BEDSIDE ASSESSMENT OF DEMENTIA TYPE USING THE QUALITATIVE EVALUATION OF DEMENTIA

Problem: We present a novel dementia assessment instrument, the Qualit ative Evaluation of Dementia (QED), designed to discriminate dementia type at the bedside. The QED is a brief, clinically based checklist wh ich operationalizes the approach of a geriatric psychiatrist to the qu alitative assessment of dementing illnesses. Scores range from 0 = pur e ''subcortical'' illness, to 30 = pure ''cortical'' disease. Internal consistency (Chronbach's alpha = .69) and interrater reliability (r = .93) are acceptable. Blinded raters agree on dementia type in > 90% o f cases. When QED scores are mapped against measures of general cognit ive function a qualitative picture of dementia typology emerges. Metho d: The QED's ability to discriminate NINCDS ''Probable'' Alzheimer's d isease (AD) from those with ''no dementia'' and ''dementia without cor tical features'' was assessed in a cross-sectional sample of 118 conse cutive patients presenting to a multidisciplinary geriatric assessment clinic and consultation service. Results: The QED accurately discrimi nated air three study groups. When combined with the Executive Intervi ew (EXIT), 87.7% of subjects were correctly classified. Misclassificat ion rates ranged from 11.8% for ''Probable'' AD to 13% for ''no dement ia.'' Substituting the Mini-Mental State Examination (MMSE) correctly classified 74% of subjects, with misclassification rates ranging from 11.8% for ''Probable'' AD to 40.0% for ''dementia with no cortical fea tures.'' The EXIT was more sensitive than the MMSE to mild cognitive i mpairment, and subcortical dementia. Conclusions: These results sugges t that dementia subtypes can be accurately and reliably assessed using a combination of the QED and either the EXIT or the MMSE. This evalua tion can be completed in 20 minutes, and requires no access to brood o r laboratory work. The use of the EXIT in this assessment reduces misc lassification rates and improves sensitivity in early stages. The impl ications of this technique for clinical assessment, subject recruitmen t, and the conceptualization of dementia are discussed.