Person-specific dose-finding for a digital messaging intervention to promote physical activity

Area of Science:

• Behavioral Science
• Control Systems Engineering
• Digital Health

Background:

• Digital messaging is a known method for promoting physical activity.
• Systematic dose-finding approaches are underutilized in behavioral intervention development.
• Control systems engineering tools can model dynamical systems for behavioral interventions.

Purpose of the Study:

• To apply system identification from control systems engineering to estimate dynamical models.
• To inform decision rules for digital messaging interventions aimed at increasing physical activity.
• To explore personalized and context-sensitive approaches to digital health interventions.

Main Methods:

• 45 insufficiently active emerging and young adults participated over 6 months.
• Participants wore activity monitors and received 0-6 daily digital messages (move more, sit less, inspirational).
• Message context (weather) and individual responses were analyzed using system identification.

Main Results:

• A generic model showed limitations in capturing heterogeneous individual responses.
• No message type was universally ineffective; responses varied by person and context.
• Exterior temperature moderated the effects of certain messages on physical activity.

Conclusions:

• Generic models offer initial insights for adaptive interventions but can be inefficient.
• Person-specific models, updated with accumulating data, are crucial for optimizing interventions.
• Interventions should evolve with individual experiences and environmental exposures for sustained physical activity promotion.