Depression in dementia sufferers--comparison of diagnostic criteria.

Abstract

To the Editor: Depression occurs in 5 to 40% of dementia sufferers, causing distress to patients,1 detrimentally influencing cognition2 as well as instrumental activities of daily living,3 and reducing life expectancy.4 The information collected depends upon the criteria utilized. The Diagnostic and Statistical Manual of Disease, 3rd Edition Revised (DSM-III-R) and the Research Diagnostic Criteria (RDC) have been used in a number of studies,1 although neither was designed to evaluate depression in dementia. The Dementia Mood Assessment Scale and the Cornell Depression Scale (CDS) have been developed for this purpose. The CDS utilizes a standard list of symptoms that are assessed from informant and patient interview. No studies have examined the ability of clinical criteria to predict a persistent mood disorder. Consecutive dementia patients referred to old age psychiatry services and a memory clinic were assessed using the Geriatric Mental State Schedule, History and Aetiology schedule (HAS), the Cognitive Section of the Cambridge Assessment for Mental Disorders in the Elderly (CAMCOG), a physical examination, and a blood screen. Symptoms of depression were rated on the CDS with additional questions regarding duration of symptoms and disability. Depression was diagnosed according to RDC, DSM-III-R, and a CDS score > 9. Dementia was diagnosed according to the McKhann et al. criteria (NINCDS ADRDA), DSM-III-R, the Hachinski scale, the McKeith et al. criteria, and the Computerised Diagnostic Package linked to the HAS Schedule (HAS AGECAT). The method is described elsewhere.5 Informants were contacted at monthly intervals for 1 year to complete the CDS. Depression was diagnosed using the RDC. Depression had to be absent for 3 months to be considered resolved. Ninety per cent of patients participated. Eighty nine (71.8%) completed the follow-up. Mean age of patients was 79.7, and their mean CAMCOG score was 46.2. Sixty-three (70.8%) of the patients had probable or possible Alzheimer's disease, 15 (16.9%) had vascular dementia, and nine (10.1%) had dementia with Lewy bodies. Two had unspecified dementia. At the baseline assessment, 21 (16.9%) patients had DSM-III-R major depression, 30 (24.2%) had RDC major depression, and 33 (26.6%) scored more than 9 on the CDS. The Kappa values for the agreement between the systems are shown in Table 1. The mean CDS scores for the groups were similar (DSM-III-R, 13.0; RDC, 12.9; CDS, 13.2). Sixteen patients with DSM-III-R major depression completed the follow-up. Fourteen (87.5%) did not experience any RDC major depression, one experienced 1 month and one experienced 2 months of depression. Combining the number of months of RDC major and minor depression, only two (12.5%) of the patients experienced 6 or more months of depression. Of the 15 resolved cases, six (40%) experienced a recurrence. Twenty-two patients with RDC major depression at baseline completed the follow-up. Nineteen (86.4%) did not experience any RDC major depression, two (9.1%) experienced 1 month, and one (4.6%) experienced 2 months. Combining RDC major and minor depression, three (13.6%) of the patients experienced 6 or more months of depression. Six (30%) of the 20 resolved cases experienced recurrence. Twenty-six patients with CDS score greater than 9 at baseline completed the follow-up. Twenty-three (88.5%) did not experience any months of RDC major depression, one (3.8%) experienced 1 month, and two (7.7%) experienced 2 months. Combining RDC major and minor depression, five (19.2%) experienced 6 or more months of depression. Seven (30.4%) of the 23 resolved cases experienced a recurrence. The sample is representative of dementia patients in clinical settings. The patients were assessed with a standardized battery of instruments and were followed-up monthly for 1 year. The DSM-III-R criteria were the most restrictive showing a prevalence of major depression of 16.9% compared with a prevalence of 24.2% for RDC major depression and 26.5% for a CDS depression. Although more restrictive, they do not identify a group of patients with more severe or more persistent depression. Patients identified according to each set of criteria had similar CDS scores and were unlikely to experience any further major depression. A similar proportion experienced further minor depression. There is a remarkable similarity between the severity and outcome of depression identified using the different criteria. It is clear that none of them identify meaningful cases of depression from a cross sectional assessment. Perhaps a minimum duration of symptoms needs to be specified before a diagnosis of depressive disorder is made. Three months may be a suitable period.