Poster #35 - Within-Person Relationships between Mother-Infant Bonding and Postpartum Depressive Symptoms: A Birth Cohort Study

• In Event: Poster Session 04

Abstract

During the perinatal period, sudden environmental and physical changes increase the risk of maternal mental health deterioration. Previous research suggests that mother-infant bonding and depressive symptoms change over the postpartum period, and these two changes may interact (Kasamatsu et al. 2020; O’Higgins et al. 2013). However, previous studies have only examined the relationship at the between-person level, and whether the pattern is generalizable to dynamic relations at the within-person level remains unknown. In this study, we applied the random intercept cross-lagged panel model [RI-CLPM; Hamaker, Kuiper, and Grasman (2015); Figure 1] to examine the intra-individual reciprocal relations between mother-infant bonding and postpartum depressive symptoms, across an 18 month postpartum period.

A total of 434 mothers participated in a birth cohort study, the Chiba Study of Mother and Child Health (Sakurai et al. 2016). They completed the Mother-to-Infant Bonding Scale (MIBS) and Edinburgh Postnatal Depression Scale (EPDS) at the following time points: a few days after delivery (0mo), one month after delivery (1mo), four months after delivery (4mo), ten months after delivery (10mo) and 18 months after delivery (18mo). The MIBS and EPDS are self-report questionnaires designed to assess maternal bonding with their babies, and postpartum depression. A high score indicates poor mother-to-infant bonding and high depressive symptoms.

Full information maximum likelihood was used to treat missing data, and data for 360 individuals were included in the analysis. A model, in which the cross-lagged, auto-regressive paths and correlations between within-person components were constrained to be equal across time, was employed because it showed good model fit, and did not significantly worsen the goodness of fit relative to the unconstrained models (CFI = .985, RMSEA = .038, ⊿χ2(16) = 23.426, p = .102). In this model, the cross-lagged coefficients from mother-infant bonding failure to postpartum depressive symptoms were positive and significant (Table 1). This suggests that when mothers experience greater difficulty in bonding with their child than individual average, their depressive symptoms are worse than the individual average at the following time points. In contrast, deviations in postpartum depressive symptoms did not exhibit significant effects on changes in mother-infant bonding impairments. At the between-person level, the time-invariant component of mother-infant bonding impairment was positively related to the time-invariant component of postpartum depressive symptoms. These results did not change significantly after controlling for the participants’ age at delivery, income, education level, previous childbearing experience, and sex of the child.

This study showed that difficulties in mother-infant bonding preceded the increase in depressive symptoms at the within-person level, and that the inverse relationship was not significant. Although previous studies have indicated that prior depressive symptoms exacerbate bonding with children, we suggest that findings from inter-individual relationships may not necessarily be generalizable to intra-individual relationships. The results provide more robust support for the possibility that the depressive symptoms of the same individual can be alleviated by interventions related to bonding with their fetus or infant.

Authors

• Daiki Hiraoka, University of Fukui
  Presenting Author

• Akiko Kawanami, Chiba University

• Kenichi Sakurai, Chiba University

• Chisato Mori, Chiba University