Open Collections

Abstract

Introduction: Physical inactivity contributes significantly to the development of non-communicable diseases such as type 2 diabetes (T2D). Despite updates to the WHO guidelines recognizing physical activity of any duration, many individuals with T2D remain insufficiently active. "Exercise snacks" - brief, vigorous activity bouts embedded into daily routines - may be a feasible modality. Supported by epidemiological evidence linking short vigorous activity measured by accelerometers to lower mortality risk, exercise snacks could meaningfully improve health. Purpose: To determine if accelerometers could be used to detect exercise snacks in people living with T2D and to explore whether an exercise snacks intervention impacted accelerometer-measured physical behaviours. Methods: This thesis was a secondary analysis of a randomized controlled trial (RCT) in which participants with T2D were assigned to either an exercise snacks group (n=32) or a light-intensity placebo group (n=38). Prior to the intervention, participants completed an acute randomized crossover trial under diet-controlled conditions involving exercise (snack or light intensity placebo) and non-exercise days. The 12-week intervention was supported via a mobile app that provides reminders, exercise options, and tracked adherence. Physical activity behaviour was assessed using activPAL4 accelerometers at baseline (Week 0) and Week 11 of the intervention. Detection metrics for identifying exercise snacks were developed from two designated exercise days and cross-referenced with participant logbooks. Results: Detection using raw acceleration alone identified exercise snacks with moderate sensitivity (61.5%) but low accuracy (32.7%), often misclassifying unrelated vigorous movements. Acutely under controlled conditions, exercise snacks significantly increased physical activity intensity compared to non-exercise days, as measured by MX Metrics. At Week 11, participants in the exercise snacks group spent more time in high-intensity acceleration bins when compared to the light-intensity placebo group (350–400 mg; 0.427 [95% CI: 0.209, 0.647], p < 0.001 and >400 mg; 0.719 [0.482, 0.956], p < 0.001). Conclusion: ActivPAL4 accelerometers can detect exercise snacks, but improved detection algorithms are warranted. Exercise snacks acutely increase the distribution of physical activity intensity and, over time, significantly increase time spent in vigorous physical activity, demonstrating their potential to encourage vigorous activity for people living with T2D.