JAMA Psychiatry Study Finds Wearable Sleep-Activity Data Can Forecast Depression Relapse

Overview

• Published Feb. 11, 2026, the McMaster-led study analyzed research-grade actigraphy in 93 adults with remitted major depression across Canada.
• Participants wore devices a mean of 46 weeks, producing about 32,000 person-days of data, with readings from the two weeks before and after relapse excluded to limit reverse causation.
• Lower sleep regularity was linked to nearly double the relapse risk, and the strongest predictor was a smaller gap between daytime activity and nighttime rest (lower relative amplitude).
• Other significant markers included lower sleep efficiency, more time awake after sleep onset, higher nighttime activity, and greater composite phase deviation, with reported hazard ratios up to 1.86.
• Researchers say passive monitoring could support early warnings and targeted outreach between visits, and the work was supported by the Ontario Brain Institute, Janssen R&D, the Ontario Research Fund, and CAN‑BIND.