O-089 an Mhealth Intervention to Promote Physical Activity Among Workers Using a Deep Learning Model for Passive Monitoring of Depression and Anxiety: A Single-Arm Feasibility Trial

Introduction Promoting physical activity effectively prevents depression and anxiety. Although mobile health (mHealth) technologies offer promising results in promoting physical activity, conflicting evidence exists on mHealth effectiveness, and workers face barriers to using mHealth services. The present study aimed to preliminarily investigate the effectiveness and implementation of a newly developed smartphone app, named ASHARE, employing a deep learning model to monitor depression and anxiety using physical activity. Methods A single-arm intervention study was conducted. From March to April 2023, workers aged 18 and older who were not absent were recruited. The participants were asked to install and use the app for one month. Before and after using the app, self-reported physical activity and psychological distress were measured. Physical activity duration from the app was also recorded digitally. Implementation Outcome Scales for Digital Mental Health were used to measure acceptability, appropriateness, feasibility, satisfaction, and harm. Results The study included 24 workers. On average, the app was used for 12.54 days. After using the app, no significant change was observed in physical activity (p=0.309) or psychological distress (p=0.926) either. The participants who had a severe level of psychological distress decreased significantly (p=0.011). The digitally recorded duration of physical activity was increased during the intervention period (p=0.078). Discussion Scores in acceptability, appropriateness, and satisfaction were lower compared to previous mHealth studies while those in feasibility and harm were better. Conclusion The app ASHARE was not enough to promote physical activity or improve psychological distress. Improving acceptability, appropriateness, and satisfaction is considered a key issue to implement this app.