Format

Send to

Choose Destination
See comment in PubMed Commons below
Int J Geriatr Psychiatry. 2000 Sep;15(9):824-30.

Validation and comparison of three brief depression scales in an elderly Chinese population.

Author information

  • 1Department of Geriatric Medicine, Tan Tock Seng Hospital, Singapore. joy-lim@hotmail.com

Abstract

The study aimed to establish the diagnostic accuracy of the Geriatric Depression Scale (GDS), the Even Briefer Assessment Scale for Depression (EBAS DEP), and the single question test for depression in our elderly Chinese population, and to determine if any one instrument was to be preferred. Ninety-eight community-living, socially active and non-depressed elderly and 75 patients diagnosed with depression were administered the three depression scales. Receiver operating characteristics (ROC) were employed to determine the optimal cut-off scores for the GDS and EBAS DEP, and the diagnostic performance of all three instruments were then compared. ROC analysis indicated an optimal cut-off score of 4 and above for the 15-item GDS, with a sensitivity of 84.0% and a specificity of 85.7%, while the EBAS DEP had 77.3% sensitivity and 89.8% specificity at the optimal cut-off score of 3 and above. The sensitivity and specificity of the single question were 64.0% and 94.9%, respectively. The non-parametric test of the areas-under-the-curve showed no significant difference between the diagnostic performances of the GDS and the EBAS DEP; visually, however, the ROC plot of the GDS was superior. The GDS, the EBAS DEP, and the single question were all valid screening tools for depression in the elderly Chinese population. For busy physicians, there is rationale to first use the single-question test, supplemented where necessary with either the GDS or the EBAS DEP, as an efficient diagnostic strategy for identifying depression amongst older Chinese patients.

Copyright 2000 John Wiley & Sons, Ltd.

PMID:
10984729
[PubMed - indexed for MEDLINE]
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Loading ...
    Write to the Help Desk