Differential item functioning analysis on the Geriatric Depression Scale-15: An iterative hybrid ordinal logistic...

Area of Science:

• Geriatric Psychology
• Psychometrics
• Public Health

Background:

• Global increase in elderly population presents significant economic, social, and health challenges.
• Depression in late life is a prevalent issue requiring accessible screening tools.
• The 15-item Geriatric Depression Scale (GDS-15) is a common screening tool, but its measurement invariance is not well-established.

Purpose of the Study:

• To assess the measurement invariance of the GDS-15 across various demographic factors in a general population.
• To identify potential differential item functioning (DIF) within the GDS-15.
• To ensure the GDS-15's validity and reliability for diverse elderly populations.

Main Methods:

• A sample of 1473 participants from diverse Iranian ethnic groups completed the GDS-15 and demographic questionnaires.
• The lordif package in R (version 3.1.3) was utilized for differential item functioning (DIF) analysis.
• Items exhibiting uneven performance across demographic variables were identified.

Main Results:

• Significant DIF was observed for several GDS-15 items.
• Women, homebound patients, individuals with lower financial status, and those with a non-Persian mother tongue scored higher on certain psychological symptoms.
• These demographic groups showed higher scores compared to others with similar overall depression levels.

Conclusions:

• The GDS-15 demonstrates measurement bias, affecting its accuracy across different demographic groups.
• Psychologists should consider removing or replacing biased items before utilizing the GDS-15 for screening geriatric depression.
• Further validation studies are needed to ensure equitable and accurate depression screening in diverse elderly populations.