The Geriatric Depression Scale and the Cornell Scale for Depression in Dementia. A validity study

Abstract

The study is a validation study of two psychogeriatric depression rating scales, The Geriatric Depression Scale (GDS) and the Cornell Scale for Depression in Dementia (CSDD). The sensitivity and specificity, and the convergent and criterion validity of the two scales as well as the inter-rater reliability of the CSDD are reported. Two independent clinicians using the ICD-10 for depression and dementia, the Clinical Global Impression (CGI), the Hamilton Depression rating scale 17-items and the Mini-Mental-State Examination (MMSE), interviewed each patient or control subject. One hundred forty-five persons of 65 years or more of age were included, 73 were depressed only, 36 depressed and demented; 36 persons were control subjects, 11 of these were demented. The inter-rater reliabilities were high or very high equalling perfect correlation. There was very high convergent validity between the screening tools and the severity scales; the shorter versions of the GDS (15-, 10- or four-item version) had lower though still almost perfect correlations. The criterion validity in the total population showed the CSDD as the better scale with sensitivity and specificity of 93% and 97% with a cut-off value of > or =6. The GDS versions had sensitivities and specificities ranging from 82% to 90% and 75% to 94% respectively with cut-off values > or =9, 4, 3 and 1. The CSDD retained its validity and specificity as a screening tool for depression in a population of demented, while the GDS versions all diminished in validity. The GDS and the CSDD are both valid screening tools for depression in the elderly; however, the CSDD alone seems to be equally valid in populations of demented and non-demented.

A Comparison of the Validity of the Cornell Scale and the MADRS in Detecting Depression among Memory Clinic Patients

“…The mean score of 8 in the group of patients with minor depression in that study has been used as a cutoff point. Kørner et al [25] reported a very high sensitivity (0.93) and specificity (0.97) with a cutoff point of 6 in a group of outpatients including patients both with and without dementia, and the AUC was 0.98. In this study, spouses of the patients were used as controls, which according to Sackett’s definition, is not a valid contrast group [31].…”

“…Some studies have shown it to be valid in patients both with and without dementia [25,36], while others have found it incapable of differentiating between depressed and not depressed patients [37]. The cutoff points with the best sensitivity and specificity differ, and probably depend on cultural differences, patient populations and different study designs.…”

“…The CSDD has been validated in different clinical settings among cognitively impaired patients [25,26], and it is presumed to be suitable for both patients with and without dementia [22,27]. To our knowledge, the MADRS has not been comprehensively validated in cognitively impaired patients.…”

Serum Thyroid-Stimulating Hormone Levels and Frailty in the Elderly: The Progetto Veneto Anziani Study

“…Disability was evaluated with the number of activities of daily living (ADL) preserved using the index proposed by Katz et al 12 , while cognitive status and depressive mood with 30-items minimental state examination (MMSE) 13 and geriatric depression scale (GDS) 14 , respectively.…”

Implementation of a brief anxiety assessment and evaluation in a Department of Veterans Affairs geriatric primary care clinic

“…GDS-15 scores range from 0 to 15, with higher scores indicative of greater depression severity. Studies provide support for internal consistency, reliability, and convergent validity with other measures of depression and discriminant validity [25]. Adequate internal consistency was found for the GDS-15 (KR-20 = 0.67) in the present sample of older VA patients.…”