Longitudinal measurement invariance of EURO-D scale across 27 countries in SHARE wave 8 and wave 9: A cross-country alignment study

Abstract

Background: The EURO-D is a widely used measure of 12 depressive symptoms among middle-aged and older adults in Europe. The present study aims to evaluate the psychometric properties and measurement invariance of EURO-D across gender, age, time, and countries. Methods: This study utilized population-representative samples of participants aged 50 and above from Waves 8 and 9 (N = 52,369–67,744, Mean age = 69.9, 57 % female) of the Survey of Health, Aging, and Retirement in Europe (SHARE) across 27 countries. The factorial validity, reliability, and convergent validity of EURO-D were assessed using confirmatory factor analysis. The alignment approach was used to examine the longitudinal measurement invariance of EURO-D across countries in Waves 8 and 9. Results: The modified 1-factor model, with two residual correlations, provided a parsimonious and acceptable fit in both Waves 8 and 9. EURO-D demonstrated exact scalar invariance across gender, age groups, and time points, and was significantly associated with greater loneliness, functional limitations and lower cognitive function. Across countries, EURO-D showed approximate scalar invariance at Wave 8 and Wave 9, with 30.2 % and 30.4 % of measurement parameters being non-invariant, respectively. Variations were observed at the country level in the temporal changes in the aligned EURO-D factor mean. Conclusions: The findings support satisfactory psychometric properties and measurement equivalence of EURO-D across gender, age groups, time points, and countries. These results have important clinical implications for ensuring cross-cultural comparability in aging populations in future cross-national research using the SHARE dataset.