Prenatal Attachment Inventory: expanding the reliability and validity evidence using a sample of Italian women

Background: The Prenatal Attachment Inventory (PAI) is a widely used questionnaire to measure prenatal attachment. However, its internal structure is controversial. None of the previous studies have investigated the dimensions of the Italian version of the scale using both an exploratory and a confirmatory approach. There is also a need to explore extensively the external validity of the Italian PAI.Objective: We designed a study aimed at shedding further light on the dimensionality of the questionnaire and expanding the evidence of its reliability and validity. Method: Five-hundred and thirty-five Italian expectant mothers participated in the prenatal phase; a subsample of 100 women participated in the postnatal phase of the study. The PAI was administered together with other scales measuring maternal–fetal attachment, psychological wellbeing and relational variables. Results: The hypothesised relationships with external criteria were substantiated overall. The five dimensions described in the Italian study by Barone, Lionetti, and Dellagiulia also emerged from our factor analyses, with the exception of two items. Internal consistency was adequate for the total scale and for four of the five subscales. Conclusion: As the Fantasy subscale showed poor internal consistency, we advise against its use as an independent measure. However, when used as a global score, the PAI is a reliable and valid measure of prenatal attachment in Italian women. Thus, it can be used for research purposes. The use of the PAI could also be very helpful in clinical settings, in order to identify expectant mothers who have difficulty in establishing an affective bond with their unborn infants. To this end, further research should study the characteristics of the PAI on high-risk groups and clinical samples in order to obtain clinical cut-offs