User experience and biomarker associations of remotely administered and unsupervised smartphone‐based cognitive testing in a community sample of older adults

Abstract

AbstractBackgroundRemotely administered, unsupervised smartphone‐based cognitive tests have the potential for scalable and cost‐effective screening of Alzheimer’s disease (AD) in clinical trials and primary care. Previous work has provided initial evidence for the feasibility and validity of these tools in clinical and community settings. However, more studies investigating the utility of these cognitive tests are warranted, particularly regarding user experience and the association with AD biomarkers.MethodTo date, 198 participants (CDR 0 = 85%, CDR 0.5 = 15%, Mini‐Mental State Examination = 29, average age = 77) have enrolled in this smartphone sub‐study, a part of the Swedish population‐based Gothenburg H70 Birth Cohort Studies. After participating for at least 12 months, the user experience of smartphone‐based cognitive tests (two memory tests, MDT‐OS and ORR‐LDR) was assessed. Furthermore, we investigated their associations with volumetric brain changes based on magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) biomarkers (Aβ42/Aβ40 ratio, T‐tau, P‐tau, NFL, and neurogranin). The latter will be analyzed in a subsample (n = 45 for CSF and n = 110 for MRI) of all participants. Voxel‐based analyses are being performed in SPM12.ResultMost study participants found the app easy and enjoyable to use, the instructions easy to understand, and they generally favored performing tests in their home environment. Low CSF Aβ42/Aβ40 ratio at age 70 was associated with worse performance on the ORR‐LDR at age 77 (R = .43, p<.002), while measures of t‐tau, p‐tau, NFL, and neurogranin at age 70 were not. Voxel‐wise associations between smartphone‐based cognitive tests and MRI measures will be reported at the conference.ConclusionOur results suggest that using smartphone‐based cognitive tests in a remote unsupervised setting has the potential to identify aspects of early AD in a community‐based setting. The positive user experience is promising for acceptance of these novel tools for cognition assessment. Next steps include analyses of the relation with neuroimaging‐based biomarkers.