Simplified Screening for Depression in Acutely Hospitalized Geriatric Patients: Comparison of the Two-Item Whooley Questions With the Geriatric Depression Scale-15

Abstract

Objective

The aim of the study is to investigate sensitivity and specificity of the two-item Whooley questions compared to the 15-item Geriatric Depression Scale (GDS-15) for the detection of depressive symptoms in acutely hospitalized geriatric patients.

Methods

Patients were prospectively recruited. Two-item Whooley questions and GDS-15 were performed within 24 h of admission, the GDS-15 being part of the routine geriatric assessment. Montgomery-Asberg Depression Rating Scale (MADRS) served as gold standard in the form of a structured interview which was performed within 48 h of admission. In addition, all patients were examined for independence (Barthel Index), cognition (mini-mental status examination), vision and hearing (finger rub test) and multimorbidity (Charlson Comorbidity Index).

Results

248 patients were recruited. Median (Q1/3) age was 83 (/3 79/86) years, 157 patients (63%) were female. Whooley questions had a sensitivity of 0.95 (95% CI: 0.81–0.99) and specificity of 0.49 (95% CI: 0.43–0.56) identifying moderate depressive symptoms (MADRS ≥ 20 points) with a negative predictive value (NPV) of 0.99. This is compared to a sensitivity of 0.57 (95% CI: 0.37–0.75) and specificity of 0.62 (95% CI: 0.56–0.68) using GDS-15 to identify moderate depressive symptoms. Both functional impairment and cognitive deficits had an impact on the result of GDS-15, but did not influence the result of Whooley questions.

Conclusions

Compared to the GDS-15, the two-item Whooley questions are more accurate to screen for symptoms of depression in acutely hospitalized geriatric patients.