Abstract
INTRODUCTION: Emergency department (ED) screening for cognitive impairment (CI) is crucial for early intervention, yet guidance on dementia screening protocols remains limited. Our objective was to assess the concordance between the Ascertain Dementia 8 (AD8) tool, a brief screening tool for CI, administered to ED patients and their informants. METHODS: We conducted a prospective observational study of dyads, including patients ≥65 years without a diagnosis of dementia seeking care in the ED and their informants (e.g., family, friend with close contact). Trained research assistants used the 4AT to exclude patients with concern for delirium. The AD8 was then administered to blinded patients (pAD8) and informants (iAD8), with scores ranging from 0 to 8 and scores of ≥2 indicating CI. We used the intraclass correlation coefficient (ICC) to calculate the level of agreement between AD8 scores. RESULTS: Our analytic sample included 538 dyads, of which 63.3% of patients were female with a mean age of 73.5 years. A total of 131 (24.3%) patients without a diagnosis of dementia self-identified as having CI using the pAD8, in comparison to 110 (20.4%) informants using the iAD8. The ICC of 0.519 (95% CI 0.454-0.578) indicated moderate agreement between pAD8 and iAD8 scores. CONCLUSIONS: When seeking emergency care, one in four older adults without a diagnosis of dementia and approximately 20% of informants indicated that patients were cognitively impaired. Our finding of moderate agreement between the pAD8 and iAD8 has important implications for clinicians providing care to patients with suspected CI and investigators conducting dementia-related studies.