This study used a quasi-experimental design with a program group, pamphlet group and comparison group in an urban area of Japan and was conducted from February 2017 to August 2018. In the program group, the intervention was delivered before returning to work and measured 3 months after returning to work. The program involved a breastfeeding class (90 min), use of a pamphlet, distribution of a newsletter upon returning to work, and email consultation up to 3 months after returning to work. In the pamphlet group, this was sent before returning to work and breastfeeding continuation rate was measured at 3 months after returning to work. The pamphlet was the same one that was used in the program group. No consultation was given to the pamphlet group. Breastfeeding continuation rate was measured in the comparison group only at 3 months after returning to work without intervention. The outcome measure was breastfeeding continuation rate after returning to work.
Participants and setting
The inclusion criteria of the program group and the pamphlet group were: 1) women who planned to return to work within 4–12 months after giving birth; 2) women who were breastfeeding at the time of recruitment; and 3) women who could communicate and read and write in Japanese. There were no exclusion criteria. The reasons why returning to work after at least 4 months after childbirth was included were as follows. First, breastfeeding rates in Japan rise up to 4 months after birth and are maintained for up to 6 months . Second, since complementary food is started from the age of 6 months, it was thought that women who did not have enough breast milk could continue breastfeeding while using complementary food. To recruit participants for the program group, cooperation was requested from medical and childcare facilities near the program venue. Posters and leaflets requesting participation in the study were distributed at these facilities, and women interested in cooperating were asked to contact the researcher. As a result, program group participants were recruited from seven clinics, four health centers, 19 childcare support centers, five maternity care houses, and four daycares. Prior to the implementation of the program, the purpose of the research was explained verbally and in writing, and consent to participate in the research was obtained. Participants for the pamphlet group were also recruited with the cooperation of medical and childcare facilities. Posters and leaflets requesting participation in the research were distributed at these facilities. As a result, pamphlet group participants were recruited from one hospital, one clinic, nine childcare support centers, five maternity care houses, five daycares, and one private company. The pamphlet was distributed by postal mail, along with a document explaining the purpose of the research, and consent to cooperate was obtained from all participants.
The comparison group included women who returned to work within 4–12 months after giving birth, had been back at work for at least 3 months, were breastfeeding before returning to work, and could read and write in Japanese. There were no exclusion criteria. Comparison group participants were recruited from 22 daycare facilities. A document explaining the research was enclosed with the questionnaire, and returning the questionnaire was interpreted as consent to participate in the research.
In Japan, it is not common for mothers to receive breastfeeding support before returning to work. Therefore, the breastfeeding status of mothers who had already returned to work should reflect the general breastfeeding status.
Description of the breastfeeding support program
The framework of the program was transformative learning , adult learning theory, and empowerment . The program involved a 90-min breastfeeding class, a pamphlet, a newsletter, and email consultation. The purpose of the class was to empower women returning to work by providing knowledge about the continuation of breastfeeding, allowing mothers to discuss breastfeeding with their peers, and for mothers to choose to continue breastfeeding after returning to work. Participants reflect on their breastfeeding experience, recognize the value of breastfeeding, and increase their self-efficacy. This leads to the behavior of choosing to continue breastfeeding after returning to work. At the end of the class, the women wrote an action plan on what to prepare before returning, and how to continue breastfeeding after returning to work.
Adults have a need to be independent in learning, and often realize the need for learning when trying to fulfill developmental tasks and social roles. It is presumed that the participants of this program who are about to return to work have high learning needs. Transformative learning is the process of critically self-reflecting and questioning values. Each participant had experienced breastfeeding since childbirth. After returning to work, when the mothers spent more time separated from their babies, they thought about what they wanted to do with respect to breastfeeding and what was best for the baby. Through small group discussions, mothers were able to share their feelings and worries with each other.
