A new study has found that subtle fluctuations in sleep and daily activity could offer a powerful early warning for hypomanic episodes in people with bipolar disorder, potentially opening the door to earlier and more tailored interventions. The findings were published in the Journal of Affective Disorders.
In one of the most comprehensive studies of its kind, researchers tracked 164 participants diagnosed with bipolar I or II disorder over a median period of more than 16 months using wearable technology. The study relied on devices such as Oura rings to continuously capture sleep patterns and activity levels, alongside weekly self-reported assessments of mood changes. From this long-term data, the researchers were able to isolate 85 distinct hypomanic episodes in 50 individuals.
The most striking finding was that day-to-day changes in sleep stages, especially within-night variability, emerged as the most sensitive predictor of an impending episode. Activity fluctuations throughout the day also served as strong indicators, with both types of behavioural changes preceding hypomania by up to three days. These results suggest that disruptions to biological rhythms might be a more consistent precursor to mood shifts than the subjective experience of elevated mood itself.
Rather than relying on averages, the researchers employed a novel analytical method that focused on variability over time. This approach reflects a growing consensus in the mental health field that psychiatric conditions such as bipolar disorder are better understood as dynamic systems that evolve through tipping points, much like natural systems in biology or climate science.
By using finely-grained data collected in real-time, the study also sidesteps a longstanding challenge in bipolar research: many individuals do not recognise the onset of a hypomanic episode until they are well into it. This delay often limits the effectiveness of early treatment and support. In contrast, wearable devices offer an objective, passive way to detect risk signals, which can be particularly useful in high-risk individuals or those with a history of rapid cycling.
The study’s authors note that earlier research has often struggled to identify reliable early markers due to inconsistent data or subjective recall. By tracking over 19 million activity data points and more than six million sleep data points, this latest research significantly boosts the statistical power and granularity available for analysis.
Still, the findings are not without limitations. The sample size, while large for a wearables study, remains relatively modest for drawing broader conclusions across diverse populations. Moreover, factors such as socioeconomic background and lifestyle were not controlled for in this analysis, leaving room for further exploration.
Still, the results mark an important step towards integrating technology into psychiatric care. While not intended to replace clinical judgement, wearable devices could be used to enhance it – offering alerts to patients and professionals alike when patterns start to shift in a concerning direction.
The ability to detect hypomanic episodes earlier may ultimately lead to interventions that are more timely and better tailored, reducing the personal and social costs of delayed diagnosis.
