Promotion of early breast milk expression among mothers of preterm infants in the neonatal ICU in an obstetrics and gynaecology hospital: a best practice implementation project : JBI Evidence Implementation

Secondary Logo

Journal Logo


Promotion of early breast milk expression among mothers of preterm infants in the neonatal ICU in an obstetrics and gynaecology hospital: a best practice implementation project

Huang, Xiaoyan PhD1,2; Zhang, Junping BA3,2; Zhou, Feifei BSc3,2; Yang, Yun Diploma3,2; Lizarondo, Lucylynn PhD4; McArthur, Alexa MPHC, MClinSc4

Author Information
JBI Evidence Implementation 18(3):p 278-287, September 2020. | DOI: 10.1097/XEB.0000000000000223


What is known about the topic?

  • The benefits of expressed breast milk for preterm infants and mothers have been identified.
  • Early and frequent expression of breast milk for mothers of premature infants was important.
  • Healthcare professional-led education could improve the rates of breastfeeding initiation.

What does this article add?

  • The role of a nurse specialist in breastfeeding was highlighted.
  • Involvement of multidisciplinary staff was important.
  • Utilization of online networking software was valuable.


Breast milk has been identified as the best source of nutrition for neonates and infants.1,2 Although breast milk can sometimes provide inadequate nutrition if provided alone, it is still recommended for preterm infants.1,2 Preterm infants have a weaker immune system, including fewer immune cells and a less effective inflammatory response, in comparison with full-term infants.1 Breast milk, containing a variety of immune elements, including antimicrobial proteins, maternal leukocytes, immunoglobulins, cytokines and chemokines, oligosaccharides, gangliosides, nucleotides and long-chain polyunsaturated fatty acids, can provide preterm infants with various protective immunologic benefits.1 Preterm infants who are fed with breast milk tend to experience lower incidence of adverse outcomes, such as necrotizing enterocolitis (NEC), late-onset sepsis and retinopathy of prematurity.2 In addition to improved health outcomes, feeding preterm infants with breast milk has economic benefits secondary to the reduction of healthcare costs associated with NEC and other diseases.3

Due to the vulnerability of preterm infants, especially low and very low birth weight infants, they are likely to be admitted into neonatal ICU (NICU) after birth. This means that many preterm infants are separated from their mothers, which prohibits direct breastfeeding.4 In China specifically, where healthcare resources are relatively limited due to the large population, mothers are usually not allowed to enter the NICU to feed their babies. In addition, the immaturity of preterm infants’ sucking skills makes direct breastfeeding challenging.5 In this circumstance, breast milk expression is helpful. Breast milk expression, a process to remove breast milk from the breast without an infant's direct sucking at the mother's nipple, has been identified as an important element of breastfeeding.6 Different methods of expression could be used, including hand expression, breast pumping using either an electrical pump or a hand pump, and hands-on pumping combining hand expression and breast pumping.6 Regardless of the means of expression, early initiation of breast milk expression in the first 6 h following delivery, or sooner, has been considered to be effective in the promotion of breastfeeding and supply of breast milk to infants who are unable to be directly breastfed.7 Expression of breast milk also helps mothers of preterm infants feel less guilty about their premature birth, and can be regarded as a means to connect with their infants.8 Mothers believed that their milk can provide their infants with healing function, reducing the effects of complications and keeping their infants healthy.8

Although the benefits of expressed breast milk for preterm infants and mothers have been identified, the rate of breast milk provision remains low. A cross-sectional analysis including 6997 infants in NICU in the USA indicated that the majority of very low birth weight infants (52%) were on formula milk only upon discharge, and only 6% of infants were discharged on exclusive human milk.9 A national representative survey in China involving 14 458 children younger than 2 years revealed that the weighted exclusive breastfeeding rate was only 18.6% for children under 6 months.10

While many mothers realized that expression of breast milk was necessary to meet their infants’ feeding needs, some of them reported that pumping milk was expensive, time-consuming and unpleasant compared with direct breastfeeding.11 These perceptions may hinder the practice of breast milk expression. In addition, mothers who lacked knowledge about the benefits of breast milk and had an unfavourable attitude towards breastfeeding were more likely to stop breast milk expression.4 Another barrier to early expression was the lack of nursing time and nursing personnel to assist mothers with breast milk expression.12 Researchers indicated that education for family members on how to help with breast milk expression may be beneficial.12 A Cochrane systematic review indicated that healthcare professional-led education could improve the rates of breastfeeding initiation.13 Education to mothers and training to healthcare professionals have also been effective in promoting the expression of breast milk and increasing the breast milk feeding rate for preterm infants.13

