Maternal resolution of grief after preterm birth: implications for infant attachment security

Abstract

Objective: This study explored the association between mothers' unresolved grief regarding their infant's preterm birth and infant-mother attachment security. We hypothesized that mothers with unresolved grief would be more likely to have insecurely attached infants at 16 months and that this association would be partially mediated by maternal interaction quality.

Methods: This longitudinal study focused on 74 preterm infants (age of <36 weeks) and their mothers who were part of a larger study of high-risk infants. The present analysis included assessment of neonatal and socioeconomic risks at NICU discharge; maternal depression, Reaction to Preterm Birth Interview findings, and quality of parenting at a postterm age of 9 months; and infant-mother attachment at postterm age of 16 months. Associations among findings of grief resolution with the Reaction to Preterm Birth Interview, quality of parenting interactions, and attachment security were explored by using relative risk ratios and logistic and multivariate regression models.

Results: The relative risk of developing insecure attachment when mothers had unresolved grief was 1.59 (95% confidence interval: 1.03-2.44). Controlling for covariates (adjusted odds ratio: 2.94), maternal feelings of resolved grief regarding the preterm birth experience were associated with secure infant-mother attachment at 16 months. Maternal grief resolution and interaction quality were independent predictors of attachment security.

Conclusion: Maternal grief resolution regarding the experience of preterm birth and the quality of maternal interactions have important implications for emerging attachment security for infants born prematurely.

Figures

FIGURE 1

Hypothesized partial mediation model. SSP indicates strange situation procedure. A indicates the relationship between the predictor (RPBI) and outcome (attachment); B, relationship between the predictor (RPBI) and the intervening variable (PCERA); C, relationship between the intervening variable (PCERA) and outcome (attachment).

FIGURE 2

Distribution of attachment classifications (n = 74).

FIGURE 3

Maternal positive interactions and resolved grief as predictors of infant security. SSP indicates strange situation procedure; CI, confidence interval.