Differential responses to interventions targeting physical activity compared to sedentary time: a randomized n-of-1 design

Abstract

Background: Physical activity and time spent being sedentary predict glucose control in type 2 diabetes. This study aimed to investigate how individuals respond to interventions targeting reduced sedentary time (ST) or increased bouts of physical activity (PA). Methods: Using an n-of-1 RCT design, seven participants wore an accelerometer measuring PA for 6 months that delivered randomly-allocated prompts on alternate days; either to improve PA, reduce ST or no prompt. The design facilitated testing differential intervention effects within each individual. No prompts were delivered on days subsequent to receiving either PA or ST prompts, to examine carry-over effects. Bootstrapped time series analyses were used to assess the within-participant effect of each type of prompt over time, and to explore carry-over effects. Findings: Each participants varied in response to the PA and ST interventions. Carry-over effects were observed across all participants after both types of interventions. Discussion: Distinct within-participant responses to PA and ST interventions further highlights the need to understand how best to tailor behaviour change interventions. N-of-1 RCTs provide a robust means to explore tailoring, though carry-over effects pose challenges.