The Challenge of Relapse in Major Depressive Disorder
Major depressive disorder (MDD) is a chronic and recurrent condition that imposes a staggering burden on global health systems. Despite the availability of various pharmacotherapeutic and psychotherapeutic interventions, a significant proportion of patients experience a relapsing-remitting course. Predicting these relapses remains one of the most formidable challenges in clinical psychiatry. Historically, clinicians have relied on subjective patient self-reports or intermittent clinical interviews—methods that are often hindered by recall bias and the late appearance of symptoms.
There is an urgent clinical need for objective, continuous, and non-invasive biomarkers that can signal an impending depressive episode before a full clinical relapse occurs. Recent advances in wearable technology, specifically actigraphy, have opened new avenues for digital phenotyping. By monitoring rest-activity rhythms (RARs) and sleep-wake cycles in a naturalistic environment, actigraphy provides a high-resolution window into the biological rhythms that are frequently disrupted in mood disorders.
Study Overview: The Tonon et al. Longitudinal Cohort
In a comprehensive observational cohort study published in JAMA Psychiatry (2026), Tonon and colleagues investigated whether actigraphy-derived parameters could serve as reliable markers for MDD relapse. The study, conducted from July 2016 to January 2019, followed a multicentric sample of 102 adults from psychiatric and primary care clinics across Canada.
Participants were required to have a diagnosis of MDD and be in a state of relative clinical stability at baseline, defined by a Montgomery-Åsberg Depression Rating Scale (MADRS) score of 14 or less. The researchers utilized continuous actigraphy data, averaged every two weeks, over a follow-up period of one to two years. This longitudinal approach allowed for the observation of approximately 32,000 complete actigraphy days, providing an unprecedented depth of data on the temporal dynamics of rest and activity.
Defining Relapse and Clinical Course
The primary outcome of the study was relapse, which was rigorously defined and adjudicated by an independent panel. Relapse criteria included a MADRS score of 22 or higher for two consecutive weeks, psychiatric hospitalization, the emergence of suicidal intent or behavior, or the necessity for antidepressant treatment escalation.
To further refine the analysis, the researchers categorized the clinical courses of participants into three distinct groups:
1. Relapse: Those meeting the formal criteria for a new depressive episode.
2. Ultrastable: Participants who maintained a MADRS score below 14 throughout the entire study period.
3. Unstable: Participants who exhibited transient MADRS scores between 14 and 22 but did not meet the full threshold for relapse.
Key Findings: Baseline Markers of Vulnerability
The results of the Cox proportional hazards models, adjusted for age, sex, season, and baseline MADRS scores, identified several actigraphy parameters as significant predictors of relapse.
Sleep Regularity and Efficiency
Baseline sleep regularity was strongly associated with long-term stability. Specifically, lower sleep regularity at the start of the study was a harbinger of future relapse (Hazard Ratio [HR], 0.46). Similarly, lower sleep efficiency (HR, 0.57) and higher levels of wake after sleep onset (WASO; HR, 1.77) were significant risk markers. These findings suggest that the fragmentation of sleep and the lack of a consistent sleep-wake schedule are not merely symptoms of an active episode but may represent a trait-like vulnerability to MDD recurrence.
Relative Amplitude and Nighttime Activity
One of the most robust predictors identified was Relative Amplitude (RA), which measures the difference between activity during the most active 10 hours and the least active 5 hours of the day. A lower RA, indicating a ‘flatter’ rest-activity rhythm, was significantly associated with a higher risk of relapse (HR, 0.45). Conversely, higher nighttime activity (HR, 1.86) was also a strong predictor of clinical destabilization. Essentially, a dampened rhythm—where the distinction between daytime alertness and nighttime rest is blurred—serves as a critical warning sign for clinicians.
Longitudinal Dynamics and Time-Varying Risk
Beyond baseline measurements, the study utilized time-varying models to assess how changes in actigraphy parameters over time correlated with the risk of relapse.
