Digital Phenotypes for Predicting Depression

• Conditions: Depression - Major Depressive Disorder; Depression Bipolar; Bipolar Disorder (BD); Mood Disorders

• Registration Number: NCT07151846
• Lead Sponsor: Korea University Anam Hospital

Brief Summary

This longitudinal study aims to identify and validate digital phenotypes that can predict recurrence of major depressive episodes using passively collected, real-time sensing data from smartphones and wearable devices. Over a 12-month period, 540 participants-including patients with mood disorders and healthy or high-risk controls-will complete five clinical assessments at 3-month intervals, wear a Fitbit device daily, and log daily mood ratings via a mobile app. The study includes the development of AI-based predictive models and the construction of an anonymized wearable big-data repository for mood disorders.

Detailed Description

This is a 12-month longitudinal follow-up study designed to identify digital phenotypes capable of predicting the recurrence of major depressive episodes using passively collected real-time data from smartphones and wearable devices. A total of 540 participants will be enrolled, including 200 patients diagnosed with mood disorders who have experienced at least one depressive episode in the past two years, and 340 healthy or high-risk control participants without a clinical diagnosis.

Clinical interviews and self-report measures will be conducted five times at 3-month intervals. Throughout the study period, participants will wear a Fitbit device daily and use the "My Mental Hero" mobile application to record their mood once a day at a self-selected time. Data collected will be used to develop machine learning models for predicting depressive recurrence and to build a large-scale, anonymized wearable big-data repository for mood disorders.

Participants will receive no interventions beyond data collection, and neither randomization nor blinding will be implemented. Trained psychiatrists, clinical psychologists, or research nurses will conduct the assessments. Initial visits will include informed consent, device setup, and app training, taking approximately 1-2 hours. Subsequent assessments will take about one hour, with additional brief interviews (≤30 minutes) conducted if a mood episode is suspected.

While follow-up assessments for patients and high-risk participants will be conducted in person, those for healthy controls may be conducted remotely (via Zoom or phone) under secure and private conditions without screen recording. Remote assessments may also be used for patient and high-risk groups if necessary.

Study Timeline

• Study Start: 2025-02-27
• Study First Submitted: 2025-07-23
• Status Verified: 2025-08
• Study First Submitted QC: 2025-08-25
• Last Update Submitted: 2025-08-25
• Study First Posted: 2025-09-03
• Last Update Posted: 2025-09-03
• Primary Completion: 2026-08-31
• Study Completion: 2026-08-31

Recruitment & Eligibility

• Status: ENROLLING_BY_INVITATION
• Sex: All
• Target Recruitment: 540

Inclusion Criteria

Not provided

Exclusion Criteria

Not provided

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
The Patient Health Questionnaire (PHQ-9)	Baseline, month 3, month 6, month 9, month 12	The Patient Health Questionnaire (PHQ-9) is a self-report screening tool designed to assess the severity of depressive symptoms. PHQ score range is 0 to 27. Higher PHQ score indicates worse depressive symptoms.
Inter-visit mood episode occurance	month 3, month 6, month 9, month 12	Mood episodes that occur between scheduled study visits will be assessed by trained raters. If an inter-visit episode is suspected, a structured clinical evaluation will be conducted using standardized case report forms (CRFs) to determine: 1. Episode period (start and end dates),; 2. Presence of DSM-5 mood symptoms (e.g., depressed mood, anhedonia, decreased need for sleep, inflated self-esteem),; 3. Episode severity as rated by the Clinical Global Impressions (CGI) scale (1-7; higher scores indicate greater severity).; Scoring and Interpretation: * CGI scale range: 1 (Normal, not ill) to 7 (Among the most extremely ill patients).; * Higher CGI scores represent worse mood symptom severity.; Diagnostic criteria for depressive episodes (≥1 core symptom plus ≥5 symptoms, ≥2 weeks) and (hypo)manic episodes (≥1 core symptom plus ≥3-4 symptoms, ≥4 days) will be applied using the CRF.
MADRS (Montgomery-Åsberg Depression Rating Scale)	Baseline, month 3, month 6, month 9, month 12	The Montgomery-Åsberg Depression Rating Scale (MADRS) is a clinician-administered tool used to assess the severity of depressive symptoms. It consists of 10 items rated on a 0-6 scale and is widely used in clinical trials due to its sensitivity to treatment-related change. MADRS score range is 0 to 60. Higher MADRS score indicates worse depressive symptoms.
YMRS (Young Mania Rating Scale)	Baseline, month 3, month 6, month 9, month 12	The Young Mania Rating Scale (YMRS) is a clinician-rated scale used to measure the severity of manic symptoms in individuals with bipolar disorder. It includes 11 items covering mood, behavior, and cognitive functioning, with higher scores indicating greater symptom severity. YMRS score rang is 0 to 60.; Higher YMRS score indicates worse manic symptoms.
CGI (Clinical Global Impression Scale)	Baseline, month 3, month 6, month 9, month 12	The Clinical Global Impression (CGI) Scale is a standardized assessment tool used by clinicians to rate overall illness severity (CGI-S). CGI-S is rated on a 7-point scale for each of three components: Depression, Mania, Overall. CGI scale range: 1 (Normal, not ill) to 7 (Among the most extremely ill patients).

Trial Locations

Locations (1)

Korea University Anam Hospital; 🇰🇷 Seoul, Seongbuk-gu, South Korea