The Obstetrics and Gynecology Hospital of Fudan University, also known as Shanghai Red House Obstetrics & Gynecology Hospital, is the first women and children's hospital in Shanghai, China. It has been accredited by the Baby Friendly Hospital Initiate since 1992. There are six obstetric wards with a total of 172 beds, and one NICU with around 15–25 beds in the Yangpu location of the hospital. Approximately 800 babies are born every month in the Yangpu location. Annually, there are about 350 preterm infants admitted in the NICU in the Yangpu location. These preterm infants are separated from their mothers immediately after they are born, and mothers are usually not allowed to see their babies until they are about to be discharged from the hospital. This means that direct breastfeeding is unlikely when the baby is in the hospital and therefore expression of breast milk is necessary to sustain lactation during the separation period.

Best practice recommendations for expression of milk for preterm infants7 (see Appendix I, highlighted the importance of early and frequent expression of breast milk for mothers of newborn premature infants. They also emphasized that relevant healthcare practitioners should provide education to mothers and encourage fathers to provide practical support to their partners during breastfeeding/milk expression.7

Despite the existence of best practice recommendations, the practice in a clinical setting may not adhere to the available evidence. There is evidence to suggest that the knowledge of healthcare professionals regarding breastfeeding for preterm infants was limited.14 Anecdotal evidence suggests that breast milk expression among mothers of pre-term infants in the Yangpu location of the Obstetrics and Gynecology Hospital of Fudan University is not ideal. Therefore, a best implementation project regarding the promotion of early breast milk expression among mothers of preterm infants in the NICU was conducted between June 2018 and October 2018.

Aim and objectives

The aim of this best practice implementation project was to promote early breast milk expression among mothers of preterm infants in the NICU in an obstetrics and gynaecology hospital, with the specific objectives stated as follows:

  • (1) To assess baseline compliance with evidence-based best practice regarding promotion of early breast milk expression among mothers of preterm infants in the NICU.
  • (2) To identify barriers and develop and implement strategies to improve compliance with best practice regarding promotion of early breast milk expression for preterm infants in the NICU.
  • (3) To assess the change in compliance with evidence-based best practice after implementation of strategies.


The JBI Practical Application of Clinical Evidence System (PACES) and Getting Research into Practice (GRiP) audit and feedback tool were used in this best practice implementation project. Three phases were involved:

  • (1) A project team was established and a baseline audit was undertaken using the JBI PACES to determine the current compliance to evidence-based practice recommendations;
  • (2) Barriers to compliance were identified and strategies to address these barriers were designed and implemented using the JBI GRIP framework;
  • (3) A follow-up audit was conducted to evaluate changes in compliance to evidence-based practice recommendations following implementation, and issues to be addressed in subsequent audits were identified.

The project was registered as a quality improvement activity within the hospital, and therefore ethical approval was not required. Written informed consent was obtained from all participants.

Phase 1: Team establishment and baseline audit

A project team was established in June 2018. The team consisted of the Director of Nursing Department of the hospital, the head nurse in charge of the whole obstetric nursing department, the head nurse of the NICU, a nurse specialist of breastfeeding who was registered as an International Board Certified Lactation Consultant (IBCLC), and two nursing researchers. A face-to-face meeting was held to plan the project, and identify audit criteria based on evidence-based recommendations, sample and methods for measuring the compliance with best practice. The criteria were based on the evidence summary 7 and the WHO's 10 steps to successful breastfeeding (revised 2018).15

The audit criteria used in this project for both baseline and follow-up audit are shown in Table 1. The sample and methods used to measure the compliance with audit criteria are also outlined in Table 1.

