Effects of the experience of breastfeeding-friendly practices and breastfeeding intention and self-efficacy on breastfeeding behavior: a cohort study in Taiwan

Study design and participants

A longitudinal cohort study was conducted over 6 months after childbirth from September 2019 to August 2020. Experience of breastfeeding-friendly practices and breastfeeding self-efficacy were measured using self-reported questionnaires three to 5 days after giving birth in the postpartum wards and at one, two, four and 6 months after childbirth. This study consecutively recruited women 20 years or older who gave birth to healthy newborns at a gestational age of 36 to 42 weeks at two Baby-Friendly-accredited hospitals in Taiwan; had initiated breastfeeding; and could read or write Traditional Chinese. The exclusion criteria were as follows: (1) abnormal maternal or neonatal health conditions after delivery that required hospital admission; (2) prohibition of breastfeeding according to medical recommendation; and (3) multiple births. The sample size was estimated with the rule of thumb: logistic regression and Cox models should be used with a minimum of ten outcome events for each predictor variable [22]. The planned attrition rate was calculated at 20% according to related studies [14], and the estimated sample size was 170.

Instruments

The instruments consisted of four self-reported questionnaires on maternal characteristics, the experience of breastfeeding-friendly practices, breastfeeding self-efficacy, and breastfeeding behavior. The maternal characteristics, including the participant’s age, education level, marital status, occupation before childbirth, time of return to work, occupation after childbirth, parity, delivery type, experience of breastfeeding, and breastfeeding intention, were recorded using a self-developed questionnaire. Breastfeeding intention was defined as the duration for which the mother was willing to breastfeed [14], and the responses for breastfeeding intention were grouped into 1–4, 4–6, 6–12, and >  12 months [9, 14]. The range of the item content validity index (I-CVI) was 0.9–1; the average content validity index (S-CVI) of scales was 0.99.

The experience of breastfeeding-friendly practices in this study was defined as the mother’s experience of the Baby-Friendly practices implemented directly for mothers or infants to support breastfeeding and was measured using a seven-item self-developed questionnaire. In reference to the WHO Ten Steps to Successful Breastfeeding in 2018 [4], the experience of breastfeeding-friendly practices in this study was categorized into the experience of “immediate postnatal care” during maternity hospitalization and “breastfeeding establishment and infant feeding practices” from maternity hospitalization to 6 months after childbirth. Immediate postnatal care comprised the two items of skin-to-skin contact (categorized as Yes and No) and timing of breastfeeding initiation (categorized as < 1, 1–4, 4–24, and >  24 hours). The second section on the experience of breastfeeding-friendly practices was composed of five items that measured breastfeeding establishment and infant feeding practices on a 5-point Likert scale for (1 = never, 2 = rarely, 3 = sometimes, 4 = often, and 5 = always); and the total score of the experience of breastfeeding-friendly practices was 25 points. Items for breastfeeding establishment and infant feeding practices included the experience of rooming-in, maintaining the production of breast milk, breastfeeding on demand, avoiding feeding bottles and teats, and avoiding the use of pacifiers. The S-CVI of the experience of breastfeeding-friendly practices during hospitalization was 1, and that at one, two, four and 6 months was 0.96.

Breastfeeding self-efficacy was defined as a mother’s perceived confidence in her ability to breastfeed her infant [23] and was measured using the traditional Chinese version of the Breastfeeding Self-Efficacy Scale-Short Form (BSES-SF), as translated by Hu [24]. The original Breastfeeding Self-Efficacy Scale (BSES) was developed by Dennis and Faux in 1999 and is based on breastfeeding self-efficacy theory. In 2003, Dennis revised the original 33-item BSES into the 14-item BSES-SF, and psychometrically, the revised BSES-SF was divided into two domains, namely technical and intrapersonal thoughts [25]. The BSES-SF employed a 5-point Likert scale, with the score of each item ranging from 1 (not at all confident) to 5 (always confident). The total BSES-SF score ranged from 14 to 70, with a higher score representing a higher level of maternal self-efficacy in breastfeeding. Cronbach’s alpha was 0.94 for the original BSES-SF (n = 667) [25] and 0.93 (n = 48) for the traditional Chinese version of the 14-item BSES-SF [24]. In this study, Cronbach’s alpha for the traditional Chinese version BSES-SF was 0.95 (n = 155).

Breastfeeding behavior was assessed using a structured questionnaire developed by the authors. Based on the definition of breastfeeding indicators by the WHO in 2008 and standardized indicators of breastfeeding described by Greiner in 2014, breastfeeding behavior was measured using point-in-time data (24-hour recall) and data reflecting breastfeeding behaviors since birth [26, 27]. The indicators of breastfeeding behavior in this study included feeding patterns and the duration of breastfeeding, and three types of breastfeeding patterns were referenced according to what the infant was fed in the past 24 hours [27, 28]. Exclusive breastfeeding indicated that the infant was fed no other food or drink, not even water, other than breast milk, although they may have received prescribed medicines, oral rehydration solutions, or vitamins and minerals [28]. Partial breastfeeding was defined as the feeding pattern in which the infant receives breast milk supplemented with baby formula or solid or semisolid foods. Replacement feeding referred to the feeding pattern in which the infant receives baby formula milk or solid or semisolid foods without breast milk [26, 29]. Exclusive breastfeeding and partial breastfeeding were grouped under “any breastfeeding.” Breastfeeding duration was defined as the length (in weeks) of any breastfeeding within 6 months after childbirth. Both the I-CVI and S-CVI calculated by five experts of maternal nursing and instrument development were 1.

Data collection and study procedure

After approval to conduct the study was obtained from the institutional review boards of the two participating institutions, eligible women were recruited from these hospitals. Women who met the eligibility criteria were approached by the researcher two to 4 days after giving birth in the respective maternity wards. The study participants provided informed consent after a verbal explanation of the study procedure and objectives was provided. An online survey system was used to deliver the questionnaire through e-mail or mobile instant messaging applications during participant hospitalization and at the follow-ups of one, two, four and 6 months after childbirth. In Taiwan’s National Health Insurance program, women can receive 3 days of hospital care after vaginal birth and 5 days after a Cesarean section. Therefore, we conducted the first survey three to 5 days after childbirth, depending on the participant’s length of stay. The first survey comprised questions on maternal characteristics, experience of breastfeeding-friendly practices, breastfeeding self-efficacy, and breastfeeding behavior. Follow-ups at one, two, four and 6 months after childbirth were conducted for data collection on the experience of breastfeeding-friendly practices, breastfeeding self-efficacy, and breastfeeding behavior.

Data analysis

Statistical analysis was performed using R software v4.0.5, and descriptive statistics were obtained to explain the demographic data. Factors associated with breastfeeding behaviors in different periods after childbirth were examined using one-way analysis of variance (ANOVA) and the Kruskal–Wallis, Chi-square, Fisher’s exact test, and Wilcoxon rank sum tests based on the data type of the variables. Multivariable multinomial logistic regression and multiple logistic regression were conducted to identify the determinants of breastfeeding behavior at four and 6 months after childbirth. In addition, Cox regression analyses were performed to identify factors associated with breastfeeding duration; a nomogram measuring the probability of breastfeeding termination at 6 months was also generated. A P-value of less than 0.05 represented the level of significance.

留言 (0)

沒有登入
gif