The psychometrics of the Center for Epidemiologic Studies Depression Scale (CES-D) in Chinese Primary Care Patients

Abstract

INTRODUCTION: The aim of this study was to establish the factor structure, construct validity, reliability, sensitivity and responsiveness of the CES-D for Chinese patients.in primary care setting. METHOD: 3686 subjects were included in this prospective longitudinal study. The factor structure was evaluated by CFA. The internal construct validity was assessed by corrected item-total correlation. The convergent validity was assessed by examining the correlations between CES-D, PHQ-9 and SF-12 v 2 MCS. Instrument reliability was assessed by examining the internal consistency and 2-week test-retest reliability. Sensitivity was determined by calculating the AUC. The internal and external responsiveness was assessed. RESULTS: The data fitted the four-factor structure. Corrected item-total correlations were 0.4 for most items. The CES-D had a strong correlation with the PHQ-9 and SF-12 v2 MCS. The CES-D showed good internal consistency (Cronbach’s alpha coefficient = 0.89) and good test-retest reliability (Intraclass correlation coefficient: 0.91, paired t-test p-value >0.05). The CES-D was sensitive in detecting differences between groups, with the AUC >0.7. Internal responsiveness to detect positive and negative changes was satisfactory. The CES-D was externally responsive, with the AUC>0.7. DISCUSSION: The CES-D is valid and reliable. The instrument is sensitive enough to screen for depression and responsive enough to longitudinally monitor patients in primary care.