Digital Biomarkers for Passive Remote Monitoring of Bipolar Disorder: Systematic Review

Area of Science:

• Digital Health
• Psychiatry
• Biomarkers

Background:

• Bipolar disorder (BD) is characterized by mood episodes (mania, depression, euthymia).
• Detecting mood transitions is challenging due to infrequent clinical contact.
• Digital health technologies offer continuous, real-world monitoring of behavior and physiology.

Purpose of the Study:

• To systematically review passively collected digital biomarkers for BD mood states.
• To characterize devices, modalities, and analytical approaches used in BD research.
• To identify gaps and priorities for clinical translation of digital biomarkers.

Main Methods:

• Systematic review following PRISMA guidelines, searching major databases.
• Inclusion of studies on adults with BD I/II using passively collected digital biomarkers.
• Exclusion of active measures and studies with mixed diagnoses; risk of bias assessment.

Main Results:

• 45 studies identified, with most having small cohorts and short monitoring periods.
• Key biomarker domains include physical activity, sleep, socialization, geolocation, and speech.
• Consistent patterns link depression to reduced activity/socialization and mania to increased activity/communication; circadian features showed predictive value.

Conclusions:

• Passive digital biomarkers show promise, particularly those reflecting behavioral and circadian patterns.
• Future research requires longer monitoring, standardized reporting, external validation, and privacy-preserving data minimization.
• Development of reliable tools for earlier BD detection and management necessitates these advancements.