Psychometric Comparisons of Three Depression Measures for Patients With Stroke

Abstract

Abstract Importance: Psychometric examinations for patients with stroke remain insufficient. The Center for Epidemiologic Studies Depression Scale (CES–D), Beck Depression Inventory–II (BDI–II), and Geriatric Depression Scale (GDS) are promising outcome measures. Objective: To examine and compare the reliability and validity of three depression measures in a sample of patients with stroke. Design: Repeated-measures design. Setting: A hospital in southern Taiwan. Participants: Fifty-nine outpatients, who completed three depression measures. Outcomes and Measures: Cronbach’s α and intraclass correlation coefficients (ICCs) were used to examine the internal consistency and test–retest reliability, respectively, of the three measures. An independent-samples t test was conducted to compare two groups of patients with different levels of disability to investigate discriminative validity. Pearson’s rs were calculated among the three measures to examine concurrent validity. Results: The three measures had good internal consistency (α = .85–.92) and sufficient test–retest reliability (ICC = .84–.91). The minimal detectable change (percentage of minimal detectable change) was 10.6 (63.3%), 13.5 (98.3%), and 5.8 (49.9%) for the CES–D, BDI–II, and GDS, respectively. There was a statistically significant difference between the two groups in CES–D score (p = .032) and no significant differences on the other two measures (p = .095–.187). The correlations among the three measures ranged from .79 to .89. Conclusion and Relevance: All three depression measures had sound internal consistency, test–retest reliability, and concurrent validity in patients with stroke. What This Article Adds: Of the three measures, the CES–D had better discriminative validity, and the GDS demonstrated greater reliability and smaller random measurement error in patients with stroke.