Greater composite phase deviation (HR, 1.76) and consistently low RA (HR, 0.45) remained significant in these models. Remarkably, the association between low Relative Amplitude and relapse remained significant (HR, 0.60) even after the researchers adjusted for concurrent MADRS scores. This indicates that changes in rest-activity rhythms are not simply a reflection of worsening mood but may actually precede or independently contribute to the relapsing process.
Furthermore, actigraphy measures successfully differentiated individuals who experienced a relapse from those with an ultrastable or unstable clinical course. This suggests that actigraphy can distinguish between benign fluctuations in mood and the more profound physiological shifts that lead to clinical relapse.
Mechanistic Insights: The Circadian-Mood Nexus
The findings by Tonon et al. align with the growing body of literature regarding the ‘circadian clock’ theory of depression. The suprachiasmatic nucleus (SCN) of the hypothalamus regulates nearly all physiological processes, including the sleep-wake cycle, hormone secretion, and mood. In MDD, the molecular machinery of the circadian clock is often ‘de-synchronized’ or dampened.
A robust Relative Amplitude represents a healthy, entrained circadian system. When this amplitude decreases, it often reflects a weakening of the SCN’s output or a lack of sensitivity to external ‘zeitgebers’ (time-givers) like sunlight and social activity. The association between sleep phase variability and relapse further highlights the importance of ‘rhythmopathy’ in psychiatry. When the timing of sleep shifts frequently (high phase deviation), it creates a state of internal desynchrony, which can impair emotional regulation and cognitive function, ultimately lowering the threshold for a depressive relapse.
Expert Commentary and Clinical Implications
This study represents a significant step toward the ‘precision psychiatry’ model. By using actigraphy, clinicians can potentially identify high-risk individuals during periods of apparent clinical remission.
Scalability and Monitoring
Unlike polysomnography, which is expensive and labor-intensive, actigraphy is highly scalable. With the ubiquity of consumer-grade wearables and medical-grade sensors, continuous monitoring is now feasible for a broad patient population. The ability to detect a decline in Relative Amplitude or an increase in sleep irregularity in real-time could trigger automated alerts for both the patient and the physician.
Targeted Interventions
Identifying these markers allows for timely, personalized interventions. For instance, a patient showing decreased sleep regularity could be prescribed Social Rhythm Therapy (SRT) or light therapy to stabilize their circadian system before their mood deteriorates. This shift from reactive to proactive care could significantly reduce the disease burden of MDD.
Study Limitations
While the study is robust, it is important to note certain limitations. The sample was a referred population from Canadian clinics, which may limit generalizability to other demographic groups or healthcare settings. Additionally, while actigraphy is an excellent proxy for sleep-wake cycles, it does not provide the architectural sleep data (such as REM vs. NREM stages) that polysomnography offers. Future research should focus on integrating actigraphy with other digital biomarkers, such as heart rate variability (HRV) and smartphone usage patterns, to create a more comprehensive risk profile.
Conclusion
The study by Tonon et al. provides compelling evidence that actigraphy-derived measures of sleep phase variability and daily activity amplitude are potent markers of depressive relapse. By quantifying the ‘invisible’ disruptions in biological rhythms, actigraphy offers a scalable and objective tool for clinicians to identify patients at the highest risk of recurrence. As we move toward a more digitally integrated healthcare system, these biomarkers will likely play a central role in enabling timely, personalized, and preventive strategies for managing major depressive disorder.
References
1. Tonon AC, Nexha A, Cunningham JEA, et al. One-Year Actigraphy Study of Sleep and Rest-Activity Rhythms as Markers of Relapse in Depression. JAMA Psychiatry. 2026 Feb 11:e254453. doi: 10.1001/jamapsychiatry.2025.4453.
2. Alloy LB, Ng TH, Titone MK, Boland EM. Circadian Rhythm Disruption in Bipolar Mania and Depression: Current Status on Mechanisms and Treatment. Curr Opin Psychol. 2017;14:31-36.
3. Walker RK, et al. Digital Phenotyping in Depression: Maintainance and Relapse Prediction. Lancet Psychiatry. 2020;7(12):1010-1012.
4. Lam RW, et al. Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines for the Management of Adults with Major Depressive Disorder. Can J Psychiatry. 2016;61(9):510-523.