Psychometric Properties of The Persian Version of The Prenatal Attachment Inventory in Pregnant Iranian Women

BACKGROUND In 1993, Muller developed the Prenatal Attachment Inventory (PAI) which has been used widely in many studies and translated into several languages. The current study aimed to translate the PAI into Persian, assess the underlying structure of the PAI, and the appropriateness of the one-factor solution proposed by Muller. MATERIALS AND METHODS In this cross-sectional study, we recruited a total of 322 primi- gravidae in their 27(th) to 34(th) gestational weeks that referred to private and governmental prenatal clinics in Tehran, Iran. All participants completed the Persian versions of the PAI and a demographic questionnaire. Participants were re-tested 2 weeks after the initial test- ing. The following psychometric properties of the PAI were investigated: construct validity by confirmatory factor analysis (CFA), internal consistency reliability with Cronbach's alpha, and test-retest reliability according to the intraclass correlation coefficient (ICC). RESULTS The CFA results indicated that a single-factor model provided good fit to the data, which confirmed the original model by its developer. The Cronbach's alpha coef- ficient for PAI was 0.856 and the test-retest reliability with ICC was 0.784. Consider- ing the duration between marriage and pregnancy, women with low duration scored significantly higher than women with high duration on PAI (P=0.043). CONCLUSION The Persian version of the PAI showed that one factor structure was ad- equate and could be used for measuring psychological affectionate attachment between Iranian mothers and their fetuses.