Smartphone sensor-based mobility and social behavioral correlates of anxiety severity in adults

Background Smartphone-based sensing enables objective assessment of daily mobility and social behavior; however, evidence linking these behaviors to anxiety symptoms, particularly across weekdays and weekends, remains limited. Objective To assess associations between anxiety symptoms and smartphone-derived mobility and social behavioral markers, and to assess whether these associations differed between weekdays and weekends. Methods This cross-sectional study monitored 170 adults in the Netherlands for seven days using a smartphone application that captured global positioning system-based mobility and phone usage data (i.e., calls, messages, and social media). Sensed behavioral markers included time at home, distance traveled, number of locations visited, communication frequency, and social media use. Anxiety symptoms were assessed using the Generalized Anxiety Disorder-7 (GAD-7) questionnaire. Generalized additive regression models were fitted for the overall sample and separately for weekdays and weekends, adjusting for sociodemographic factors. Results The mean GAD-7 score was 3.9 (standard deviation = 4.5). Participants spent an average of 14.3 hours/day at home and visited 2.15 locations/day, with lower mobility levels on weekends. Visiting fewer locations on weekends was associated with higher GAD-7 scores, whereas no association was observed on weekdays or in the overall models. Null associations were found for time spent at home, distance traveled, communication frequency, and social media use. Conclusions Visiting fewer locations on weekends was associated with more anxiety symptoms, suggesting that the psychological meaning of mobility behaviors may depend on temporal context. Other smartphone-derived indicators showed no associations with anxiety. These findings highlight the importance of incorporating contextual and temporal dimensions into smartphone sensing and digital phenotyping research.