Using Passive Smartphone Sensing for Improved Risk Stratification of Patients With Depression and Diabetes:...

Area of Science:

• Digital Health
• Mental Well-being
• Diabetes Comorbidity

Background:

• Depression is underdiagnosed in diabetes patients, worsening outcomes.
• Smartphone sensing captures behavioral patterns linked to well-being.
• Digital tools can aid in managing comorbid depression in diabetes.

Purpose of the Study:

• Analyze smartphone sensing data for depression symptoms in diabetes.
• Explore risk stratification of diabetes patients with depression using digital data.

Main Methods:

• Cross-sectional observational study with 47 diabetes participants.
• Passive smartphone sensing of activity, mobility, sleep, and communication.
• Self-reported depression assessed via Patient Health Questionnaire-9 (PHQ-9) and machine learning classification.

Main Results:

• High depression prevalence (63%) observed in participants.
• Significant differences in daily activity and social contact between depression states.
• Extreme gradient boosting classifier achieved 81.05% accuracy in detecting depression symptoms.

Conclusions:

• Diabetes patients with depression show reduced activity and social contact.
• Smartphone sensing and predictive modeling show potential for early depression detection.
• Further randomized controlled studies are needed to validate findings.