The association between breastfeeding and attachment: A systematic review

In the popular and scientific literature an association between breastfeeding and attachment is often supposed although this has not been systematically investigated. This is the first review that provides a systematic overview of the current state of research on the association between breastfeeding and attachment. The objective is to determine whether breastfeeding is associated with child attachment as well as if maternal attachment is associated with breastfeeding behaviour.

A systematic electronic literature search for English language articles published from 1963 to 2019 using the databases Web of Science, PubMed, PsycInfo, and PsycArticles was conducted. Studies were included if (1) the association between breastfeeding and attachment of the child or maternal attachment style and breastfeeding was investigated, (2) participants were mothers or mother-child dyads, and (3) quantitative measures of breastfeeding and attachment were used. Study quality was rated using the NIH Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies.

Studies examining mothers or mother-child dyads were reviewed. Sample sizes varied between N = 60 and N = 8900 participants. On average, mothers were between 27.56 and 32.50 years old.

Eleven studies were eligible for inclusion. Of seven studies examining breastfeeding and child attachment, four found that a longer duration of breastfeeding was significantly associated with higher levels of attachment security after controlling for various covariates. No significant differences in attachment security between breastfeeding and bottle-feeding were reported, but one study found higher levels of attachment disorganization for bottle-feeding, although mean levels were below a clinically relevant level. The initiation of breastfeeding directly after childbirth was not related to child attachment. Of four studies examining maternal attachment and breastfeeding, three found a significant association between secure attachment of the mother and breastfeeding behaviour. Secure attached mothers initiated breastfeeding more often and preferred breast- over bottle-feeding than insecure attached mothers. Mixed results were found for breastfeeding duration. Most studies had limitations regarding adequate sample size, and valid and reliable measurement of breastfeeding.

The findings provide some evidence that breastfeeding might contribute to child attachment security. Furthermore, maternal attachment style might play a role in breastfeeding behaviour. However, more prospective studies are needed to draw a solid conclusion.

With regard to breastfeeding recommendations, health professionals should be aware of and communicate towards clients that the effect of breastfeeding on child attachment might be rather small. Screening of maternal attachment style as a part of maternity practice could be useful to support insecure attached mothers experiencing breastfeeding difficulties.