482 - Postpartum bonding and early child growth trajectories

Publication Number: 4473.482

Cristina R.. Fernandez, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Austen Curtin, Columbia University Medical Center, New York, NY, United States; Hope Hendry, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Elena Arduin, Columbia University Vagelos College of Physicians and Surgeons, New York, NY, United States; Andreane Lavallee, Columbia University, New York, NY, United States; Dani Dumitriu, Columbia University, New York, NY, United States

Background: Optimal early growth trajectories are critical for child health and development. Strong early relational health (ERH)-defined as the quality of caregiver-child relationships-may support ideal growth starting in infancy. Yet the role of postpartum bonding, an ERH construct reflecting the parent-to-infant emotional tie, in shaping early growth trajectories is unclear.

Objective: To determine associations between postpartum bonding and early child growth trajectories.

Design/Methods: Secondary longitudinal data were drawn from the Columbia University COMBO Initiative study. Postpartum bonding as primary exposure was measured at 4 months via the validated 25-item Postpartum Bonding Questionnaire (PBQ; total score 0-125, higher scores= more bonding difficulties). Repeated measures of body mass index (BMI) and weight-for-length (WFL) z-scores as primary outcomes were calculated from weight and length measurements obtained during outpatient clinic visits through 18 months. Linear mixed models, with random intercepts to account for repeated measures and PBQ x age interactions, tested whether associations between exposure and growth outcomes varied across child age. Models were adjusted for child sex, gestational age, exclusive breastfeeding through 4-6 months, child age at clinic visit as a continuous variable, and maternal pre-pregnancy BMI.

Results: N=186 children had complete PBQ and growth data. Mean gestational age was 38.6±2.2 weeks; 44% were female, 28% exclusively breastfed through 4-6 months, and maternal BMI averaged 32.3±6.9 kg/m2. Mean PBQ total score was 6.7±5.6. Distribution of growth trajectories show lower PBQ scores, indicative of stronger bonding, correlate with BMI and WFL z-scores within a healthy range (Fig. 1). Higher PBQ scores, indicative of more bonding difficulties, were significantly associated with lower BMI z-scores (β=-0.037, 95% confidence interval: -0.07, -0.01, Fig. 1A). A significant PBQ x age interaction showed that bonding difficulties were associated with lower WFL z-scores among infants < 8.3 months but not among older infants (Fig. 1B).

Conclusion(s): Stronger postpartum bonding correlated with BMI and WFL z-scores in a healthy range. Postpartum bonding difficulties were linked to lower BMI z-scores, and to lower WFL z-scores among younger infants but the effect was no longer significant as infants grew. Additional research is needed on the relationships between repeated measures of PBQ and other ERH constructs, both milk and complementary feeding, and longitudinal growth trajectories into childhood.

Figure 1. Relationship between Postpartum Bonding Questionnaire total score at 4 months postpartum and trajectories of infant body mass index z-scores and weight-for-length z-scores through 18 months-old, N=186
All models controlled for maternal pre-pregnancy BMI and infant sex, gestational age, exclusive breastfeeding through 4-6 months-old, and age at primary care clinic visit (as a continuous variable). For visualization, infant age at clinic visit is displayed in upper, middle, and lower terciles.
Distribution of growth trajectories show lower PBQ scores, indicative of stronger bonding, correlate with higher relative body mass index (BMI) and weight-for-length (WFL) z-scores within a healthy range. Panel A: Higher Postpartum Bonding Questionnaire (PBQ) total scores (x-axis) were significantly associated with lower infant BMI z-scores (y-axis). Panel B: A significant negative interaction was found between PBQ scores (x-axis) and infant age at clinic visit for WFL z-scores (y-axis) such that higher PBQ scores (more bonding difficulties) were significantly associated with lower WfL z-scores among children <8.3 months-old (lower and middle age terciles) but not in infants >8.3 months-old (upper age tercile).