Mobile Toolbox: Remote, unsupervised longitudinal assessment of a diverse cohort in the UCSF Brain Health Registry

Abstract

Listen

Translate article

AbstractBackgroundA critical need in the AD field is development and validation of efficient, scalable cognitive assessments to identify older adults with cognitive impairment. The Mobile Toolbox (MTB), a novel suite of app‐based cognitive assessments, was developed to address this need. The goal of this study was to assess the feasibility, usability, compliance, longitudinal retention, and validity of MTB assessment in a large, ethnoculturally and educationally diverse cohort in a remote, unsupervised setting.MethodParticipants were recruited from the UCSF Brain Health Registry (BHR), an online study. BHR participants were invited by email, screened for eligibility (English fluency and access to a compatible smartphone), and provided online instructions for MTB app download. All participants could complete a baseline administration. A subset was invited to a longitudinal study, in which they complete MTB assessments at a short‐term timepoint (day 7, 14, or 21; study arms sequentially assigned), and then at 6‐month intervals. Enrollment goals were N = 800 with equal distribution across 10‐year age bands (ages 18‐80+); 60% with <16 years of education; 10% non‐Latinx Black, 15% Latinx, and 5% non‐White other ethnocultural identity.ResultOver a 5‐month period, 48,110 BHR participants were invited to the study. Of those, 8287 (17%) expressed interest, 3554 (7%) completed baseline assessment, and 1327 (3%) were assigned to the longitudinal study. Participants submitted 1205 help tickets asking for email support to complete MTB. The baseline and longitudinal cohorts underrepresented Black individuals and younger age bands, despite targeted recruitment efforts (Table 1). Longitudinal completion rates were 59.0% for the short‐term timeline and 55.2% for the 6‐month timepoint, and varied by ethnocultural group (Figure 1). Compared to those invited, those who enrolled in the longitudinal study were older, had higher education, and were more likely to identify as female non‐Latinx White (p<0.05).ConclusionResults demonstrate feasibility and usability of remote, unsupervised MTB assessment. Underrepresented populations were less likely to enrolled and to complete assessments at baseline and longitudinally, emphasizing the need to develop more inclusive recruitment and engagement efforts. This approach can be scaled up to efficiently assess cognition in many research, clinical trial, and healthcare settings.