JMIR Publications

ABSTRACT

Background:

Due to increasing rates of Alzheimer’s disease and related dementias (ADRD), there is a great need for easily deployable tools for quantifying risk. Digital technology such as smartphones and smartwatches have ushered in a new era of potential monitoring for ADRD due to their unobtrusiveness and allowance for continuous data collection, but there is limited information regarding feasibility of digital data collection among older adults.

Objective:

This paper examines the feasibility of a four-week combined smartwatch and (ecological momentary assessment) EMA study in a racially diverse sample of older adults.

Methods:

Forty-six older adults (55+) with either mild cognitive impairment (MCI) or healthy cognition completed an informed consent comprehension quiz, baseline cognitive testing, brief training regarding digital data collection, and questionnaires during an in-person session. Then, participants were instructed to wear a Garmin Vivosmart4 smartwatch 23 hours/day for four-weeks, sync two smartphone apps (Garmin, Labfront), and complete a daily EMA survey. Training time, daily smartwatch wear time, adherence to the daily EMA survey, and performance on the consent quiz were quantified. Feasibility metrics were analyzed in relation to cognition, self-reported functional decline, and demographic factors. Self-reported usability of the apps and smartwatch was collected at the end of the study.

Results:

: During the four-week study, participants wore the smartwatch during waking and sleep hours for an average of 20 hours per day with an average 93% response rate to daily EMA surveys. Participants also scored well on the consent comprehension quiz (M=97.33% correct, SD=6.86)). Older participants required longer study training times (r=.405, P=.006) and participants who reported greater cognitive/functional decline (ECog) had lower daily average wear time (r=-.387, P=.009). Those with MCI (Median=20.64, IQR=2.74) had significantly lower daily average wear time than those with healthy cognition (Median=21.45, IQR=1.29; U=79.00, z=-2.048, P=.041, r=-.31). Black participants (Median=20.61, IQR= 3.18) had significantly lower smartwatch wear time than White participants (Median=21.67, IQR=.97, U=310.00, z=3.357, p<.001).On the usability survey all participants (100%) indicated they would participate in a future study asking them to wear the study smartwatch and over 80% agreed they had a positive experience with the smartwatch. Over 90% of participants were satisfied or very satisfied with the syncing process of both smartphone apps.

Conclusions:

Smartwatch monitoring, including daily wear and smartphone syncing, is highly feasible with older adults, with strong adherence to daily wear and EMA surveys and high satisfaction reported on usability surveys. Older participants and participants with mild cognitive decline will require more support in learning study procedures and in increasing daily smartwatch wear time. Overall, longitudinal monitoring with the Garmin Vivosmart4 smartwatch and Labfront application is acceptable and feasible to collect nearly continuous data in Black and White older adults with and without cognitive impairment.