Web-based Educational Intervention on Breastfeeding Self-efficacy and Breastfeeding Outcome

Study Design A randomized controlled trial design, two-arm, with a 1:1 allocation ratio, will be conducted. The participants will be allocated in the intervention group or control group using block randomization with a block size of 4. The statistician, who will not join in the recruitment process, will generate the sequence of random allocation. To maintain the allocation concealment, we will use opaque sealed envelopes that contain the sequence numbers. Due to the nature of the study, it is difficult to blind the participants. Only the data collector will be blinded to the participant's group assignment.

Participants and setting The mother and her spouse will be invited to participate in this study during the prenatal period from a public hospital and health care center in Jakarta. The inclusion criteria of mothers are: (a) primiparous and singleton pregnancy at 12-34 weeks' gestation, (b) age ≥ 20 years, (c) tend to breastfeed the infant, (d) the husband willing to participate in the educational program, (e) capable to use a cell phone, (f) can access the internet, (g) fluent in reading and writing Bahasa. We will not include the mothers and their partners in the study if they have no time to join in the online educational intervention. Mothers will be excluded if they have conditions as follows: (a) presence of congenital anomaly for the baby, (b) chronic disease or high-risk complications during pregnancy and after delivery. To recruit suitable participants, we will approach the eligible mothers and fathers using online recruitment via whatsapp from the patient list during antenatal check-up and flyer at the hospital and health center. All participants will be provided online informed consent.

Sample size calculation The sample size will be estimated based on a previous study of breastfeeding self-efficacy intervention (Ansari, Abedi, Hasanpoor, \& Bani, 2014). A two-sided two-sample unequal variance t-test is needed (Julious, 2004) to detect a medium effect size in self-efficacy between intervention and control group (Ansari, Abedi, Hasanpoor, \& Bani, 2014) using 80% power at a 5% significance level (alpha). Considering the drop-outs rate and pandemic situation in Indonesia, we will recruit 160 pairs of participants (200 pairs of mothers and fathers for each group) in this study.

Intervention The intervention program will be pilot tested before the formal study. The content validity of the intervention will be conducted by the breastfeeding expert panel including one obstetric specialist, two lactation consultants, one educator and one pediatrician (Appendix 7). The participants in the intervention group will receive standard usual care in primary health care and the web-based educational intervention. The development of the educational intervention is based on the Breastfeeding Self-Efficacy Theory (BSE) (Dennis, 1999) The strategies will be developed from the four sources of breastfeeding self-efficacy framework (Dennis, 1999) including (a) performance accomplishment (past breastfeeding experience); (b) vicarious experience for example observation of others individual performance, whether live, recorded, or printed) (Dennis, 1999; Kingston et.al., 2007); (c) verbal persuasion (i.e. encouragement from a lactation consultant); (d) physiological and emotional arousal (i.e. reducing negative interpretation like pain, fatigue, anxiety or stress). The educational program will be consisted of six sessions which cover 4 elements of breastfeeding self-efficacy theory.

To support the educational program, the researchers will discuss with the expert of information system to prepare the website. The website will be created by using hosting and domain services. Each session of education program will be delivered by written information and video. Additionally, massage board for each session will be provided to allow mothers to ask the question regarding the topic. The website will be designed to be compatible with mobile phone, computer, and also tablets. After the expert approve the applicability of the website, the pilot study will be conducted with 10 mothers and fathers.

The researcher will recruit the mothers and fathers using flyer and the internet advertisement for the potential eligible mothers and fathers. The research team using e-mail or communication application will contact mothers and fathers who are interested to participate in the study. The information regarding the purposes of the research will be given to the mothers and fathers who eligible for this study. After participants agree to participate in the research, the researcher will contact the participants to give the username and the password. The research assistant will guide the participant to log in and fill the personal information. The participants will be reminded to enter the website within 3 days by whatsapp and can complete the education session within 4 weeks. Mothers could not start the other session before completing the previous one.

The total of section is 6-10 and will be followed up by telephone counselling at 2, 4, and 6 weeks postpartum. The duration of each session is 10-20 minutes, while duration of telephone counselling is 30 minutes. Telephone counselling intends to support mothers in sustaining exclusive breastfeeding until six months. The lactation consultant will ask about mother's feeling after delivery, possible missconceptions, dan give emotional support for mothers and fathers. Additionally, these telephone counselling also aim to monitor breastfeeding progress and also identify potential problems related to breastfeeding.

Control group Mothers in the control group will receive standard treatment with pamphlets during the prenatal period. After delivery, standard postpartum care will be provided by the midwife, including pamphlets, rooming in and encourage mothers to breastfeed the baby.

Measurement The primary outcome of this study is maternal and paternal breastfeeding self-efficacy. The secondary outcomes including infant feeding attitude, depressive symptoms, anxiety symptoms, and breastfeeding rate.

Demographic characteristics and obstetric variables Maternal socio-demographic characteristics (i.e. age, level of education, working status, religion and wealth index) and obstetric variables (i.e. parity, mode of delivery and frequency of antenatal check-up) will be measured at baseline (recruitment).

