Screening für Demenzen bei Hochaltrigen: Eine Erprobung in Hausarztpraxen

Abstract

Background: The aim of the presented study was to investigate which dementia screening test differentiates best between a "noticeable demented" versus a "non-noticeable demented" population. Another aspect was to investigate which items/subtests of the screening method suit best to distinguish the above mentioned groups within the highly aged (> 75 years) range of the population.Methods: The study included the following screening tests: MMST, MMT, TFDD, KMS, DemTect, RDST and Clock- Test. The study was performed in the Psychiatric Department of the University Hospital Goettingen/Germany during the period of March 2003 till April 2004. A total of 90 patients aged above 75 years were examined. The average age of the probands was 80.5 years. Patients were recruited in nine different doctor"s surgeries and in the memory clinic of the University of Goettingen. They were examined based on a screening method that is composed from subtests of various common screening methods. After performing these tests the existence of dementia was assessed independently by a colleague of the memory clinic. At that point interviews and neuro-psychological tests were used to evaluate the different screening methods regarding different criteria. Taking all these test results in consideration the probands were finally classified in "non dement" and "minor demental syndrome" by the reference investigator.Results: All screening methods demonstrated a very good significance around 0.00, i.e. all of them are basically well suited to distinguish between these patient groups. However the DemTect and the MMT appear most suitable after considering the statistical significance and the effect size (Cohens-d-value). In addition to this both test methods demonstrated a limited education dependency with a correlation of 0.26 respectively 0.25. The KMS showed the highest education dependency. The investigation of the sub- tests/single items using regression analysis showed that the subtests "word fluency", "word list learning and delayed recall", "clock-test" and "Serial Seven" are most suitable for the group differentiation in very old age. On this basis a new test method was developed. This test method consists of the subtests "Serial Seven", "Shulman-Clock", "Word list learning and delayed recall" and "Word fluency animals".Conclusion: The development of this new screening method allows to diagnose dementia in the above 75 age group as most commonly used methods are not validated in the highly aged population. But exactly this group of patients aged above 75 years will provide the highest number of dementia patients in the future. Additionally this potential test method meets the criteria that define a good test method. The subtests did not show any effect of eduction. The individual items in this test analyse predominantly those areas which are often and early affected by dementia. These areas are disturbance of memory, aphasia, ability of abstraction and visual construction. The test showed a high acceptance by the patients. The test can be completed within 10 to 15 minutes and is therefore most suitable for the use in the daily routine of the general practitioner"s surgery. In addition to this the short duration of 10 to 15 minutes does not have a negative influence on the results due to lack of concentration.