Macrotemporal Variables Influence the Effect of a mHealth Intervention for Adolescent Physical Activity (Preprint)

Abstract

Background: An adaptive text messaging intervention to promote adolescent physical activity has demonstrated encouraging feasibility, acceptability, and preliminary efficacy in a recent proof-of-concept study. To inform future intervention development, a secondary analysis of the data examined how physical activity is influenced by mood, environment, and physical feelings of energy and fatigue.

Objective: To understand how both macro- and microtemporal variables (e.g.., psychological and environmental variables at both levels) influence the efficacy of a brief mHealth intervention (i.e., NUDGE) for physical activity.

Methods: Using a matched control design, we evaluated the effect of daily positive and negative affect, perceptions of the weather, energy, and fatigue as moderators of the effect of the intervention on 21 intervention participants and 21 matched controls.

Results: Consistent with study hypotheses, macrotemporal (levels of the variable on a three-week timescale) moderators of intervention effectiveness were observed for positive affect (p < .001), negative affect (p = 0.031), energy (p < .001), fatigue (p < .001), and perceived weather barriers (p < .001) for moderate-to-vigorous physical activity. These effects were more consistently observed for moderate-to-vigorous physical activity than for sedentary behavior, which was only significant for energy (p < .001). No effects for microtemporal (variables at the day level) were observed.

Conclusions: There appears to be an optimization opportunity for mHealth physical activity interventions that can be achieved by personalizing intervention features and content based on approximately monthly assessments of affect, physical feeling states, and perceived weather barriers.