Factor structure of the Geriatric Depression Scale and measurement invariance across gender among Chinese elders

Abstract

Background: The Geriatric Depression Scale (GDS) is commonly used in research and clinical settings for screening of depression. The current study aimed to examine the best-fit factor structure model of the GDS among Chinese elders and to evaluate the measurement invariance of the GDS across genders.

Methods: Participants included 1,553 elderly residents from the Hunan, Shandong, and Beijing provinces. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were conducted in tandem to determine the structure of the GDS on a large scale. Multigroup CFA (N = 1553, 45.24% male, mean = 71.33 ± 8.06) was utilized to test the measurement invariance of the depressive symptom structure, which was generated by EFA and confirmed by CFA across gender.

Results: A three-factor model with 15 depression, 9 apathy and 4 vigor items presented the best fit indices. Measurement invariance of the new proposed model across gender was supported fully assuming different degrees of invariance.

Limitations: Our sample was entirely Chinese, and thus may not be representative of populations outside of China. Our results are based on a cross-sectional study, which did not take into consideration changes that may occur over time within individuals.

Conclusions: A three-factor model best fits the depressive symptom structure of the GDS among elderly Chinese, with measurement invariance across genders.

Keywords: Chinese elders; Depression; Gender differences; Geriatric Depression Scale; Measurement invariance.