A large study suggests that sleep stages may offer important clues about long-term health, beyond the total number of hours spent asleep. Researchers analyzed data from 95,559 UK Biobank participants, whose wrist movements were recorded for seven days between 2013 and 2015. A machine-learning model used the recordings to estimate sleep stages.
Over a median follow-up of 8.9 years, participants who spent more time in rapid eye movement (REM) sleep--the stage associated with vivid dreaming--were less likely to develop 83 conditions. These included heart failure, high blood pressure, dementia, Parkinson's disease and multiple sclerosis. The analysis found that associations with 54 conditions remained significant when REM sleep was considered as a proportion of total sleep time.
For participants whose estimated REM sleep differed by about 48 minutes per night, the group with more REM had a 26% lower relative hazard of heart failure and a 46% lower relative hazard of dementia. These figures describe associations in the study; they do not show that increasing REM sleep would prevent disease. More deep sleep was also associated with a lower risk of seven conditions, while sleep duration was linked to a range of health outcomes.
Findings need cautious interpretation
The research, led by teams from Peking University and Capital Medical University, was published in PLOS Medicine. It cannot establish cause and effect: health conditions may affect sleep, and other factors could influence both. Sleep stages were estimated from wrist movement rather than directly measured brain activity, and participants were monitored for just one week. The model's accuracy was moderate, and the study's participant group may limit how widely the results apply.
The findings reinforce that sleep quality involves more than hours alone, while underscoring the need for further research. Better long-term sleep measurements could help clarify how sleep patterns relate to health and support more tailored approaches to sleep care.
