Longitudinal Relationships Between Depressive Symptom Severity and Phone-Measured Mobility: Dynamic Structural...

Area of Science:

• Digital psychiatry
• Computational social science
• Behavioral data science

Background:

• Individual mobility, tracked via phone location data, is associated with depression.
• Longitudinal relationships between depressive symptom severity and phone-measured mobility require further exploration.

Purpose of the Study:

• To investigate the temporal direction of relationships between depressive symptom severity and phone-measured mobility over time.
• To identify specific mobility features that may predict changes in depression.

Main Methods:

• Utilized data from the Remote Assessment of Disease and Relapse-Major Depressive Disorder study across 3 European countries.
• Measured depressive symptom severity (PHQ-8) and collected mobile phone GPS/sensor location data bi-weekly.
• Employed dynamic structural equation modeling to analyze longitudinal associations between 11 mobility features and depression severity.

Main Results:

• Included 2341 records from 290 participants; significant negative correlations between depressive symptom severity and mobility were observed, particularly within individuals.
• Homestay duration, Location Entropy, and Residential Location Count predicted subsequent PHQ-8 score changes.
• Changes in PHQ-8 scores significantly influenced the periodicity of mobility.

Conclusions:

• Phone-derived mobility features show potential for predicting future depression.
• These findings support clinical applications in relapse prevention and remote mental health monitoring.
• Mobility patterns can serve as objective indicators in real-world mental health assessments.