Subscribe to our "Population in Perspective" Substack Here!
Sleep patterns differ more between individuals at risk of psychosis, and between individuals with schizophrenia spectrum disorders, than they do between healthy individuals, according to new research led by researchers from the Leverhulme Centre for Demographic Science, with colleagues at Oxford Health NHS Foundation Trust, the University of Oxford's Department of Psychiatry, and the Digital Psychiatry Unit at Harvard Medical School.
The study, published in JAMA Network Open, suggests that looking beyond average sleep patterns to differences between individuals could help researchers identify more informative digital markers of psychosis risk and progression.
Sleep disturbances are common among people experiencing psychosis and can emerge before more overt clinical symptoms. However, most previous research has focused on whether people with psychosis sleep more or less, on average, than people without the condition.
The new study, co-authored by LCDS researchers Rosario Aronica and Melinda Mills, instead examined how much sleep patterns differed between people within these groups.
Rosario Aronica said:
"Averages can hide an important signal. Two groups can have very similar average sleep, while one of them contains both people sleeping very little and people sleeping much longer. By looking at variation between individuals, we can start to identify patterns that may be useful for understanding the different ways psychosis develops and affects people."
The researchers conducted a systematic review and meta-analysis of 18 studies involving 1,358 participants, including 202 people considered at clinical high risk for psychosis, 574 people with schizophrenia spectrum disorders, and 582 healthy control participants.
All studies used wrist actigraphy - wearable devices that continuously measure movement and can provide information about sleep in everyday settings.
People at clinical high risk for psychosis showed greater variation between individuals in time spent in bed, wakefulness after falling asleep, and sleep efficiency than healthy controls. Similar patterns were found among people with schizophrenia spectrum disorders, who also showed greater variation in total sleep time.
The findings show why focusing only on group averages can miss potentially important information. Previous research, for example, found little difference in average time spent in bed between people at clinical high risk for psychosis and people with schizophrenia, compared to healthy controls. The new analysis shows that, despite similar averages, sleep patterns vary more within clinical groups of people with psychosis.
The researchers say this variation could point towards distinct groups of people with different sleep profiles, although individual-level and longitudinal data will be needed to establish such subgroups.
Wearable technologies could prove particularly valuable for future research. By continuously monitoring sleep outside clinical settings, devices such as wrist actigraphs can help researchers investigate whether particular sleep patterns are associated with the onset, progression or recurrence of psychosis.
The authors stress that the findings do not allow using sleep variability to predict psychosis in individuals. Instead, they support further research into sleep as a potential digital biomarker that could ultimately contribute to more personalised approaches to care.
This work was supported by the Wellcome Trust.
