Predicting Affective Episodes Through Digital Monitoring of Circadian Rhythm Disturbances: A Systematic Review of Modern Technology Applications

Introduction: Affective disorders are serious, long-lasting, and often relapsing conditions that strike an individual unpredictably. The ideal psychiatric monitoring mechanism would predict coming episodes before their full-blown clinical symptoms develop. This would be a tremendous advance and may be possible using our current understanding of the nature of affective disorders, portable device-based data collection capabilities, and remote data analysis. There are numerous reviews of the literature that cover the symptom recognition of affective disorders using mobile devices, but there are no reviews on the actual prediction of affective episodes. This review aims to cover that gap. Methods: A systematic review of five databases (2016-2025) was conducted and included 10 observational studies, mostly designed as prospective cohorts, that concentrated on the prognosis of affective disorders and utilized wearable devices. The variables under study were circadian parameters, sleep metrics, amounts and intensity of physical activity, and light exposure. Although the methodological diversity among the studies made direct comparisons problematic, the studies allowed for the identification of certain findings that appear promising for predicting the occurrence of affective episodes. Results: The review encompassed 10 observational studies (1416 subjects). Our synthesis showed that it is possible, even feasible, to predict affective episodes using mathematical models. These models assess the types of characteristic disturbances that individuals with affective disorders have in their circadian rhythms. When we applied some standard methods for doing this kind of analysis (Accuracy Metrics and AUC Values), the results gave us AUC values that ranged from about 0.67 to 0.98, depending on several factors. Conclusions: Predicting affective episodes is possible using wearable technology or smartphone app, which can detect disturbances in a person's circadian rhythm. Currently, the best methods for doing this merit a look because they could indeed allow for early intervention before the onset of manic or depressive symptoms.