Smartwatch Detects Depression Relapse Risk Through Sleep & Activity Data

Could a smartwatch be an early warning system for depression relapse? New research from McMaster University suggests that disruptions in a person’s sleep and daily activity, as detected by a simple wrist-worn device, can signal an increased risk of relapse into major depressive disorder (MDD).

The findings, published in February 11, 2026, in JAMA Psychiatry, highlight a potentially powerful, passive way to monitor relapse risk in individuals living with MDD, often detecting the probability of a relapse weeks or even months before symptoms fully return. Approximately 60% of people with MDD experience a relapse within five years, even with ongoing 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?’”

Benicio Frey, Professor, Department of Psychiatry and Behavioural Neurosciences, McMaster University

The study followed 93 adults across Canada who had previously recovered from depression. Participants wore a research-grade actigraphy device – similar to commercially available fitness trackers like Fitbits or Apple Watches – for one to two years, generating over 32,000 days of sleep and activity data. Researchers analyzed this data to identify patterns preceding depressive episodes.

Key Indicators of Potential Relapse

The research identified several key indicators associated with an increased risk of depression relapse. Individuals with a more irregular sleep profile had nearly double the risk. This irregularity wasn’t simply about the amount of sleep, but the consistency of sleep-wake cycles.

Perhaps the strongest predictor of relapse was the degree of difference between daytime activity and nighttime rest. When the body showed less distinction between being active and being at rest, the risk of relapse increased. This suggests that a disruption in the body’s natural rhythms – a blurring of the lines between activity and recovery – can be a significant warning sign.

The amount of time spent awake during the night after initially falling asleep also proved to be a predictive factor. Increased wakefulness after sleep onset was associated with a higher risk of relapse. Participants’ sleep schedules became demonstrably more erratic in the period leading up to a relapse.

Passive Monitoring and the Potential for Personalized Care

This research underscores the untapped potential of wearable technology for individuals recovering from MDD. Unlike traditional monitoring methods that rely on self-reported symptoms – which often appear later in the relapse process – wearable devices collect data passively and continuously, providing ongoing insight between clinical appointments. This continuous data stream could allow for earlier intervention.

Researchers emphasize opportunities for innovation within the healthcare system. Wearable-derived alerts could help clinicians target care to those most at risk, potentially improving outcomes and reducing the burden of recurrent depressive episodes. The ability to passively detect these abnormal patterns using smart sensors opens a new window for personalizing care for conditions prone to recurrence.

MDD is a common and serious medical condition affecting millions worldwide. It impacts how a person feels, thinks, and functions, often manifesting as persistent low mood, loss of appetite, feelings of guilt, and a diminished interest in activities. The cyclical nature of the illness – the risk of relapse – presents a significant challenge to long-term management.

Beyond Sleep: Physiological Parameters and Early Detection

The McMaster University study builds on broader research exploring the use of wearable technology for mental health monitoring. A clinical trial (NCT06789822) is currently investigating a mobile application, Dalia, which uses a smartwatch to monitor a range of physiological parameters – including heart activity, sleep quality, and even self-reported moods – to detect early signs of relapse or recurrence. This suggests a growing trend toward multi-faceted monitoring approaches.

Further research, including a study published in PubMed, demonstrates the role of consumer-grade activity trackers in estimating relapse risk and depression severity in people with recurrent MDD. Variability in sleep duration and midpoint were identified as potentially useful targets for tailored interventions.

The study was supported by the Ontario Brain Institute, Janssen Research &amp. Development, and the Ontario Research Fund – Research Excellence, in partnership with the Canadian Biomarker Integration Network in Depression (CAN‑BIND).