O2-07-02: Is the geriatric depression scale a reliable screening tool for depressive symptoms in elderly patients with cognitive impairment?

The Geriatric Depression Scale (GDS) was developed to rate depression in elderly. Its major advantages are that it is simple and fast (max. 10 min) to administer and that it does not require a trained interviewer. However, its reliability and validity in patients with cognitive impairment remains a matter of debate. In order to determine the reliability and validity of the 30-item GDS for the assessment of depressive symptoms in Mild Cognitive Impairment (MCI) and probable Alzheimer's disease (AD) as compared to the Cornell Scale for Depression in Dementia (CSDD), we set up a prospective study. Diagnosed according to strictly applied diagnostic criteria, patients with MCI (n=156) and AD (n=247) were enrolled. At inclusion, depressive symptoms were assessed by means of the GDS and the CSDD. Patients underwent a neuropsychological examination that consisted of amongst others a Mini-Mental State Examination (MMSE) according to which the AD group was subdivided in mildly (MMSE≥18) (n=117), moderately (MMSE18 and ≥10) (n=89) and severely affected (MMSE 10) (n=38) AD patients. As a CSDD total score of 8 or more suggests significant depressive symptoms (Burns et al., 2004), this cut-off score was applied to dichotomize the patients included for ROC curve analysis. In MCI, moderate but highly significant correlations were found between GDS and CSDD scores (Spearman: r=0.615; P 0.001). In mildly (r=0.313; p 0.001), moderately (r=0.229; p=0.031) and severely (r=0.336; p=0.039) affected AD patients, only weak correlations between GDS and CSDD scores were calculated. ROC curve analysis showed that sensitivity and specificity values of respectively 95% and 67% were achieved when a GDS cut-off score of 8 was applied in MCI patients. In AD patients, low sensitivity and specificity values did not allow selecting an optimal cut-off score by means of ROC curve analysis. The GDS is a reliable screening instrument for depressive symptoms in MCI patients. Using a cut-off score of 8, sensitivity values of 95% were achieved.