REM Sleep and Disease Risk: What the 83-Disease Study Found

Protect a regular sleep window of seven or more hours, keep it steady, and let your sleep stages sort themselves out. That’s the practical takeaway from a new study of more than 95,000 adults that linked more REM sleep to a lower risk of 83 diseases. The headline number is striking, but the tool behind it can only estimate sleep stages, so the useful message is about total, consistent sleep rather than chasing one stage.

What the study did

Researchers used data from 95,559 UK Biobank participants, average age 56, who wore a wrist accelerometer (a movement sensor) for a week between 2013 and 2015. An algorithm called SleepNet turned that movement data into estimates of total sleep, REM sleep, light sleep and deep sleep. The team then followed participants for a median of 8.9 years and checked those sleep estimates against 1,049 different health outcomes.^[1]

REM (rapid eye movement) sleep is the stage when most vivid dreaming happens. Deep sleep, also called slow-wave or N3 sleep, is the stage when your brain activity is slowest and you’re hardest to wake.

What it found

The researchers estimated risk per interquartile-range increase in REM sleep, which works out to about 48 extra minutes of estimated REM a night. More REM was associated with a lower risk of 83 conditions across 12 disease categories. Those included heart failure (a hazard ratio of 0.74, or roughly 26% lower risk), atrial fibrillation (0.83) and dementia (0.54). A hazard ratio below 1 means the event happened less often with more REM over the follow-up period. More deep sleep was associated with a lower risk of 7 conditions, including type 2 diabetes (0.89) and major depressive disorder (0.86).^[1]

Total sleep mattered too. For 69 conditions, the lowest risk sat mostly in the 6-to-8-hour range. People sleeping under 5 hours a night accounted for 37 of the 41 higher-risk associations the team found when comparing sleep-length groups with the 6-to-8-hour group. Day-to-day variation in sleep duration was linked to higher risk of 3 conditions.^[1]

Why to read it with care

• It shows a link, not a cause. This is an observational study. The authors state plainly that it can’t prove sleep stages caused the differences in disease risk, and that unmeasured factors could explain some of them.
• The sleep stages are estimates. A wrist sensor reads movement, not brain waves. The authors note that SleepNet tends to underestimate REM and overestimate non-REM sleep compared with a lab sleep study, and that the numbers should be read as movement-based estimates.
• Illness may have come first. When the team dropped the first two years of follow-up to reduce the chance that early, undiagnosed disease was already disturbing sleep, 95 of the original 156 associations held up. The rest didn’t.
• One week is a snapshot. Sleep was measured once, for seven days. Diagnoses came only from hospital inpatient records, and UK Biobank volunteers tend to be healthier and less diverse than the general population.

None of that makes the study useless. It adds a large, objective data set to a long line of research tying poor sleep to poor health. It just doesn’t mean that adding REM minutes on your tracker will prevent any specific disease.

What to do with this

• Aim for seven or more hours of sleep, most nights. The American Academy of Sleep Medicine and Sleep Research Society recommend that adults sleep 7 or more hours per night on a regular basis.^[2] Work backward from your wake-up time and set a lights-out reminder.
• Keep the timing steady. In this study, bigger day-to-day swings in sleep duration showed up as a risk marker. The study did not measure bedtime or wake time, so treat a steady schedule as general advice rather than a finding from this paper. Aim for roughly the same bedtime and wake time on workdays and days off.
• Go easy on alcohol near bedtime. A review of studies in healthy volunteers found that alcohol delays the first REM period at all doses, and that moderate to high doses reduce total REM across the night.^[3]
• Don’t chase your tracker’s REM score. This is our inference, not a finding of the study: if a research-grade algorithm can’t measure stages precisely, a consumer wearable is a rough guide at best. Watch the trend in total sleep and consistency, and notice if checking the data starts making you anxious. That pattern has a name: orthosomnia.
• Talk to a clinician if sleep stays short or broken. Regularly sleeping under five hours, loud snoring, or waking unrefreshed despite enough time in bed are worth raising with your doctor.

You don’t need to engineer the perfect night. Protect the window, keep it regular, and let your body run its own cycles. That covers most of what this study, and the research before it, supports.

References

[1] Luo J, Liu R, Yin J, Cao W, Sun S, Chen R. Accelerometer-derived real-world sleep stages and risk of incident diseases: A UK Biobank cohort study and phenome-wide association analysis. PLOS Medicine, 2026;23(9):e1005213. Population: 95,559 UK Biobank participants, mean age 56.2, wrist accelerometer worn 7 days (2013-2015), median follow-up 8.9 years. DOI: https://doi.org/10.1371/journal.pmed.1005213. Finding as used: per interquartile range of REM sleep (47.6 min), lower risk of 83 incident diseases, including heart failure (HR 0.74, 95% CI 0.68-0.80), atrial fibrillation (HR 0.83, 0.79-0.86) and dementias (HR 0.54, 0.47-0.62); per interquartile range of deep sleep (47.5 min), lower risk of 7 diseases, including type 2 diabetes (HR 0.89, 0.85-0.93) and major depressive disorder (HR 0.86, 0.82-0.91); minimum-risk sleep duration for 69 phenotypes concentrated within 6-8 hours; under 5 hours accounted for 37 of 41 adverse associations versus 6-8 hours; irregularity in daily sleep duration (standard deviation) linked to higher risk of 3 diseases; 95 of 156 associations remained significant after a 2-year washout; authors state the design precludes causal inference and that SleepNet underestimates REM and overestimates NREM versus polysomnography. ↑︎

[2] Watson NF, Badr MS, Belenky G, et al. Recommended Amount of Sleep for a Healthy Adult: A Joint Consensus Statement of the American Academy of Sleep Medicine and Sleep Research Society. J Clin Sleep Med, 2015;11(6):591-592. Population: adults (expert consensus statement). DOI: https://doi.org/10.5664/jcsm.4758. Finding as used: adults should sleep 7 or more hours per night on a regular basis to promote optimal health. ↑︎

[3] Ebrahim IO, Shapiro CM, Williams AJ, Fenwick PB. Alcohol and Sleep I: Effects on Normal Sleep. Alcohol Clin Exp Res, 2013;37(4):539-549. Population: qualitative review of studies in healthy volunteers. DOI: https://doi.org/10.1111/acer.12006. Finding as used: onset of the first REM period is significantly delayed at all alcohol doses; total-night REM percentage is decreased in the majority of studies at moderate and high doses, with no clear trend at low doses. ↑︎

Kurt Greiner

Kurt is a digital strategist and IT professional blending emerging technology with practical application to help businesses and individuals streamline their digital presence. His current work focuses on the intersection of intentional living and technological resilience, exploring how individuals can leverage modern tools to navigate the second half of life with purpose.