Aim
The aim of the study was to examine whether infants' sensory responsiveness is associated with exclusive breastfeeding and whether this can predict non-exclusive breastfeeding at 6 months of age.
Design
This study is a cohort prospective study, designed to have data collected at three time points: at 2 days, 6 weeks, and 6 months after birth. It acquired relevant study factors on infants, utilizing mother's self-reports, and observational assessments of the infants at their homes.
Participants
Mothers hospitalized in the maternity ward between June 2019 and August 2020 were recruited 2 days after birth, using a convenience sampling method. Inclusion criteria were the desire to initiate breastfeeding; older than 20 years of age; no language barriers; healthy; and having given birth to a healthy single newborn between 36–42 weeks of gestation. Exclusion criteria were undergoing chemotherapy; HIV positive; having given birth to a newborn who was born small for gestational age (below the 3rd percentile), needed to be fed partially or fully with a tube, or scored below the cut-off for typical cognitive and language development at the age of 6 months as assessed by the Bayley-III.
Sample size was calculated based on power analyses via G*Power 3 Software derived from a p-value of .05 and statistical power of .80, yielding N = 150.
Adopting an even more conservative approach, 174 mothers and their newborn infants were recruited, thus ensuring greater statistical power and a decreased risk of type II error. Ten mothers did not reach the third data collection point (6 months after birth) therefore we were unable to determine their breastfeeding status (EBF vs. NEBF). These mothers were excluded from the analyses, resulting in a sample of 164 mothers. Of these 164 mothers, 6 were unavailable to fill the questionnaires at 6 weeks after birth, however continued to participate in the study and filled the questionnaires at 6 months. Analyses which included measures from the age of 6 weeks were performed on a sample of 158 mothers. Note that only 104 out of 164 infants received at-home assessments at age 6 months due to COVID-19 restrictions.
At 6 months after birth, after completing the assessments, the participating mothers provided information about their breastfeeding status via a questionnaire and their infants were divided into two groups accordingly: i) EBF group who exclusively breastfeed, and ii) NEBF group who did not breastfeed at all or partially breastfed (one or more formula feedings each day). The EBF group consisted of 105 infants, and the NEBF group consisted of 59 infants.
Data Collection
The study was conducted between June 2019 and January 2021 in a leading medical center where the hospitalization period is 48 hours following a vaginal birth and 96 hours after a cesarean birth. On the second day after birth in the maternity ward, all participating mothers provided written consent and completed self-administered paper questionnaires handed out by the main researcher. Six weeks and 6 months after birth, participating mothers completed online self-administered questionnaires. Following the submission of the 6-month questionnaires, a qualified occupational therapist conducted infant assessments at the participant’s home. In all phases of the study, the researcher collecting the data and the examiner assessing the infants at home were blinded to the breastfeeding status and duration. The examiner was not exposed to data collected at the three-time points via the questionnaires. At 6 months (the 3rd time point) the mothers were instructed to feed their infants before the home evaluation and not to reveal their infant's breastfeeding status. The examiner did not discuss feeding methods or breastfeeding status with the mothers before completing and documenting the infant's evaluation.
Measurements
Demographic and delivery information questionnaire, including information about the infant, was compiled for this study, and completed by the participating mothers at the maternity ward 2 days after birth, indicating their intention to breastfeed.
Infant sensory profile 2 (ISP2), a standardized, reliable, and valid parental report questionnaire (37), was developed to assess the sensory responsiveness as reflected in daily activities. The infant questionnaire addressed ages 0–6 months, consisting of 25 items. Parents rated the frequency of their infant's behaviors on a five-point Likert scale from one (almost always) to five (almost never). The score for each sensory system was calculated for the auditory, visual, vestibular, tactile, and oral sections, and a total score was calculated as well. Higher scores indicate a higher frequency of over-responsivity, whereas lower scores indicate under-responsivity. The total score is interpreted relative to age norms: (a) typical performance; (b) atypical performance - more than others (sensory over-responsivity) or less than others (sensory under-responsivity) (1–2 SD). Participating mothers completed ISP2 6 weeks after birth. Internal consistency was demonstrated (Cronbach α = 0.75), as was test re-test reliability (ICC = 0.86) and good content and structure validity (37). Cronbach's α for the ISP2 in this study was 0.806.
