Psychometric evaluation of the Mother-to-Infant Bonding Scale in a Flemish context in early post-partum mothers.

INTRODUCTION

A strong mother-infant bond supports healthy development and leads to better long-term outcomes for both mothers and infants. Tools like the Mother-to-Infant Bonding Scale (MIBS) enable objective assessment of this bond. This study examined key psychometric properties of the Dutch MIBS in a Flemish population.

METHODS

This study comprised two subsequent parts. First, face and content validity were assessed by an expert panel of healthcare providers (n = 17) and mothers (n = 6). Second, convergent and concurrent validity were assessed based on cross-sectional data collection. Internal consistency was evaluated using Cronbach's alpha. Exploratory factor analysis (EFA) was performed. Evidence based on relationships with demographic and clinical variables was examined to support the validity of the Dutch MIBS.

RESULTS

Data were collected from 128 mothers who completed the MIBS and Postpartum Bonding Questionnaire (PBQ) on day 2 postpartum. Experts reached consensus on the face validity of the Dutch version of the MIBS in a Flemish context. Items showed adequate Item Content Validity Index (I-CVI) (≥0.79), except for the "resentful" and "aggressive" items. The scale achieved a Scale Content Validity Index/Average (SCVI/Ave) of 0.89, although internal consistency was moderate (α = 0.66). MIBS and PBQ scores correlated moderately (Kendall's Tau T = 0.27, P < 0.001). Fisher's exact test showed a significant difference in the proportion of mothers classified with disrupted bonding depending on the tool (P = 0.015). Combined feeding was associated with a higher rate of disrupted bonding (P = 0.007).

CONCLUSION

This study found that the Flemish MIBS tool had acceptable content validity, except for "resentful" and "aggressive." Reliability was moderate and convergent validity with PBQ was limited, with substantial discordant classification of disrupted bonding. The superiority to the PBQ-tool could not be demonstrated.