Factor structure and measurement invariance of the Edinburgh Postnatal Depression Scale during the perinatal period: a longitudinal study of Japanese women.

This article focuses on examining the factor structure and measurement invariance of the Japanese version of the Edinburgh Postnatal Depression Scale (EPDS) across the perinatal period in a longitudinal study of Japanese nulliparous women. The study found that a 6-item, 3-factor model—comprising subscales of anhedonia, anxiety, and depression—originally proposed by Kubota et al., demonstrated better fit and scalar measurement invariance from late pregnancy through six months postpartum compared to other models and the full 10-item EPDS. While the 6-item model showed stability across time points, residual invariance was not confirmed, indicating some limitations in strict factorial invariance. The authors suggest that this shorter EPDS version is easier to use and more reliable for clinical and research screening of perinatal depression in Japanese women, though caution is advised when comparing mean scores over time due to incomplete invariance testing.