Validating remote cognitive tests in the Wisconsin Registry for Alzheimer's Prevention.

Abstract

Introduction: Remote cognitive assessment addresses barriers to research participation for older participants, but continuity in longitudinal studies is a challenge. We examined whether scores from telephone-based assessments (T-COG) were valid and reliable estimates of in-person traditional neuropsychological findings.

Methods: Participants in the Wisconsin Registry for Alzheimer's Prevention (WRAP) who had completed in-person testing within the prior 12 months were invited to complete a follow-up T-COG visit. Bland-Altman plots and intraclass correlations were used to assess repeatability and bias.

Results: Correlations between in-person and T-COG scores were of moderate to large magnitude (Pearson r: 0.57 to 0.80, intraclass correlation coefficients: 0.52 to 0.79). Bland-Altman plots revealed moderately wide limits of agreement, but no clear bias.

Discussion: The WRAP-abbreviated T-COG battery can be completed in less than an hour and shows good concordance with in-person test administration. Availability of the T-COG option may reduce missing data and/or enhance retention when obstacles to in-person attendance arise.

Highlights: Remote testing had adequate validity compared to in-person assessment.A telephone preclinical Alzheimer's composite correlated well with the original.Remote options may reduce missing data when barriers exist to in-person testing.Teleassessment may facilitate participant recruitment and retention.