Exploring Message Receptivity and Protocol Adherence in a Clinical Study: A Micro-Randomized Trial Protocol

Background Digital health studies increasingly combine passive sensing with active self-reports like ecological momentary assessments (EMAs). However, adherence to active data collection often declines over time, affecting data quality. Reminder messages using personalization or loss-aversion principles may improve adherence, but their relative effects and optimal timing remain unclear. This micro-randomized trial (MRT), embedded in the Glow Up prediabetes detection study, examines message receptivity and adherence to EMA protocols. Objective To examine how notification content (state-adapted, loss-aversion, or neutral) and physiological factors influence adherence to nutritional tracking in an observational study, aiming to inform future notification strategies. Methods This MRT within the Glow Up study evaluates adherence to an EMA protocol administered to 200 participants over a 4-week period via a smartphone app. Participants will be randomized three times daily to receive one of three reminder conditions: state-adapted, loss-aversion, or neutral. The measured proximal outcome is survey completion, indicating protocol adherence. Additionally, data captured by wearable devices is analyzed to explore associations between physiological factors and momentary states of receptivity. Results A 14-day pilot study with 20 participants in June 2025 will refine procedures. Main data collection begins in September 2025. Interim analyses will be conducted after 20% completion.