Well-Being Tracking via Smartphone-Measured Activity and Sleep: Cohort Study

Area of Science:

• Digital health
• Mental health technology
• Computational psychiatry

Background:

• Personalized treatment for mood disorders like depression and bipolar disorder can be improved by automatically tracking mental well-being.
• Smartphones offer a unique opportunity to monitor behavior and infer mental well-being due to their ubiquity.

Purpose of the Study:

• To evaluate the effectiveness of smartphone-based activity and sleep tracking for monitoring individuals' mental well-being.
• To determine the correlation between smartphone-derived activity and sleep patterns and self-reported mental well-being.

Main Methods:

• Recruited 106 individuals to install a smartphone app for daily well-being surveys and behavioral tracking.
• Utilized smartphone accelerometer data to infer activity and sleep measures in a subset of 53 participants.
• Related inferred activity and sleep measures to participants' self-reported well-being and used them for prediction.

Main Results:

• Daily physical activity approximations from smartphones positively correlated with mood (P=.004) and energy levels (P<.001).
• Sleep duration positively correlated with mood (P=.02), but sleep disturbance measures did not show significant relationships.
• Predictive models using activity and sleep measures showed significant improvement over baseline models for well-being prediction (P<.01).

Conclusions:

• Inferred smartphone activity and sleep measures are strongly related to and predictive of mental well-being.
• Physical activity and sleep are crucial factors for predicting mood and enabling automatic mood monitoring.
• While model improvements were modest, the findings support the utility of smartphone data for mental well-being assessment.