Applying Rasch methodology to examine and enhance precision of the Edinburgh Postnatal Depression Scale

Area of Science:

• Psychometrics
• Mental Health Assessment
• Statistical Modeling

Background:

• The Edinburgh Postnatal Depression Scale (EPDS) is widely used but has limited precision due to ordinal scaling.
• Rasch methodology offers potential to enhance the psychometric properties of depression scales.

Purpose of the Study:

• To apply Rasch methodology to improve the psychometric properties of the EPDS.
• To develop an algorithm for converting ordinal EPDS scores to interval scores.

Main Methods:

• The Partial Credit Rasch model was used with data from 621 mothers.
• Local dependencies between EPDS items were addressed by creating super-items.

Main Results:

• Model fit was achieved by combining six locally dependent items into three super-items.
• The modified EPDS demonstrated strong reliability (PSI = 0.86) and unidimensionality.
• Ordinal-to-interval conversion algorithms were generated.

Conclusions:

• The EPDS can meet Rasch model expectations with internal modifications.
• Ordinal-to-interval conversion enhances the precision and statistical conformity of EPDS scores.
• This improves the accuracy of monitoring clinical status and outcomes in mothers.