Can Sleep and Activity Patterns Help Predict Relapse in Depression?
March 27, 2026
Major depressive disorder is often a recurrent condition, with many individuals experiencing relapse even after periods of recovery. Identifying objective, early markers of relapse risk remains a key challenge in psychiatry. A recent CAN-BIND study published in JAMA Psychiatry investigated whether long term patterns of sleep and daily activity, measured using wearable devices, could serve as indicators of impending relapse in individuals with depression.
What are Rest Activity Rhythms (SES)?
Rest activity rhythms are objective measures of the body’s circadian system, reflecting patterns of movement and rest across the 24-hour day. These rhythms are typically measured using actigraphy, a method that uses wrist-worn devices to continuously record activity levels over extended periods.
Key features of rest activity rhythms include their stability, amplitude, and timing. Stability refers to how consistent activity patterns are from day to day. Amplitude reflects the difference between periods of high and low activity, often interpreted as the strength of the rhythm. Timing captures when peak activity and rest occur within the daily cycle. Disruptions in these rhythms, including reduced regularity and altered timing, have been associated with mood disorders and may reflect underlying circadian dysregulation.
How the Study Worked
This longitudinal study followed individuals with a history of major depressive disorder over a one-year period. Participants were in remission at baseline and continuously wore actigraphy devices to capture objective measures of sleep and rest activity rhythms.
The researchers derived multiple rhythm metrics, including measures of interdaily stability, intradaily variability, relative amplitude, and timing of activity. These metrics were analyzed in relation to subsequent depressive relapse, which was assessed prospectively during the follow-up period. By examining temporal changes in these objective markers, the study aimed to determine whether changes in circadian and behavioural rhythms preceded the onset of relapse.
“Sleep and circadian rhythm disturbances are highly prevalent and associated with worse outcomes in MDD, including lower remission rates, increased suicidality, and relapse after successful treatment. Furthermore, disturbances in rest-activity rhythms often emerge soon before clinical onset of a depressive episode. Digital biomarkers obtained through wearable devices have demonstrated bidirectional relationships between depressive symptoms and disturbances in sleep and rest-activity rhythms.”
What Did This Study Reveal?
The findings demonstrated that specific alterations in rest activity rhythms were associated with increased risk of depression relapse.
- Lower interdaily stability, indicating less consistent day to day activity patterns, was linked to higher relapse risk. This suggests that irregular daily structure may reflect vulnerability to recurrence.
- Higher intradaily variability, reflecting greater fragmentation of activity and rest within the day, was also associated with relapse. This pattern indicates more frequent transitions between activity and rest states, consistent with disrupted circadian organization.
- Reduced relative amplitude, indicating a smaller difference between active and rest periods, was associated with relapse, suggesting a blunting of normal daily rhythms.
- Shifts in the timing of activity, including delayed or inconsistent phase patterns, were also observed among individuals who relapsed.
Importantly, these changes were detectable prior to the onset of clinically significant depressive symptoms, supporting their role as potential early markers rather than consequences of relapse.
Why This Matters
These findings provide evidence that objective, continuously measured behavioural rhythms may serve as digital biomarkers for depression relapse. Unlike subjective symptom reporting, actigraphy offers real time data that can capture subtle physiological and behavioural changes. The ability to detect these early warning signals of relapse could enable timely clinical intervention, such as adjusting treatment strategies or increasing monitoring. This approach aligns with efforts to develop more proactive and personalized models of mental health care.
“Most current relapse prediction models use clinically oriented features, mainly symptom severity and symptom dimensions, but the models that have been critically evaluated have shown poor predictive capacities. Actigraphy-derived measures can provide indications of underlying biological processes, which cannot be fully captured by clinical assessments alone. Importantly, the use of such passive sensing technology can inform specific therapeutic targets to reduce relapse vulnerability in remitted patients…“
Limitations to Consider
Although actigraphy provides continuous and objective data, it infers sleep and rest from movement and does not directly measure neurophysiological sleep processes. The study design is also observational, meaning that while associations were identified, causality cannot be established between rhythm disruption and relapse. Additionally, external factors such as psychosocial stressors, medication changes, and lifestyle behaviours may influence both activity patterns and relapse risk, but were not fully controlled. Future research is needed to validate these findings in larger and more diverse samples and to determine whether interventions targeting circadian rhythms can reduce the risk of relapse in depression.
Final Takeaway
The findings from this study suggest that changes in daily sleep and activity patterns may provide early warning signs that depression could return. Subtle shifts in how regular, active, or consistent a person’s daily routine is can appear before noticeable symptoms come back. With further research, tracking these patterns using wearable devices could help clinicians identify risk earlier and take steps to prevent relapse, moving toward more proactive and personalized care for depression.
Citation: Tonon, A. C., Nexha, A., Cunningham, J. E. A., d’Eon, J., Chakrabarty, T., Farzan, F., Foster, J. A., Harkness, K. L., Hassel, S., Ho, K., Lam, R. W., Milev, R., Minuzzi, L., Müller, D. J., Nunes, A., Parikh, S. V., Quilty, L. C., Rotzinger, S., Soares, C. N., … Frey, B. N. (2026). One-year actigraphy study of sleep and rest-activity rhythms as markers of relapse in depression. JAMA Psychiatry, 83(4), 379–388. https://doi.org/10.1001/jamapsychiatry.2025.4453