Effectiveness of a Just-In-Time Adaptive App to Increase Daily Steps: An RCT.

Abstract

Introduction: Addressing the public health problem of physical inactivity, this study evaluates SNapp, a just-in-time adaptive app intervention to promote walking through dynamically tailored coaching content. It assesses SNapp's impact on daily steps and how users' perceptions regarding ease of use and usefulness moderated its effectiveness.

Methods: SNapp was evaluated in an RCT from February 2021 to May 2022.This trial was preregistered in the Dutch Trial Register (NL7064). Analyses were conducted in November 2022. A total of 176 adults (76% female, mean age of 56 years) were randomized to a control group receiving a step counter app (n=89) or an intervention group receiving the app plus coaching content (n=87). SNapp's coaching content encompasses individually tailored feedback on step counts and advice to engage in more walking, taking preferences regarding behavior change techniques into account. Additionally, SNapp provides contextualized content calling attention to suitable walking locations in the user's environment. The primary outcome was daily step count as recorded by the step counter app. User perceptions regarding ease of use and usefulness were assessed via survey at 3-month follow-up.

Results: Mixed models indicated that the intervention did not significantly impact step counts on average over time (B= -202.30, 95% CI= -889.7, 485.1), with the coefficient indicating that the intervention group walked fewer steps per day on average, though this difference was not statistically significant. Perceived ease of use did not moderate the intervention effect (B_{group × perceived ease of use}=38.60, 90% CI= -276.5, 353.7). Perceived usefulness significantly moderated the intervention effect (B_{group × perceived usefulness}=344.38, 90% CI=40.4, 648.3).

Conclusions: SNapp increased steps only in users who deemed the app useful, underscoring the importance of user perceptions in app-based interventions.