A large study published in PLOS Medicine on 17 September 2026 suggests that sleep patterns, including estimated REM and deep sleep, could be informative markers of future disease risk. The research, which involved nearly 96,000 participants over almost nine years, used a deep-learning model to infer sleep stages from wrist movement.
Experts noted that while the study is robust in its analysis and scale, the sleep stages were estimates rather than direct measurements. The authors of the study acknowledged that these should be regarded as movement-based estimates. Additionally, the study can only show associations between sleep aspects and health conditions, not causation.
Dr. Greg Elder, Assistant Professor in Psychology at Northumbria University, highlighted that the study adds to evidence showing the importance of objectively measured sleep for health. He stated that a regular sleep pattern with sufficient sleep is important for physical and psychological health as people age.