The Test of Sensory Functions in Infants (TSFI) is a standardized, reliable, and valid tool (38) hat was developed to asses sensory responsiveness in infants aged 4–18 months. The TSFI comprises 24 items that evaluate responses in 5 subdomains: tactile deep pressure, visual-tactile integration, adaptive motor function, ocular motor function, and reactivity to vestibular stimulation. Each of the subdomains produces a specific age-normed score, and a total score is calculated as the sum of all subdomains. The total score ranges from 0 to 49, a higher score indicates a more typical sensory responsiveness whereas a lower score indicates sensory over-responsivity behaviors. Participating infants were assessed using the TSFI at 6 months of age. The TSFI has been demonstrated to have test-retest reliability (ICC = 0.88–0.99) and inter-rater reliability (ICC = 0.26–0.84). Content and construct validity were established (39).
Bayley Scales of Infant and Toddler Development − 3rd Edition (Bayley-III) is a gold standard, standardized, reliable, and valid tool used to evaluate the development of infants and toddlers from 1 to 42 months of age. The tool comprises cognition, language, motor, social-emotional, and adaptive-behavior subtests (40). For this study, only the cognitive and language (receptive and expressive communication) subtests were used, solely to test the exclusion criteria. Each test item was scored as "credit" or "not credit", and the credit scores were summed to obtain the total raw scores for each scale. The raw scores were converted into norm-based composite scores. To interpret the results, the composite score was used with an average of 100 points and standard deviation of 15 points. Cognitive development was considered appropriate when the results of the composite score ranged from 85 to 115 points. In this study, participating infants were assessed using Bayley-III at 6 months of age. The Bayley-III has demonstrated internal consistency (Cronbach α > 0.86), test-retest reliability (ICC > 0.67), and concurrent validity compared to several developmental diagnostic tests (40).
Breastfeeding status measure. In a one-item self-report developed for this study, participating mothers reported on their breastfeeding status (exclusive, partial, or none). In the current study, the breastfeeding status measure was obtained at 6 months.
Ethical considerations
All aspects of the study were approved by the Institutional Ethics Review Board of the Medical Center (reference number 0302-14-MMC) and the University Review Board, and written informed consent was obtained from participating mothers. All participating mothers were assured that participation was voluntary and that they could choose to withdraw from the study at any time. Mothers' and infants' privacy was ensured, and their details were kept confidential.
Data analysis
Statistical analyses were performed with SPSS® V27 (IBM Corp., Armonk, NY, USA). Data was aggregated with descriptive statistics by data type. Categorical variables were compared between the EBF and NEBF groups using a chi-square test, and the continuous variables were compared using MANOVA tests. To assess the extent to which exclusive breastfeeding could be explained by the predictors, we employed a hierarchical logistic regression scheme using a stepwise method. The variable selection was based on t-test and MANOVA analyses, and the data was entered in chronological order: The demographic and delivery-related variables were entered at the first step of the regression. The second step included the sensory profile of the infants, assessed at 6 weeks, (typical/atypical performance). The TSFI-total score, measured at 6 months, was entered at the third step. Odds ratios and their 95% confidence intervals are presented. All statistical tests were two-sided, and a p value of < .05 was considered statistically significant. Nominal p values, Cohen's d and Partial Eta Squared effect sizes are presented. Cohen’s d values are defined as small (0.2), medium (0.5), or large (0.8); Partial Eta Squared values are typically referred to as small (0.01), medium (0.06) and large (0.14) (41).