Study design and participants
This was a randomized controlled clinical trial. The current study was part of a great trial in which the effect of prenatal counseling on the breastfeeding practice of mothers with former failed breastfeeding was studied and the breastfeeding self-efficacy and frequency of breastfeeding problems were measured as the secondary outcomes. This trial leaded from November 2017 to May 2018 on 108 pregnant women go to the health centers in Tabriz, Iran. Insertion criteria were ability to read and write, stay in the city of Tabriz, former failed lactation, monogamy pregnancy, possibility of constant presence at counseling meetings, and being in the third trimester. Exclusion criteria were known fetal abnormalities, high-risk pregnancies, and breastfeeding contraindications.
The sample size was calculated based on the Kordi et al.’s study (29) and the “exclusive breastfeeding” variable. 49 persons were intended for each group by a part of software and considering P1=40% (exclusive breastfeeding frequency in intervention group), P2=17.5% (exclusive breastfeeding frequency in control group), α= 0/05 and power = 80%. 54 persons were computing as the ending sample size for each group, with a 10% potential attrition rate.
Sampling and random allocation
After earning an ethics code from Tabriz University of Medical Sciences’ Ethics Committee (IR.TBZMED.REC.1396.595) and recording the study in the Iranian Registry of Clinical Trials (IRCT20100109003027N19), sampling was prepared by the accessible technique from whole healthcare centers in Tabriz. After presentation an introduction letter, the researcher went to whole healthcare centers and assembled a list of multipara pregnant women in their third trimester. The research aims were clarified to eligible participants through phone calls and they were questioned if they had previous unsuccessful breastfeeding experience. Then they were observed in terms insertion and exclusion criteria if they responded positively to the former question and presented a readiness to contribute in the study. If they have qualifying criteria, they would be invited to a briefing meeting. The study aims and method were described and on paper informed consent was attained from the participants. A socio-demographic information questionnaire was then completed. Participants were assignmented to two groups, i.e. intervention recipient and control, using web www.random.org -based randomized block plan with block sizes of 4 and 6 and a 1:1 allocation ratio. Random allocation was done by an uninvolved person in sampling and data cumulation. Allocation concealment was done via inscription the kind of allocation on sections of paper and placing them in consecutively numbered, opaque, sealed packets. Packets were opened in the order of participant entry in the study and lastly, they were allocated to either the counseling or control group.
Intervention
In the intervention group, collection breastfeeding counseling sessions in clusters of 5-7, with each session for a period 60-90 minutes was begun. Four counseling sessions were held within a one-week’ time and the control group only received routine care of healthcare center such as common brief training about breastfeeding. The educational context in intervention group included breast milk profits and combinations, the psychological benefits of breastfeeding, breast structure and physiology, breastfeeding hormones, usual reasons of breastfeeding discontinuation and failure, common breast conditions and disorders, maternal nourishment during breastfeeding, and breast pumping tips among others. Also, an instructive booklet was given to the intervention group in the end of primarily session. Phone or, if necessary, face to face counseling was scheduled to mothers until the day 15 and the end of the month 4 after childbirth in case of got into trouble by the same consultant . A researcher-made checklist of frequency of breastfeeding problems and standard breastfeeding self-efficacy questionnaire (BSES) were completed for both groups on day 15 and months 2 and 4 postpartum.
Data Collection Tool
In this study, self-efficacy was measured using a self-efficacy measure (BSES: Breastfeeding self-efficacy scale) for breastfeeding. Breastfeeding self-efficacy has 33 items and is scored on Likert spectrum with score 1 for “totally disagree” to score 5 for “totally agree”. The distance between scores is from 33 to 165, and higher scores indicate a higher level of self-efficacy in breastfeeding. Bandura (1997) designed this tool, and Faux and Denis first used it for breastfeeding (16). Shahri et al. (2015) conducted the psychometric evaluation of the Persian version of breastfeeding self-efficacy measure (30). To determine the scientific reliability of the self-efficacy questionnaire, test-retest analysis was conducted. The questionnaires were filled by 20 clients once, and the filling was repeated after 2 weeks, then reliability was obtained by determining in-line correlation coefficient; Cronbach alpha was measured to be 0.826 and 0.834, respectively, and confirmed by confidence interval of 95%. In this study, the questionnaire was completed at three times in the 15th, 2nd and 4th month after delivery in two intervention and control groups. Breastfeeding problems were measured using a checklist of breastfeeding abundance problems designed by the research team. Breastfeeding problems such as inadequate milk, refusing the breast, common problems of breasts, and problems of the infant were compared at three time intervals of 2 and 4 months and 15 days after delivery in two intervention and control groups.
Data analysis
SPSS 24 software package was employed to analyze the results. The normal distribution of quantitative data was examined using the Kolmogorov–Smirnov test, indicating the normal distribution of all data. The socio-demographic information of the two groups was compared using the chi-square test, the chi-square test for trend, independent t-test, and Fisher’s exact test. The analyses of the inter-group breastfeeding self-efficacy were conducted using the chi-square test and repeated measure ANOVA test, while controlling such variables as willingness to pregnancy, mother’s level of education, and wife’s level of education. The inter-group frequency of breastfeeding problems were compared using the chi-square and Fisher's Exact test. P<0.05 was considered as the significance level.