Assessing Longitudinal Invariance of the Center for Epidemiologic Studies-Depression Scale Among Middle-Aged and Older Adults.

Abstract

The longitudinal invariance of the Center for Epidemiologic Studies-Depression (CES-D) scale among middle-aged and older adults is unknown. This study examined the factorial invariance of the CES-D scale in a large cohort of community-based adults longitudinally. 1,204 participants completed the 20-item CES-D scale at 4 time points 1 year apart. Structural equation modeling was used to identify best fitting model using longitudinal data at baseline and at 1-, 2-, and 3-year follow-up. The 4-factor model showed partial invariance over 3 years. Two of the 6 noninvariant items were consistently noninvariant at the 3 follow-up points. Special consideration should be given to these 2 items when using the CES-D scale in healthy adults (45-75 years old).