Evaluation of the Edinburgh Post Natal Depression Scale using Rasch analysis.

Area of Science:

• Psychometrics
• Mental Health Assessment
• Perinatal Psychiatry

Background:

• The Edinburgh Postnatal Depression Scale (EPDS) is widely used but its validity as a single scale is questioned.
• Concerns exist that the EPDS may measure both depressive feelings and anxiety separately.

Purpose of the Study:

• To evaluate the psychometric properties of the Edinburgh Postnatal Depression Scale (EPDS) using Rasch measurement models.
• To determine if the EPDS is a unidimensional scale for measuring depression or if it comprises distinct factors.

Main Methods:

• Rasch analysis was applied to data from 324 primiparous women in Melbourne.
• The analysis tested for category ordering, item bias (Differential Item Functioning - DIF), and unidimensionality.
• The EPDS was revised by removing items that did not fit the Rasch model.

Main Results:

• The original 10-item EPDS showed a lack of fit to the Rasch model (chi-square = 82.8, df = 40, p < .001).
• Removing items 7 and 8 resulted in an 8-item scale (EPDS-8) that fit the Rasch model (non-significant Item-Trait Interaction).
• The revised EPDS-8 demonstrated no DIF, supported unidimensionality, and maintained case identification with revised cut-points.

Conclusions:

• The original 10-item EPDS is not ideal for unidimensional depression measurement.
• A revised 8-item EPDS (EPDS-8) provides a more psychometrically robust scale for assessing postnatal depression.
• The EPDS-8, with revised cut-points, accurately identifies cases consistent with the original EPDS-10.