Digital Biomarkers for Passive Remote Monitoring of Bipolar Disorder: Systematic Review
Bibliographic record
Abstract
Background: Bipolar disorder (BD) features episodic shifts among (hypo)mania, depression, mixed states, and euthymia. Timely detection of mood transitions is difficult due to infrequent clinical touchpoints. Digital health technologies, including wearables and smartphones, offer a unique opportunity to passively and continuously monitor behavior and physiology that could reflect underlying mood dynamics in real-world settings. Objective: We aim to systematically review passively collected digital biomarkers for BD mood states, characterize devices/modalities and analytic approaches, appraise risk of bias, and identify design gaps and priorities for clinical translation. Methods: Following PRISMA guidelines (PROSPERO CRD42024607765), we searched MEDLINE, PsycINFO, Scopus, IEEE Xplore, and ACM Digital Library (February 7, 2025). We included peer-reviewed studies of adults with BD I/II that measured passively collected digital biomarkers and related them to depressive, (hypo)manic, mixed, or euthymic states. Active-only measures (e.g. lab tests, ecological-momentary assessment) and studies entangling BD with other diagnoses were excluded. Two independent reviewers screened studies and extracted study characteristics and results. We grouped digital biomarkers into categories and conducted narrative synthesis. Risk of bias was assessed with PROBAST (predictive models) and the Newcastle-Ottawa Scale (observational studies). Results: Of 8,355 records, 45 studies met criteria. Most enrolled ≤50 participants (64%) and monitored ≤100 days (49%); 29% collected data only in-clinic. Nine biomarker domains emerged: physical activity, heart rate (HR), electrodermal activity (EDA), geolocation, keyboard use, light exposure, sleep, socialization, and speech. Consistent patterns linked depression to reduced mobility and social interaction, later/variable sleep, and lower daytime light; (hypo)mania was associated with higher and more variable activity, shorter/advanced sleep, and increased communication. Circadian features derived from sleep/activity repeatedly aided prediction. EDA tended to be lower in depression; HRV findings were mixed across settings and methods. Keyboard and speech features (e.g., timing, prosody) showed associations and performed well in classifiers. Fifteen studies used ML; several reported strong performance for episode prediction/classification (AUROC ≈0.80-0.98 in larger cohorts), yet external validation was absent, samples were small, monitoring windows were often short relative to episode timescales, clinical labels were infrequent/misaligned, and missingness was rarely modeled despite likely informativeness. Conclusions: Passive digital biomarkers for BD show promise, with the most robust signals aligning with DSM-5 behavioral and circadian features (sleep-wake patterns, activity/mobility, socialization/geolocation, and speech). To move from promise to practice, future studies should adopt longer within-subject monitoring, align label cadence with sensing granularity, standardize features/reporting, pre-register analyses, externally validate models, minimize data to protect privacy, and expand physiological measurement beyond heart rate and electrodermal activity. These steps are essential to develop reliable, actionable tools for earlier detection and management of BD mood episodes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.013 | 0.063 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.008 | 0.006 |
| Bibliometrics | 0.012 | 0.011 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".