Table 1:
Audit criteria, sample and methods used to measure compliance

The baseline audit was conducted between 2 July 2018 and 13 August 2018. In the baseline audit, documents of the hospital were reviewed for criterion 1. Healthcare professionals, including the nurses working in the obstetric wards and NICU, midwives working in the delivery rooms, and doctors working in the NICU participated for the audit of criterion 2. Mothers who delivered preterm infants (gestation <37 weeks) admitted into the NICU were recruited for the audit of criteria 3–7. Those who had diseases that prohibited breast milk expression were excluded. A convenience sample of 146 healthcare professionals and 30 mothers were chosen. Questionnaires for healthcare professionals and mothers consisting of demographic information and questions concerning the audit criteria were developed by the research team after the first team meeting. A breast milk expression diary was also designed to record the time, frequency, volume, method, and other issues relevant to breast milk expression; the diary was completed by nurses in the baseline audit and by mothers in the follow-up audit.

Phase 2: design and implementation of strategies to improve practice (Getting Research into Practice)

The project team held two face-to-face meetings during the project period to discuss the baseline audit results, identify barriers to successful implementation of best practice recommendations, propose strategies to overcome the barriers, and determine resources required for successful implementation. A WeChat group including the team members was also established to facilitate regular communication, and online discussions were conducted when a face-to-face meeting was impossible. The barriers, strategies and resources identified were documented in GRiP; the GRiP report was shared within the project team and relevant staff to engage key stakeholders, gather opinion/feedback to facilitate the project and encourage the involvement of related healthcare professionals.

Phase 3: follow-up audit post implementation of change strategy

The same evidence-based audit criteria, sample size and methods as in the baseline audit were used in the follow-up audit, which was undertaken from 14 August 2018 to 2 September 2018. The results were shared within the project team in a face-to-face meeting.


Phase 1: baseline audit

The results of the baseline audit are shown in Fig. 1. As can be seen from Fig. 1, a protocol describing the benefits of breast milk and the principles of milk expression was available and thus a 100% compliance was observed for criterion 1. For criterion 2, only 71% of healthcare professionals (n = 104) reported that they had received education about breastfeeding. For criterion 3, a 0% compliance was detected; six mothers (20%) reported that although they received education about breastfeeding during the antenatal period, no repeat sessions were provided. Most mothers (n = 26) received education in the immediate post-partum period, and thus a compliance of 87% was identified for criterion 4. Although 26 mothers were educated to use a breast pump, only 18 mothers had access to an appropriate breast pump; therefore, the compliance for criterion 5 was 60%. A compliance of 23% was achieved for criterion 6, with only seven mothers able to express breast milk within 6 h after delivery. The average time to initiate expression of breast milk was 13.43 ± 8.91 h. The compliance for criterion 7 was 0%, as none of the mothers was able to express milk 8–12 times in the first 24 h and only two mothers (6.7%) met the criterion for the second 24 h. The average number of expressions of breast milk in the first 24 h was 2.73 ± 1.82 ml, with the number increased to 4.80 ± 2.14 ml in the following 24 h.

Figure 1:
Compliance with audit criteria in baseline audit (%).

Phase 2: strategies for Getting Research into Practice

Four barriers to compliance with best practice recommendations were identified; corresponding strategies and resources were developed to address these barriers (listed in Table 2). First, healthcare professionals were found to have inadequate knowledge about breastfeeding, especially for preterm infants. Therefore, PowerPoint slides about information related to breastfeeding for preterm infants and expression of breast milk were developed by the project team and were used in the face-to-face training for relevant healthcare professionals. Three sessions of similar training were provided to healthcare professionals by the nurse specialist, and an attendance sheet was completed for each training session. Extra training sessions were provided by head nurses in each ward to those who were unable to attend the training sessions of the nurse specialist.

Table 2:
Getting Research into Practice matrix

Second, mothers of preterm infants lacked information about breastfeeding, as many did not attend the classes during the antenatal period due to premature delivery and long distance between their houses and the hospital. In addition, the importance of breastfeeding was not regularly re-emphasized to mothers for whom premature delivery was likely. To overcome the difficulties in educating mothers during the antenatal period, the project team designed an education brochure for mothers of preterm infants, including information about the ingredients of preterm breast milk, the benefit of breast milk for preterm infants, the expression of breast milk, and the storage and transfer of breast milk. The related information was added to antenatal classes. The relevant information was also sent to mothers via WeChat public account notification, which meant that even when mothers did not attend the antenatal classes, they had access to and were reminded to read the information online. The nurse specialist in breastfeeding provided one-to-one education before and after the delivery when the mothers were admitted in the obstetric wards. When the nurse specialist was unavailable on weekends, the nurses on duty in the wards took over the responsibility, and the mothers would be re-educated by the nurse specialist if necessary. The related information was re-emphasized by the midwives when mothers were transferred into the delivery rooms.