Maternal breastfeeding self-efficacy Maternal breastfeeding self-efficacy is defined as the confidence of a mother in her ability to feed the baby (Dennis, 1999). Bahasa Indonesia version of the Breastfeeding Self Efficacy Short Form (BSE-SF) maternal version will be used to measure self-efficacy among breastfeeding mothers (Sandhi, Lee, Chipojola, Huda, \& Kuo, 2020).

This questionnaire is a 14-items, self-reported questionnaire that measures breastfeeding self-efficacy on a 5-point Likert scale from 1 (not at all confident) to 5 (very confident) (C. L. Dennis, 2003). Total scores are 14-70, with higher scores that suggest greater self-efficacy of breastfeeding. The Cronbach alpha of the BESE-SF Indonesia version was 0.90. We will measure the data at baseline, and follow up at 38 weeks of pregnancy, 1 week, 1, 3, and 6 months after delivery through an online questionnaire.

Paternal breastfeeding self-efficacy Paternal versions of the Breastfeeding Self- efficacy Scale-Short Form (BSES-SF) will be used to measure confidence with breastfeeding (Dennis, Brennenstuhl, \& Abbass-Dick, 2018). This is a 14-item measure where items are rated on a 5-point Likert scale with response format ranging from not at all confident (1) to very confident (5). Items summed to produce a total score ranging from 14 to 70, with higher scores indicating higher breastfeeding self-efficacy. We will measure the data at baseline, and follow up at 38 weeks of pregnancy, 1 week, 1, 3, and 6 months after delivery through an online questionnaire.

Depression The Bahasa Indonesia version of Edinburgh Postnatal Depression Scale (EPDS) will be used to assess postpartum depressive symptoms in mothers (Paramita, Faradiba, \& Febrayosi, 2018). EPDS consist of 10-items which comprises of a 4-point scale ranging from 0 ("no") to 3 ("most of the time"). The total EPDS score ranges from 0-30 whisch highest score indicates a higher depression level. The Bahasa Indonesia version of the EPDS had a Cronbach's alpha 0.86. A cut-off point is ≤ 12 (low scorers) and > 13 (high scorers) (J. L. Cox, Holden, \& Sagovsky, 1987). We will measure the data at baseline, and follow up at 38 weeks of pregnancy, 1 week, 1, 3, and 6 months after delivery through an online questionnaire.

Anxiety The Zung Self-Rating Anxiety Scale (SAS) Bahasa Indonesia version questionnaire will be used to measure anxiety (Setyowati et al., 2019). The SAS consists of 20 items with 4-point scale ranging from 0 ("no") to 3 ("most of the time"). The total score range from 20-80, which the higher score indicating more anxiety. Cronbach's alpha for the Bahasa Indonesian translated of SAS subscales was 0.658. We will measure the data at baseline, and follow up at 38 weeks of pregnancy, 1 week, 1, 3, and 6 months after delivery through an online questionnaire.

Iowa Infant Feeding Attitude Scale (IIFAS) IIFAS Bahasa Indonesia version consists of 16-items questionnaire to measure attitudes of mothers (Utami, 2016). The IIFAS was designed to assess maternal attitude towards infant feeding and to predict breastfeeding intention and exclusivity (Mora, Russell, Dungy, Losch, \& Dusdieker, 1999). Participants will respond for each items on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The total IIFAS score range from 16 to 80 with highest scores reflecting positive attitude towards breastfeeding. The Cronbach alpha of this scale is 0.87 (Utami, 2016). We will measure the data at baseline, and follow up at 38 weeks of pregnancy, 1 week, 1, 3, and 6 months after delivery through an online questionnaire.

Breastfeeding rate Exclusive breastfeeding is defined as mothers who only fed breast milk to their infants. This could also include the delivery of breast milk and vitamin or mineral supplement syrups. Predominant is defined as breast milk-receiving infants as the, defined as infants receiving any liquid or semi-solid food from a bottle of nipple.

A self-report questionnaire will use to assess the mothers' breastfeeding rate. The breastfeeding rate will be collected at 1 week, 1, 3and 6 months of infant age through online or telephone interview.

Data Collection At baseline, demographic characteristics, paternal and maternal breastfeeding self-efficacy, infant feeding attitude, anxiety as well as depression will be measured. We will follow-up those variables at 38 weeks of pregnancy, 1 week, 1, 3, and 6 months after delivery through online questionnaire. Breastfeeding rate will be collected at 1 week, 1, 3and 6 months of infant age through online or telephone interview.

Data analysis Data analysis will perform using Statistical Package for the Social Sciences (SPSS) vers. 21 (IBM, Chicago, IL, USA). The distribution of the continuous variables will report using means and standard deviations, while the continuous data will adopt frequency and percentages. Chi-square test will be used to compare the demographic characteristic of the categorical variables (i.e. level of education, working status, and religion, parity), while t-test for continues variables (i.e. age) between intervention and control group.

The main analyses for this study will use the Generalized Linear Model to test the main effect of educational intervention on breastfeeding self efficacy, breastfeeding attitude, anxiety and depression for group, time as well as group and time interaction. Chi-square test or Fisher exact test will be used. Odds Ratio (OR) and Confidence Intervals (CI) will be calculated to examine the association between intervention and control group.