A large prospective study published in PLOS Medicine on September 17, 2026, links rapid eye movement (REM) and deep sleep stages—rather than just total sleep duration—with the future risk of various health conditions. The research tracks nearly 96,000 participants over almost nine years, offering new data on how specific sleep architecture relates to long-term disease outcomes, according to findings discussed by independent experts.
Study Design and Wearable Technology Methods
The research evaluates data from nearly 96,000 participants over a follow-up period of almost nine years, incorporating stringent corrections for testing more than 1,000 health outcomes alongside several sensitivity analyses, according to Dr. Nina Rzechorzek, MRC Clinician Scientist | Brain Temperature & Neural Circadian Rhythms, University of Cambridge. Participants wore monitors for seven consecutive days, providing a population-scale view of habitual sleep patterns.
However, the specific sleep stages analyzed in the paper were not measured directly using gold-standard polysomnography. Instead, these stages were inferred by a deep-learning model relying on wrist movement captured by accelerometers, as noted by Dr. Rzechorzek. Dr. Greg Elder, Assistant Professor in Psychology and Associate Director of Northumbria Centre for Sleep Research, Northumbria University, points out that while wristwatches do an effective job of determining whether an individual is asleep or awake, they do not explicitly measure brain activity or eye movements required to definitively confirm REM or deep sleep.
Expert Analysis on Limitations and Reverse Causation
Independent specialists highlight critical boundaries regarding what the data can prove. Dr. Rzechorzek cautions that reverse causation remains a significant concern within the findings. When the study authors extended their exclusion period to two years, only 95 of the original 156 significant associations remained. This shift suggests that early or undiagnosed disease may already have begun altering a participant’s sleep prior to an official clinical diagnosis.
Furthermore, Dr. Elder emphasizes that the observational design means the research can only demonstrate associations between sleep aspects and health conditions rather than proving direct causation. While short sleep and other sleep disturbances frequently accompany a wide range of illnesses, the study does not establish whether those alterations actively cause the conditions or merely correlate with them. Both researchers note that press release claims suggesting that maintaining a strict six-to-eight-hour sleep window or actively increasing REM and deep sleep can effectively reduce disease risk go beyond what the current study design can establish.
Keep reading