Assessing depression in Greek dementia patients: Which scale to use?

Depression in dementia is known to deteriorate patients' cognitive function and Quality of Life and to increase the burden of care. Although detecting depression in dementia is crucial, there is no gold standard for its screening and diagnosis. We examined the psychometric properties of 3 different scales in detecting depression in dementia. Results will be useful as community services for dementia in the country are developing and the need for reliable detection of depression in dementia patients is urgent. Our sample consisted of 136 Greek dementia patients who consulted a memory clinic. For the diagnosis of depression, DSM-IV criteria for major depression and 3 different depression measures were used: a self-assessment scale (Geriatric Depression Scale; GDS), a caregiver assessment scale (Neuropsychiatric Inventory-Depression; NPI-D) and a clinician rated scale (Cornell Scale for Depression in Dementia; CSDD). For the evaluation of the screening performance of the three depression scales receiver operating characteristic curve (ROC) analysis was applied. The DSMIV criteria served as the gold standard method for the diagnosis of major depression. CSDD showed the best psychometric properties for the diagnosis of depression in dementia. The ROC curve analysis revealed that among the three measures, the CSDD had the wider AUC (0.919), second in the width of the AUC was the GDS (0.871), and last was the NPI-D (0.812). The prevalence of depression ranged from 18.4% according to DSM-IV criteria to 42.6% using the NPI-D. Using the GDS (cut off point: 7/8) and the CSDD (cut off point: 6/7), depression was present in 26.9% and 33.1% of the patients, respectively. Correlations between scales used were significant (r from 0.432 to 0.660; p<0.001). Caregivers tend to report more depressive symptoms in dementia compared to patients' and clinicians' ratings. CSDD should be used in specialized centers, but GDS may be an alternative in patients able to complete the assessment. The need to establish valid criteria for the diagnosis of depression in dementia is urgent.

The burden of caring for patients with dementia and its predictors.

Dementia not only affects patients but also care providers. The assessment of Caregivers' Burden (CB) has grown exponentially in the dementia field, as studies have shown that it is higher in dementia than in other diseases. Dementia care in Greece is different compared to other developed countries, as most of the patients receive care at home from family members. The aim of the present study was to examine the level of burden in Greek caregivers who live in Athens, and its association with patient and caregiver factors. This was a cross-sectional study of 161 primary caregivers of dementia patients living in the community and attending a secondary clinic. CB was assessed with the Zarit Burden Interview (ZBI) and caregivers' depression with the Center for Epidemiological Studies Depression Scale (CES-D). Clinical characteristics of the patients were also assessed using validated scales (cognitive status, functional ability, neuropsychiatric symptoms). In order to find predictors of caregiver Burden, we conducted a 3-step hierarchical regression analysis. Most patients were suffering from Alzheimer's Dementia (n=101; 62.73%) and had moderate and severe dementia according to the MMSE score (mean MMSE=11.50), with patients being unable to perform 2 basic activities of daily living on average. 45 patients (27.95%) had depression according to the CSDD; only 5 patients didn't have any behavioral problem in the NPI, while patients had more than 5 behavioral problems on average. Caregivers were involved in their role for 3.6 years on average and the mean weekly caregiving time was more than 70 hours. Nearly half (n=80; 49.06%) of the 161 caregivers demonstrated high CB (ZBI>40) and nearly one fourth had depression according to the CES-D scale. All blocks of variables entered into the regression model independently predicted caregiver burden's variance (demographics, clinical factors and caregiving characteristics). Lower caregiver's age, high behavioral symptoms of dementia patients and caregivers' depression were found to be independently associated with CB. The final regression model explained 47.2% of the variance in CB. Dementia causes a great burden in caregivers. CB is a complex issue that is associated with several patients and caregivers' factors. The level of CB should be assessed in everyday dementia clinical practice.