Electronic Thesis and Dissertation Repository

Thesis Format

Integrated Article

Degree

Doctor of Philosophy

Program

Health Information Science

Supervisor

Bauer, Michael A.

Abstract

This dissertation presents findings from three longitudinal studies examining indicators of positive undergraduate mental health primarily during COVID-19 pandemic conditions at a large, residential, urban university in Canada (Western University) using apps with personal sensing data collection capabilities (e.g. GPS, step counts) over March 2019-August 2021. The apps featured mental health-related questionnaires – responses were representative of psychosocial outcomes for participants. Personal sensing data were representative of participant behavior/lifestyle. Questionnaires and personal sensing data were collected at the same time; personal sensing data was also collected hourly in the background for the final two studies. Associations between psychosocial outcomes and behavior/lifestyle found via mixed linear modelling revealed indicators of positive mental health. The first study, Smart Healthy Campus 1.0, began March 2019, concluding with limited data when COVID-19 was declared a pandemic; possible associations were identified. Initial plans were to improve SHC 1.0 with an upgraded SHC 2.0 study, but the Student Pandemic Experience (SPE) study was launched first, with questionnaires more fixated on pandemic conditions. Following this, SHC 2.0 was also launched shortly after, still relevant, although using shorter generic questionnaires. SPE completed with 315 participants who were primarily female (76%) iOS users (85.3%). Data collected for 40 weeks (11/2020 – 09/2021) included 4851 questionnaire responses and 25985 sensor samples with up to 68 individual values per sample. Mixed linear models were fit relating 15 mobile device (phone or tablet) sensors (e.g. step count) to 12 mental health-related questionnaire scores. SHC 2.0 ran with 94 participants were who primarily male (76.6%) iOS users (86%). Data collected for 30 weeks (01/2021 – 08/2021) included 1722 questionnaire responses and 6518 sensor samples with up to 68 individual values per sample. Mixed linear models were fit relating 12 mobile device (phone or tablet) sensors (e.g. step count) to the SHC 2.0 questionnaires. From these studies, it was found that device sensors had statistically significant associations with the selected mental health-related questionnaires for undergraduates during a major pandemic. These findings suggest directions for mental health-related programs (e.g. apps or physical activity) and interventions during a pandemic for a comparable group and setting.

Summary for Lay Audience

This research consisted of three studies run over several months which tried to find lifestyle patterns in university students who felt better, coped better, and had better moods during the COVID-19 pandemic. This research occurred at a large university in Canada situated in part of a city mostly with housing (Western University). Students allowed themselves to be monitored with special Apple or Android apps over March 2019-August 2021. The apps featured questionnaires asking about how they coped and how their mood was. The apps also collected sensor data automatically using student phone or tablet hardware, like GPS and steps. Using statistical analysis, relationships were found between responses to the mood/coping questionnaires and sensor data. The first study, Smart Healthy Campus 1.0, began March 2019 and ended with a small amount of data when COVID-19 became a pandemic, but some trends in student mood and sensor data were found. Plans were to improve SHC 1.0 with a better SHC Version 2.0 study, but we launched the Student Pandemic Experience (SPE) study first, because SPE asked about students coping specifically during the pandemic. Eventually, SHC 2.0 was launched because it still had useful questionnaires for the pandemic. SPE completed with 315 students who were primarily female (76%) Apple device users (85.3%). Data collected for 40 weeks (11/2020 – 09/2021) included 4851 questionnaire responses and 25985 sensor readings, which had up to 68 individual items each. Statistical models were made which related 15 phone/tablet sensors (e.g. step count) to 12 mental health-related questionnaire scores. SHC 2.0 ran with 94 students were who primarily male (76.6%) Apple device users (86%). Data collected for 30 weeks (01/2021 – 08/2021) included 1722 questionnaire responses and 6518 sensor readings, which had up to 68 individual items each. Statistical models were made which related 12 phone or tablet sensors (e.g. step count) to the special SHC 2.0 questionnaires. We found significant relationships between mood and coping-related questionnaires and phone/tablet sensor data from university students during a major pandemic. These findings suggest coping strategies for students to help them be healthier and happier, which helps with pandemic preparedness.

shcmodels_new.txt (61 kB)
R output of SHC 2.0 Initial mixed models

shcmodelsround2_new.txt (40 kB)
R output of SHC 2.0 Revised mixed models

models_new.txt (43 kB)
R output of SPE Initial mixed models

Available for download on Tuesday, May 31, 2022

Share

COinS