The new research highlights a simple, yet powerful way to passively monitor relapse risk in people living with major depressive disorder (MDD), often detecting the probability of a relapse weeks or months before the episode occurs. Approximately 60% of people with MDD relapse within five years, even with treatment.

“Advances in digital technology and AI algorithms have a great potential for relapse prevention in mental health. Imagine a future where a smartwatch can warn people with depression: ‘A new episode of depression is very likely coming within the next four weeks. How about seeing your health-care provider?'” says Benicio Frey, professor in the Department of Psychiatry and Behavioral Neurosciences at McMaster.

The study, published in JAMA Psychiatry, followed 93 adults across Canada who had previously recovered from depression and wore a research-grade actigraphy device, similar to a Fitbit or Apple Watch. Participants wore the wearable device for one to two years, generating over 32,000 days of sleep and activity data.