Third, some mothers did not have access to an appropriate breast pump, mainly because they forgot to bring the pump with them. Therefore, one bilateral electric breast pump and sterilization was provided for free in each ward and mothers were encouraged to use the pump if necessary. Mothers were also reminded to bring their breast pump when they were admitted to the hospital.

Finally, mothers lacked the support and encouragement to express breast milk when they were separated from their babies. Some strategies were adopted to motivate the mothers. A breast milk expression diary template was provided and printed to mothers for them to record the time, frequency, volume, method, and issues of each expression. Photos when preterm infants were fed with breast milk in the NICU were sent to mothers via WeChat. Fathers were also educated and encouraged to provide practical support to mothers. Doctors and nurses in the NICU were also asked to encourage the expression of breast milk. The nurse specialist helped with expression of breast milk and transferred the milk to the NICU twice a day if possible.

The tools used to collect data, and resources used for education, were all provided in Chinese language. The corresponding author can be contacted for more details.

Phase 3: follow-up audit(s)

The compliance with best practice audit criteria in follow-up audit with comparison with baseline audit is demonstrated in Fig. 2. As shown in the figure, the compliance with criterion 1 remained as 100% and the compliance with the other six criteria increased, although 100% compliance was not observed for all criteria in the follow-up audit. Specifically, the compliance with criteria 2, 4, and 5 improved to 100%, which meant that all healthcare professionals received education and all mothers were educated immediately following delivery. In addition, all mothers were able to access an appropriate breast pump and received support on how to use it.

Figure 2:
Compliance with audit criteria in follow-up audit compared with baseline audit (%).

A considerable increase from 0 to 77% was found for compliance to criterion 3. This indicated that following implementation of the best practice, 23 mothers received education during pregnancy and the information was re-emphasized when premature delivery was likely; however, there were still five mothers who were not educated during the antenatal period and another two mothers did not receive repeat information sessions before delivery. Moreover, the compliance to criterion 6 increased remarkably from 23 to 87%, indicating that in the follow-up audit, 26 mothers expressed breast milk within 6 h after delivery. The average time to the first expression of breast milk reduced significantly from 13.43 ± 8.91 to 4.90 ± 5.26 h.

In contrast to other criteria, the compliance with criterion 7 remained low, and only 50% (n = 15) of mothers were able to express 8–12 times daily in the first 48 h after the baby was born. The percentage increased to 70% (n = 21) in the second 24 h. Mothers expressed 6.77 ± 2.30 times on average in the first 24 h and the mean number of expressions increased to 7.47 ± 1.93 in the following 24 h.

In addition, the volume of breast milk expression in the first 24 h increased from 2.10 ± 5.45 to 7.72 ± 17.19 ml on average. In the following 24 h, the volume improved from 7.08 ± 12.98 to 43.77 ± 80.86 ml.


The evidence implementation project intended to promote early expression of breast milk among mothers of preterm infants in the NICU. Following an initial audit, a series of strategies, including face-to-face counselling, online education, record of milk expression using a diary, and providing feedback to mothers using babies’ photos, were adopted to overcome the barriers identified in clinical practice. From the comparison between the baseline audit and follow-up audit, a remarkable improvement was observed in the education rate of healthcare professionals and mothers, mothers’ access to breast pumps, as well as the time and frequency of breast milk expression. The outcomes of the current project were similar to that of another JBI implementation project on the same topic that was undertaken in Australia.16 Using similar best practice recommendations, culture-specific strategies were utilized to improve clinical practice.

Specifically, the healthcare professionals-led education was emphasized in this project. The importance of providing education to mothers by their healthcare providers has been recognized in many studies.13 In this project, healthcare professionals, including nurses, midwives and doctors, were trained or retrained to provide education and support to mothers. The role of the nurse specialist in providing education about breastfeeding appeared to be critical in this project. The nurse specialist, who has the qualification of IBCLC, is armed with sufficient knowledge about breastfeeding and expression of breast milk. She has the ability and time to provide appropriate and individual education to mothers and fathers. This kind of one-to-one education was of high efficiency and made the information provided consistent. In addition, the involvement of other nurses, midwives and doctors enabled the mothers to be educated at any time before and after delivery, especially when the nurse specialist was not on duty. The multidisciplinary team provided face-to-face education and online information; thus, the breast milk expression practices improved significantly in the follow-up audit.

