More REM, Better Health? Sleep Stages Predict Disease Risk

Night after night our bodies rehearse the next day. Quietly, in the dark, the architecture of sleep appears to write a ledger of future health.

How the study was done and what it measured

A team at Peking University turned to the UK Biobank to read that ledger. They analyzed wrist accelerometer data from 95,559 participants who wore devices for seven consecutive days. These sensors do not capture brain waves, but they can estimate time in different sleep stages, total sleep time, how long people stayed awake after first falling asleep, and how regular their sleep times were across nights. Participants were then tracked for an average of roughly nine years, with researchers testing associations between those sleep metrics and more than a thousand health outcomes recorded in medical records.

The headline finding was striking. Participants with more rapid eye movement sleep, commonly called REM sleep, had a lower risk of developing 83 distinct diseases later in life. The list ranges from cardiovascular to neurodegenerative conditions. Deep slow wave sleep also mattered, but to a narrower set of outcomes.

Which conditions were linked to sleep architecture

Rather than a single disease, the patterns point to broad vulnerability or resilience tied to sleep. Examples of illnesses associated with higher REM sleep included:

• Heart failure
• Dementia
• Parkinsons disease

More deep sleep was connected to lower risk for seven conditions, among them type 2 diabetes and major depressive disorder. By contrast, erratic sleep timing and longer awake periods after initially falling asleep correlated with increased risk for anxiety disorders and substance-related disorders.

There was also a clear signal about duration. People who usually slept six to eight hours per night showed the lowest overall disease risk. Sleeping less than five hours was linked to higher vulnerability across 37 conditions. Sleeping longer than eight hours was not harmless either and associated with elevated risk for some diseases.

Publication: the study appears in PLOS Medicine.

Why might different stages matter?

Sleep is not uniform. Deep slow wave sleep supports metabolic cleanup and restorative processes. REM sleep appears central to emotional processing and complex brain maintenance. If one stage is reduced, it could signal or cause disruption in systems that regulate metabolism, inflammation, and brain health.

But caution is needed. This is observational work. Accelerometers infer rather than directly measure sleep stages, and association does not prove causation. Short sleep might be a symptom of underlying disease or stressors that later lead to ill health. Still, the size of the dataset and the consistency of patterns across many outcomes strengthen the argument that how we sleep matters, not just how long.

What this means for public health and individuals

For clinicians and policymakers the implications are broad. Sleep assessments that go beyond duration could help flag patients at higher long-term risk. For individuals the messages are simple: aim for regular sleep, aim for six to eight hours, and talk to a clinician if you suffer persistent insomnia, fragmented sleep, or unusually long sleepiness.

Ultimately, sleep may be both a marker and a mechanism. It tells us about current physiological resilience and, potentially, helps build it. The study adds weight to the idea that preserving healthy sleep architecture should be part of disease prevention strategies.