Premium
Validation of a Brief Multi‐Dimensional Assessment of Dementia Severity
Author(s) -
Peipert John D.,
Jennings Lee A.,
Romero Tahmineh,
Hays Ron D.,
Wenger Neil S.,
Keeler Emmett,
Reuben David B.
Publication year - 2021
Publication title -
journal of the american geriatrics society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.992
H-Index - 232
eISSN - 1532-5415
pISSN - 0002-8614
DOI - 10.1111/jgs.16930
Subject(s) - dementia , medicine , activities of daily living , hazard ratio , gerontology , cognition , scale (ratio) , psychiatry , confidence interval , disease , physics , quantum mechanics
BACKGROUND/OBJECTIVES Briefer measures of symptoms and functional limitations may reduce assessment burden and facilitate monitoring populations of persons with dementia (PWD). DESIGN Prospective follow‐up study. SETTING University‐based dementia care management program. PARTICIPANTS 1,091 PWD. MEASUREMENTS We assessed cognition (Mini Mental State Examination (MMSE)—11 tasks), neuropsychiatric symptom severity (Neuropsychiatric Inventory Questionnaire Severity Scale (NPIQ‐S)—12 items), and functional ability (Activities of Daily Living (ADL)—6 items; Functional Activities Questionnaire (FAQ)—10 items). Item response theory was used to select subsets of items by identifying low item discrimination (<1.50), poor item fit ( χ 2 ), local dependence (LD), and with difficulty similar to other items. We estimated correlations between original and shorter scales and compared their associations with mortality. We added two symptoms (trouble swallowing, coughing when eating) reflecting late‐stage dementia complications, created a multi‐dimensional dementia assessment composite, and examined its association with mortality. RESULTS Five MMSE tasks were eliminated: two with low discrimination, two with difficulty similar to other items, and one with poor fit. The remaining tasks were correlated with the full MMSE at r = 0.82. We retained three ADLs that were correlated with the total ADL set at r = 0.95 and kept five FAQ items that were not LD (correlation with full FAQ, r = 0.97). Associations with mortality were similar between the longer and shorter scales. A higher score on the composite (range 0–100) indicates worse dementia impact and was associated with mortality (hazard ratio (HR) per scale point: 1.03 (1.02–1.04)). CONCLUSION These brief assessments and dementia composite may reduce administration time while preserving validity.