Supplementary Material

No data


Background: To estimate the effectiveness of commonly used depression scales and develop an improved strategy for depression screening among Chinese elderly patients for non-psychiatric geriatricians.

Methods: A cross-sectional study of 458 elderly inpatients aged over 65 yearswas conducted in a Beijing geriatric ward. Whooley questions, Zung Self-rating Depression Scale (SDS), and Geriatric Depression Scale-15 (GDS-15)were performed by trained geriatricians. Diagnosis of depressive disorderswas made following Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5) criteria. We calculated the area under the curve (AUC), sensitivity, specificity, positive and negative predictive value (PPV, NPV) of three depression scales when implemented alone, and in combination.

Results: The GDS-15 common cutoff point was 5, with an AUC of 0.645, sensitivity of 77.5%, specificity of 64.4%, and PPV of 83.2%. Zung SDS yielded an AUC of 0.660, with a sensitivity of 63.7%, specificity of 68.9%, and PPV of 82.3% at the standard cutoff index of 50. At least one positive answer to Whooley questions achieved an AUC of 0.611, a sensitivity of 57.8%, specificity of 64.4%, and PPV of 78.7%. The two-step approach of Whooley questions and GDS-15 showed better performance than another approach of Whooley questions and Zung SDS, by comparing the AUC (0.672 vs. 0.642).

Conclusions: GDS-15 is a routine tool for screening depression among Chinese elderly inpatients. A two-step approach using Whooley questions and GDS-15 can improve the recognition of depressed patients for non-psychiatric geriatricians.