Remote, self-administered, smartphone cognitive testing in a registry-based cohort: Feasibility, reliability, and validity findings.

BACKGROUND

Remote, smartphone-based cognitive testing may improve access to cognitive assessments for Alzheimer's disease and related dementias. We evaluated the feasibility, reliability, and validity of unsupervised smartphone-based cognitive tests in a registry-based cohort.

METHODS

Adults without a record of cognitive impairment (N = 1815; ages 18-92) were recruited from the University of California, San Francisco (UCSF) Brain Health Registry to complete three unsupervised smartphone cognitive testing sessions within 2 weeks. Reliability was assessed with correlations between sessions. Linear regression models tested associations of smartphone tasks with demographics, self- and informant-rated cognitive concerns, and web-based cognitive testing (Cogstate Brief Battery).

RESULTS

Adherence was high (82.2%) and usability favorable. Test-retest reliability was moderate to strong (ρ's = 0.61-0.85, all p's < 0.001). Lower smartphone scores were associated with older age, lower education, cognitive concerns, and worse Cogstate performance.

DISCUSSION: Findings support the feasibility, reliability, and validity of remote digital assessments in adults without a record of cognitive impairment.