Digital Phenotypes for Predicting Depression_Validation

This 12-month, single-site prospective observational study will prospectively monitor the recurrence of mood episodes while continuously collecting passive digital phenotypes in individuals with and without mood disorders. A total of 200 adults will be enrolled, including 100 patients with major depressive disorder or bipolar I/II disorder and 100 healthy control participants. Participants will complete five clinical assessments at 3-month intervals and will continuously wear a consumer wearable device and use the My Mental HERO smartphone application for passive sensing and daily mood recording. Structured clinical interviews will identify the occurrence, subtype, and day-level onset and offset of mood episodes during follow-up. The study will examine longitudinal digital phenotypes associated with mood states and subsequent mood episode onset. It is designed to provide an independent validation dataset for digital phenotypes and prediction models developed in the preceding MentalHERO prospective observational cohort (ClinicalTrials.gov Identifier: NCT07151846). Because the two cohorts follow closely aligned study protocols and assessment schedules, data from both cohorts may also be pooled for secondary analyses where appropriate.