Peer support is effective for breastfeeding support [21,22,23]. In the class, participants watched a 10-min video presenting the experience of two women who continued breastfeeding after returning to work. One of them fed her child only breast milk, and the other used mixed nutrition. The video showed mothers’ ideas of breastfeeding and the actual conditions before and after returning to work.
The number of participants was limited to 10 people at one time, and group discussions were limited to about five in consideration of group dynamics. Each class was run by two midwives and two support staff with experience in caring for babies. A researcher was in charge of class progress, and another midwife assisted. If there were 10 participants, they were divided into two groups with each midwife facilitating a discussion. If there were fewer than five participants, both midwives participated in the group.
The two support staff took care of the babies and maintained their safety so that the participants could concentrate on the class with confidence. The room environment was arranged so that women and infants could relax together.
Program participants were able to consult with the researcher by email for up to 3 months before returning to work. A newsletter was sent once to the participants before and once after returning to work. The purpose of the newsletter was to share the results of consultations with the participants and prevent them from dropping out of the study. The newsletter was one double-sided, A4-size page printed in color. The contents included bullet-point advice such as preventing problems regarding continuation of breastfeeding.
Pamphlet structure and contents
The pamphlet contents presented information that could be used before and immediately after returning to work. The pamphlet was in color and consisted of eight, A6-size pages and a cover. The information in the pamphlet included the long-term effects of breastfeeding, how to express breast milk, how to take medications while breastfeeding, weaning, laws related to mothers’ rights in the workplace and the web address of a breastfeeding support organization. In addition, the pamphlet included examples of two women who continued breastfeeding after returning to work. These examples were created based on a previous study  that included interviews with 10 women who continued breastfeeding while working and clearly showed preparation and ingenuity to continue breastfeeding, and the actual situation of breastfeeding after returning to work. The contents of the class and the pamphlet were approved by two midwives with extensive breastfeeding experience. The content validity of the pamphlet was reviewed by two midwifery researchers and two women with breastfeeding experience, and was subsequently revised based on their feedback.
The outcome was breastfeeding continuation rate at 3 months after returning to work. In this study, breastfeeding continuation was defined as breastfeeding at least once a day. The sample size was calculated assuming that the rate of breastfeeding continuation after returning to work was 60 and 30% in the program and comparison groups, respectively, and that the difference between the two groups was 30%. A power analysis was performed using two-sided analysis with an α error of 0.05 and a power of 0.8. Forty-two participants were needed for each group .
The demographic variables were maternal age, month of birth, parity, employment status, education level, smoking status, and previous breastfeeding experience. Breastfeeding-related variables were timing of return to work postpartum, working hours per day, partner’s support in child care, presence of a peer to assist with breastfeeding, consultation with midwifes, and workplace environment for breastfeeding (milk expression breaks, lactation room, refrigerator to store breast milk), and daycare environment for breastfeeding (acceptance of expressed milk). Mothers were also asked about how they were feeding their infant with response choices of breast milk only, infant formula only, and mixed feeding.
Data were analyzed using descriptive statistics. One-way analysis of variance was used for continuous variables. The chi-square analysis was used for comparison of categorical variables. When the expected frequency was five or less, Fisher’s exact test was performed.
Since it was assumed that background factors of the study participants in each group would influence the intervention outcomes, adjusted results using logistic regression analysis were obtained. The dependent variable was the breastfeeding continuation rate at 3 months after returning to work, and the intervention variable was the program intervention. There were eight independent variables considered to affect the continuation of breastfeeding after returning to work: maternal age, timing of return to work, working hours per day, education level, breastfeeding experience, partner’s support in child care, presence of a peer to assist with breastfeeding, and consultation with midwives. Statistical analyses were conducted using SPSS version 25.0 with a two-sided 5% level of significance.
This study was approved by the Research Ethics Committee of St. Luke’s International University (No. 16-A076) and Kanagawa University of Human Services (No. 10–57). The participants provided written informed consent before study participation.