[Development and validation of a test for early diagnosis of dementia with differentiation from depression (TFDD)]

Psychometric tests used for the early detection of dementia often are seen as too difficult or too complex. Classical neuropsychologic tests were not developed for this purpose. Sensitivity and specificity to discriminate "healthy" vs. "ill" are low. For measuring both dementive and depressive symptoms, so far no test has been published. The objective of this study was to develop a sensitive and specific test for dementia that is easy to administer and to evaluate. Moreover, it should discriminate dementia from depressive pseudodementia. With respect to former studies, items were selected that recognized patients in the beginning of the disease. Additionally, depressive symptoms were rated. With the items for dementia, 88 patients with dementia of the Alzheimer type, 52 patients with depressive disorder and 37 healthy elderly controls were investigated. In this group of already diagnosed patients, the test reached a sensitivity and specificity of 100 percent (healthy elderly controls vs. patients with Alzheimer's disease: n = 125, U = 0, p < 0.001; patients with depressive disorder vs. patients with Alzheimer's disease: n = 140, U = 0, p < 0.001; healthy elderly controls vs. patients with depressive disorder: n = 89, U = 485.5, p < 0.001). For the dementia items, the inter-rater-reliability was rs = 0.996 (p < 0.001, n = 18), for the depression items it was rs = 0.753 (n = 18, p < 0.001). The test-retest-reliability was rs = 0.868 (p < 0.001, n = 35) for the dementia items and rs = 0.7 (n = 8, p < 0.05) for the depression items. These validation data will make the test useful for practitioners. Its ability to discriminate patients suffering from dementia of the Alzheimer type from healthy controls is comparable to tests consuming more time.