While there are many self-report depression measures that have been used with older adults (e.g., CES-D, Zung, Beck), the Geriatric Depression Scale is by far the one with the most applicability to older persons with dementia. Historically, the usefulness of the GDS has been called into question. Burke, Houston, Boust, and Roccaforte (1989) and Kafonek et al. (1989)
reported that the clinical utility of the GDS was poor in long-term care residents with dementia. Lesher and Berryhill (1994) and Parmelee, Katz, and Lawton (1989) provided evidence that the GDS was both internally consistent and clinically valid in individuals with mild to moderate dementia.
Two more recent studies have provided further support for the importance and validity of using the report of persons with dementia in any evaluation of mood. In a study mentioned earlier (Espiritu et al., 2001), dementia patients' self-report of depression was a unique and significant predictor of the caregiver reports of patients' IADL scores. These findings held even when the sample was further divided into the more severely cognitively impaired (MMSE < 15). Perhaps the most intriguing findings about the GDS are the ones made by Mast (2002). The internal consistency of the GDS was examined among 580 geriatric patients with four levels of cognitive impairment as defined by the Mattis Dementia Rating Scale (Mattis, 1988): cognitively intact (DRS = 126+), mild cognitive impairment (DRS 103-125), moderate impairment (DRS 90-102), and severely impaired (DRS < 90). Surprisingly, the internal consistency did not decline as cognitive impairment increased. Conbach's alpha estimates were as follows: .86 (intact), .87 (mild), .87 (moderate), and .88 (severe). These findings support the reliability of the GDS among patients with severe cognitive impairment.
Chemerinski et al. (2001) investigated how well Alzheimer's patients themselves could report their own mood. Patients with dementia ( both with and without depression) had a mean MMSE score of 19, although the standard deviation was large (6). The depressed patients (as diagnosed by psychiatrists) themselves rated their own mood as significantly worse than those persons with dementia but without depression (t = 6, p < .00001). This finding is consistent with Mast's (2002) and Espiritu et al.'s (2001) finding that self-report depression information offers unique and valuable clinical information.
• Self-report information concerning mood states is generally valid in persons with dementia.
• Family caregiver reports of depression in persons with dementia are typically valid, even when caregiver is depressed.
• It is recommended that both informant ratings/observer scales and self-report information on depression be gathered and used.
Was this article helpful?