Clusters of Physical Activity, Sedentary Behavior and Alcohol Use Across Three Timescales
Restricted (Penn State Only)
- Author:
- West, Ashley Brooke
- Graduate Program:
- Kinesiology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- February 24, 2022
- Committee Members:
- Michael Russell, Outside Field Member
Diana Fishbein, Outside Unit & Minor Member
David Conroy, Chair & Dissertation Advisor
Lacy Alexander, Major Field Member
Jonathan Bates Dingwell, Program Head/Chair - Keywords:
- physical activity
sedentary behavior
alcohol use
adolescents
college students
polysubstance use - Abstract:
- Adolescence and emerging adulthood are critical life stages, marked by evolving health behavior patterns. Habits established during these transitional periods can track into adulthood, making the successful management of multiple health behaviors a marker of developmental competence. Increasing levels of physical inactivity, sedentary behavior and risky substance use present a challenge for these populations as they move into adulthood. Interventions that support multiple behavior change can be efficient and effective yet are sometimes complicated when behaviors cluster. Prior work has identified an antagonistic coupling between physical activity and alcohol use largely through cross-sectional study designs and self-reports of behaviors. Discovery work is still needed to understand if this coupling exists when examined longitudinally on varying time scales and when the behaviors are measured with self-reports and wearable sensors. Prior research is also limited with respect to how alcohol use is associated with sedentary behavior. Further, it is unknown how these behavioral clusters present within high-risk populations. This dissertation involved three studies. The first study was a systematic review of 97 studies that either reported an association between physical activity or sedentary behavior and alcohol or cannabis use within adolescents or emerging adults. This review found that physical activity and alcohol use were positively associated among emerging adults. It also revealed that sedentary behavior could be a risk factor for substance use in adolescents. The majority of the studies in this review measured these behaviors via self-reports. It was unclear if associations between these behaviors would differ if physical activity, sedentary behavior and alcohol use were captured in real-time with wearable sensors. The second and third studies were designed to address limitations found in the review, namely a need for assessment of these associations via wearable sensors and the need to examine how sedentary behavior is associated with alcohol use among emerging adults. Both studies used data from a new, 11-day ecological momentary assessment study. The second study evaluated day-level associations between physical activity, sedentary behavior and alcohol use with self-reports, an activity monitor and a transdermal alcohol sensor. Results indicated that movement behaviors were associated with time-based measures of alcohol use, but the antagonistic clustering of physical activity and alcohol use may not exist under all conditions. The third study evaluated these associations at the momentary level and indicated that, in the late evening hours, physical activity was positively associated with alcohol use, whereas sedentary behavior was negatively associated with alcohol use during traditional working hours. Collectively, these studies mapped associations between movement behaviors and substance use across adolescence into emerging adulthood and identified times of vulnerability that could inform timing of interventions for a sub-population of high-risk emerging adults. This dissertation highlighted that the antagonistic clustering between physical activity and alcohol use may not exist in at-risk populations (e.g., polysubstance-using college students) or when measuring these associations with sensors. Finally, findings from this dissertation reveal that not all parts of the day are equally implicated, and associations seen at the daily timescale could be attributed to behavioral clustering at specific hours in the day.