The education was not only provided to mothers, but also to fathers. Previous studies17,18 suggested that fathers were receptive to information about the benefits of breast milk to babies’ health, and preferred to receive information related to breastfeeding and how to practically support their partner. Mothers’ perceptions of their partner's responsiveness and fathers’ own perceptions of their responsiveness to mothers’ needs predicted longer intention and duration of breastfeeding.19 The evidence suggests that fathers should also be educated to provide practical support that meets mothers’ needs during breastfeeding or breast milk expression.17,19 Therefore, in this project, fathers were also educated about the importance of early, frequent and effective expression, making them understand the practice and to feel motivated to remind mothers to express milk during the middle of the night. Fathers were also encouraged to support mothers in a practical way, such as helping with milk expression, adjusting the position of the mother during expression, and transferring breast milk to the NICU.

In this project, WeChat, a popular social networking software in China, was used to promote group discussion, making the project running smoothly even when face-to-face meetings were unlikely. In addition, WeChat also played an important role in the education and support provided to mothers by disseminating information about breastfeeding and expression of breast milk. It was also used to send photos of babies who were fed with breast milk, which motivated mothers to do the same. When mothers understood that their expressed milk was effectively utilized by their babies rather than being wasted, they were more likely to be motivated to express breast milk early, frequently, and effectively.

Despite the overall increase in the compliance to audit criteria, a 100% compliance was not achieved for criteria 3, 6, and 7, which was similar to the results in the previous implementation project.16 The reason why not all mothers received education during the antenatal period was that some mothers were unable to attend the antenatal class due to premature delivery. In addition, some mothers did not plan to attend the class because they were living far from the hospital and knew that such education could be obtained from healthcare professionals after delivery anyway. Nonetheless, the compliance to criterion 3 of this project (77%) was higher than that in the previous project (55%).16 This may be because the nurse specialist provided one-to-one education during the pregnancy and the midwives were also involved in the education when the delivery was about to happen.

For criteria 6 and 7, the reason for the late initiation and low frequency of breast milk expression may be that most women delivered via caesarean section (73% for baseline audit and 70% for follow-up audit). Caesarean section was reported to be a factor that prohibited early initiation of breastfeeding.20 Women delivered by caesarean section had to stay in the recovery room after delivery much longer than those who had natural delivery. This meant that it was difficult for them to express breast milk earlier. In addition, the pain caused by the wound and side effects caused by anaesthesia may make early expression of breast milk more challenging. In turn, the late initiation of expression is likely to lead to low frequency of expression. Furthermore, expressing 8–12 times daily meant that mothers had to express every 2–3 h. When babies are separated from mothers, it was more difficult for the mothers to have the motivation to express breast milk, especially during the night when mothers were having a sleep. Nonetheless, after the implementation of best practice, the frequency of expression had increased to close to eight times daily, suggesting that the strategies were effective.

There were some limitations to this project. First, the sample size of the mothers was small due to the limited time frame of the project. A further audit of larger sample size and longer time frame may be undertaken in the future. Second, this project only looked at the breast milk expression practice in the first 48 h when mothers resided in the hospital. A further audit may be conducted to explore the practice when mothers were discharged from the hospital. Third, only one follow-up audit was conducted in the project. More rounds of follow-up audit may be carried on to identify other strategies to increase the compliance to criteria, and to assess the long-term effect of the implementation of best practice.


In conclusion, this project has demonstrated that evidence-based early breast milk expression among mothers of preterm infants in the NICU can be facilitated by implementing multipronged strategies that are targeted to the identified barriers. The role of a nurse specialist in breastfeeding, involvement of multidisciplinary staff and utilization of online networking software were valuable in this project. Future audits should be undertaken to evaluate the long-term effect of these strategies, and involving larger sample size and longer time-frame for implementation.


The authors express appreciation to all the healthcare professionals, mothers and fathers who participated in this project.

The current project was supported by Fudan University Nursing Research Funding (no: FNF201862).

