Can average daily Fitbit sleep and activity data predict depression in undergraduate students?
LE3 .A278 2022
Bachelor of Science
This study attempted to use daily aggregate sleep and physical activity data measured by a Fitbit Charge 4 device to detect depression in university students. Undergraduate students have high levels of depression, and this study explores the use of wearables to monitor and predict student mental health. Participants were 10 students from Acadia University and their participation involved wearing a Fitbit device for seven days and seven nights and completing a standardized measure of depression (CESD-R). The first hypothesis of the study was that later waketimes and higher variability in sleep length would be correlated with higher depression scores. The second hypothesis was that lower activity metrics such as steps taken, and distance travelled will be predictive of higher depression scores. Both hypotheses were not supported, and no significant correlations were detected, most likely due to the small sample size. Note that 8 participants agreed to share their Intraday data, and that data was accessed using the Fitbit Web API. It is recommended that more data be collected.
The author retains copyright in this thesis. Any substantial copying or any other actions that exceed fair dealing or other exceptions in the Copyright Act require the permission of the author.