A single Spanish version of maternal and paternal postnatal attachment scales: validation and conceptual analysis.

Background: Postnatal bonding constitutes a major process during the postpartum period, and there is evidence that bonding difficulties have negative consequences for parents' mental health and the child's development. However, the conceptualization of postnatal bonding presents inconsistencies, as well as problems in having instruments that encompasses the father figure. The objective was to adapt the maternal postnatal attachment scale (MPAS) and the paternal postnatal attachment scale (PPAS) to Spanish, to evaluate its validity and reliability and to analyze the construct dimensionality of both questionnaires from a gender perspective.

Methods: Instrumental design. In 2016-2017, a sample of 571 mothers and 376 fathers, with children between 6 and 11 months of age, responded to the Spanish version of MPAS and PPAS, respectively. After a process of translation-back-translation of the instrument, we empirically analyzed the internal consistency (Cronbach alpha, composite reliability (CR)) construct and concurrent validity (with regard to postpartum depression and dyadic adjustment). Additionally, we studied the instrument's content validity, using the Delphi methodology; and the differential analysis in both samples (mothers and fathers), examining the invariance.

Results: A short version of 15 items was obtained, common for mothers and fathers. The results of the Delphi methodology showed a 100% inter-judge agreement, highlighting the absence of differences in the adequacy of the items as a function of the parents' gender. Confirmatory factor analysis showed a good fit of three original factors proposed by the authors. The global Cronbach alpha coefficients in the total sample were adequate (mothers, 0.70; fathers, 0.78); and Cronbach alpha of each dimension in the case of mothers was 0.50 (Quality of bonding), 0.55 (Absence of hostility), and 0.60 (Pleasure in interaction); in the case of fathers, it was respectively 0.54, 0.64, and 0.72. CR of each dimension were: quality of bonding, 0.74 in mothers and 0.80 in fathers; absence of hostility, 0.93 in mothers and 0.94 in fathers; pleasure in interaction, 0.83 in mothers and 0.90 in fathers. With regard to the analysis of group invariance, the results revealed empirical evidence of configural and metric invariance. Concurrent validity showed moderate negative correlations for postnatal depression (mothers, r = -0.41, p < 0.001; fathers, r = -0.38, p < 0.001), and positive correlations for dyadic adjustment (mothers, r = 0.39, p < 0.001; fathers, r = 0.44, p < 0.001).

Discussion: A new version of the instrument was generated, with good psychometric properties, adequate for use both with mothers and with fathers. This scale helps evaluate postnatal maternal and paternal bonding, allowing to study it from within the family system, a necessary step forward to advance perinatal mental health.