Development of a smartphone application to measure physical activity using sensor-assisted self-report

Area of Science:

• Digital Health
• Behavioral Science
• Epidemiology

Background:

• Objective physical activity monitors have limitations, including high non-wear rates and inability to capture behavioral context.
• Adolescents increasingly use smartphones, which possess built-in motion sensors suitable for activity monitoring.

Purpose of the Study:

• To develop and describe a smartphone application, Mobile Teen, for enhanced physical activity assessment.
• To integrate objective and self-report strategies using sensor-informed context-sensitive ecological momentary assessment (CS-EMA) and end-of-day recall.

Main Methods:

• The Mobile Teen app utilizes the phone's motion sensor to detect non-wear, sedentary behavior, and physical activity.
• Sensor data transitions trigger CS-EMA surveys for real-time activity context (type, purpose).
• An end-of-day recall feature allows users to label activity periods using visual cues from sensor-detected transitions.

Main Results:

• The app automatically identifies activity states and transitions, prompting user input for context.
• It enables users to review and label their own physical activity data interactively.
• This approach augments objective monitor data by addressing non-wear gaps and adding real-time behavioral correlates.

Conclusions:

• Smartphone applications can effectively supplement objective physical activity monitors.
• Sensor-driven CS-EMA and recall features provide valuable real-time contextual data.
• These apps offer a scalable and affordable solution for large-scale physical activity research.