, depression, and breastfeeding. Early Hum Dev, 2015; 91(1):63–70; doi: 10.1016/j.earlhumdev.2014.12.005

40.

Hake-Brooks

, Anderson

. Kangaroo care and breastfeeding of mother-preterm infant dyads 0–18 months: A randomized, controlled trial. Neonatal Netw, 2008; 27(3):151–159; doi: 10.1891/0730-0832.27.3.151

41.

Gregson

, Blacker

Kangaroo care in pre-term or low birth weight babies in a postnatal ward. Br J Midwifery, 2011; 19(9):568–577; doi: 10.12968/bjom.2011.19.9.568

42.

Zhang

, Duan

, Zhao

, et al. Intermittent kangaroo mother care and the practice of breastfeeding late preterm infants: Results from four hospitals in different provinces of China. Int Breastfeed J, 2020; 15(1):1–9; doi: 10.1186/s13006-020-00309-5

43.

Demirci

, Bare

, Cohen

, et al. Feasibility and acceptability of two complementary and alternative therapies for perceived insufficient milk in mothers of late preterm and early term infants. Altern Complement Ther, 2016; 22(5):196–203; doi: 10.1089/act.2016.29072.jrd

44.

Keith

, Weaver

, Vogel

. The effect of music-based listening interventions on the volume, fat content, and caloric content of breast milk-produced by mothers of premature and critically ill infants. Adv Neonatal Care, 2012; 12(2):112–119; doi: 10.1097/ANC.0b013e31824d9842

45.

Feher

, Berger

, Johnson

, et al. Increasing breast milk production for premature infants with a relaxation/imagery audiotape. Pediatrics, 1989; 83(1):57–60; doi: 10.1542/peds.83.1.57

46.

Mohd Shukri

, Wells

, Eaton

, et al. Randomized controlled trial investigating the effects of a breastfeeding relaxation intervention on maternal psychological state, breast milk outcomes, and infant behavior and growth. Am J Clin Nutr, 2019; 110(1):121–130; doi: 10.1093/ajcn/nqz033

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.50 MB

0.00 MB