Conflicts of interest

The authors report no conflicts of interest.


1. Lewis ED, Richard C, Larsen BM, Field CJ. The importance of human milk for immunity in preterm infants. Clin Perinatol 2017; 44:2347.
2. Menon G, Williams TC. Human milk for preterm infants: why, what, when and how? Arch Dis Childhood 2013; 98:F559F562.
3. Smith H, Embleton ND. Improving expressed breast milk (EBM) provision in the neonatal unit: a rapid and effective quality improvement (QI) intervention. J Neonatal Nurs 2013; 19:149153.
4. Niela-Vilen H, Melender HL, Axelin A, et al. Predictors of breastfeeding initiation and frequency for preterm infants in the NICU. J Obstet Gynecol Neonatal Nurs 2016; 45:346358.
5. Kaya V, Aytekin A. Effects of pacifier use on transition to full breastfeeding and sucking skills in preterm infants: a randomised controlled trial. J Clin Nurs 2017; 26:20552063.
6. Flaherman VJ, Lee HC. ‘Breastfeeding’ by feeding expressed mother's milk. Pediatr Clin North Am 2013; 60:227246.
7. Lizarondo L. Evidence summary. Expression of milk for pre-term infants. The Joanna Briggs Institute EBP Database, [email protected] 2018; JBI15951.
8. Rossman B, Kratovil AL, Greene MM, et al. ‘I have faith in my milk’: the meaning of milk for mothers of very low birth weight infants hospitalized in the neonatal intensive care unit. J Hum Lact 2013; 29:359365.
9. Hallowell SG, Rogowski JA, Spatz DL, et al. Factors associated with infant feeding of human milk at discharge from neonatal intensive care: cross-sectional analysis of nurse survey and infant outcomes data. Int J Nurs Stud 2016; 53:190203.
10. Duan Y, Yang Z, Lai J, et al. Exclusive breastfeeding rate and complementary feeding indicators in China: a National Representative Survey in 2013. Nutrients 2018; 10:249257.
11. Felice JP, Geraghty SR, Quaglieri CW, et al. ‘Breastfeeding’ without baby: a longitudinal, qualitative investigation of how mothers perceive, feel about, and practice human milk expression. Matern Child Nutr 2017; 13:111.
12. Parker AL, Hoffman AJ, Darcy-Mahoney AA. Facilitating early breast milk expression in mothers of very low birth weight infants. Am J Matern Child Nurs 2018; 43:105110.
13. Balogun OO, O'Sullivan EJ, McFadden A, et al. Interventions for promoting the initiation of breastfeeding. Cochrane Database Syst Rev 2016; 11:1100.
14. Yang Y, Li R, Wang J, et al. Knowledge of healthcare providers regarding breastfeeding preterm infants in mainland China. BMC Pediatr 2018; 18:17.
15. World Health OrganizationTen steps to successful breastfeeding (revised 2018). 2018; Geneva:World Health Organization, Available from: [Accessed 16 May 2019].
16. Goodchild L, Hussey L, McPhee AJ, et al. Promoting early expression of breast milk in mothers of preterm infants in a neonatal unit: a best practice implementation project. JBI Database Syst Rev Implement Rep 2018; 16:111.
17. Denoual H, Dargentas M, Roudaut S, et al. Father's role in supporting breastfeeding of preterm infants in the neonatal intensive care unit: a qualitative study. BMJ Open 2016; 6:18.
18. Brown A, Davies R. Fathers’ experiences of supporting breastfeeding: challenges for breastfeeding promotion and education. Matern Child Nutr 2014; 10:510526.
19. Rempel LA, Rempel JK, Moore KCJ. Relationships between types of father breastfeeding support and breastfeeding outcomes. Mater Child Nutr 2017; 13:114.
20. Zhao J, Zhao Y, Du M, et al. Does caesarean section affect breastfeeding practices in China? A systematic review and meta-analysis. Matern Child Health J 2017; 21:20082024.

best practice implementation; breast milk expression; breastfeeding; preterm infant

Supplemental Digital Content

© 2020 JBI. Unauthorized reproduction of this article is prohibited.

A video commentary on implementation project titled: How do health professionals prioritise clinical areas for implementation of evidence into practice? The commentary is provided by Andrea Rochon RN, MNSc, Research Assistant, Queen's University, Ontario, Canada