Premium
Intensive measurement of cognition to support early detection of cognitive change in individuals at risk of dementia
Author(s) -
Brewster Paul,
Rush Jonathan,
Ozen Lana,
Jacobs Diane M,
Kaye Jeffrey,
Nygaard Haakon B,
Feldman Howard H,
Hofer Scott M
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.042599
Subject(s) - dementia , cognition , cohort , cognitive test , montreal cognitive assessment , mood , cognitive decline , psychology , memory span , audiology , digit symbol substitution test , working memory , medicine , clinical psychology , cognitive impairment , psychiatry , alternative medicine , disease , pathology , placebo
Abstract Background Cognitive changes in nondemented older adults are best measured longitudinally and relative to individual baselines. Effect sizes of early changes in cognition are small with most traditional measures, and so there is a need for alternative measurement approaches that can improve statistical power to detect change. Remotely administered mobile assessments can address this need by permitting much more frequent repeated measurements, which then permits statistical delineation of early slopes of decline from performance variation associated with extraneous factors (e.g., mood, practice). MyCogHealth is an iOS/Android app for self‐administered cognitive assessments and customizable surveys. Inclusion of MyCogHealth in the Canadian Thumbs Up program (CTU) will permit evaluation of feasibility and utility of intensive measurement designs for capturing early cognitive changes in older adults identified to be at increased risk of dementia. Method The CTU Platform Trial Cohort will complete five intensive measurement “bursts” in three‐month intervals using MyCogHealth. Each burst involves brief self‐administered assessments completed twice daily across seven consecutive days. Each assessment includes the Symbol Match test (processing speed), Dot Memory (visual short‐term memory), a digital Trailmaking Test analogue, and surveys assessing state factors that can affect cognition. A pilot cohort (N=17; Mean age=74; Mean MoCA score=28) was followed for 12 months using this protocol to evaluate feasibility and psychometric integrity of the Symbol Match and Dot Memory tests in this design. Result In the pilot cohort Symbol Match correlated with Digit Span Forward and the Trailmaking Test (r=.51‐.66). Dot Memory correlated with Backward Digit Span (r=.49), Trailmaking Test (r=‐.59), object naming (r=.64), and visual episodic memory (r=.58). Retest reliability was .80 for Dot Memory and .87 for Symbol Search. Intraclass correlation was .77 for Symbol Search and .42 for Dot Memory. Participant retention was 100% and compliance with burst testing was 94%. Conclusion Preliminary results support the feasibility, validity, and reliability of self‐administered burst testing deployed longitudinally. Inclusion of MyCogHealth in CTU should be informative in capturing early slopes of decline in the study cohort. Such early detection of decline will inform more powerful, personalized trials where those at the highest predicted risk would be included and those at lower risk excluded.