| Abstract No.: | C-C3104 |
| Country: | Canada |
| | |
| Title: | ASSESSMENT OF COGNITIVE IMPAIRMENT IN AGING USING THE AD8 |
| | |
| Authors/Affiliations: | 1 Lisa Koski*;
1 McGill University, Montreal, QC, Canada
|
| | |
| Content: | Objectives: The AD8 is a brief informant interview designed to aid in the detection of cognitive impairment and dementia in general practice. Our objectives with this study were to evaluate psychometric properties of the AD8 using Rasch analysis, compare its performance in English and French-language versions, and assess its relation to performance-based measures of cognitive impairment in an outpatient geriatric cognitive disorders clinic.
Methods: 63 cases administered the AD8 from October 2007 to Feb 2008 were identified and extracted, with identifying information removed, from the clinical database at two geriatric cognitive disorders clinics within the McGill University Health Centre. Other data extracted included age, sex, education, language of testing, diagnosis and clinic site, as well as individual item responses for the Montreal Cognitive Assessment and the Mini-Mental Status Exam.
Results: The AD8 and its individual items fit a unidimensional Rasch model of the construct functional cognitive decline (Chi-square P> 0.1). The item hierarchy showed acceptable reliability at 0.78. Consistent problems with memory and thinking represent the most commonly-reported symptom that is reported even by those with the least impairment, whereas disorientation to month or year and problems learning to use a tool/gadget were the least commonly reported and represented the greatest level of impairment. The English and French versions of the AD8 yielded similar hierarchies, demonstrating equivalence in the response characteristics of the two versions. The items on the AD8 span a narrow window of the total range of functional cognitive decline (from –1.0 to +1.3 logits or standard deviations). In fact, 22% of respondents obtained the maximum score possible for the scale, indicating that precise measurement of decline is not possible in this group. Compared with spouse-informants, higher scores representing greater functional decline were given by children of the patient or other family members as informants. The AD8 correlated at 0.70 with the Mini Mental Status Exam and at 0.55 with the Montreal Cognitive Assessment.
Conclusions. The AD8 can be used as a quantitative indicator of cognitive decline in less impaired patients but shows ceiling effects (is too sensitive) in more impaired individuals. The AD8 is being introduced into geriatric clinics as a tool for standardizing the evaluation of cognitive decline and provides information that is related but complementary to that obtained from performance-based measures of cognitive status.
|
| | |
| Back |
|
