A comprehensive study involving over 95,000 participants has found strong associations between specific sleep patterns and long-term health outcomes. Published on September 17 in the open-access journal PLOS Medicine, the research indicates that individuals who spend more time in rapid eye movement (REM) sleep face a significantly lower risk of developing 83 different diseases.
The study was led by Shengzhi Sun of Capital Medical University in China. Researchers analyzed data from the UK Biobank, where participants wore wrist accelerometers for seven consecutive nights to objectively measure sleep stages, including REM, deep, and light sleep, as well as total duration and irregularity. Participants were monitored for a median of 8.9 years, allowing researchers to correlate sleep metrics with health records covering more than 1,000 disease outcomes.
According to the findings, greater REM sleep—defined per interquartile range as an increase of 47.6 minutes—was linked to reduced incidence of serious conditions such as heart failure, dementia, and Parkinson’s disease. Specifically, the hazard ratio for dementia was recorded at 0.54, and for Parkinson’s disease, it was as low as 0.20. In contrast, increased deep sleep was associated with a lower risk of only seven conditions, notably type 2 diabetes and major depressive disorder.
The research also highlighted the dangers of insufficient sleep. Those who slept less than five hours per night exhibited the highest clinical vulnerability, facing an elevated risk of 37 distinct conditions. For many diseases, total sleep duration showed a non-linear relationship with health risks, with the lowest risk concentrated in the six-to-eight-hour window.
Lead author Shengzhi Sun and colleagues emphasized that these findings support current health guidelines recommending six to eight hours of sleep for middle-aged and older adults. “Sleep duration exhibited significant non-linear associations with 86 disease phenotypes, with the minimum-risk duration for the majority of these conditions precisely concentrated within a 6-8 hour window,” the authors wrote.
The focus on sleep health is particularly relevant in China, where the 2026 China Sleep Health White Paper reports that the average person sleeps 6.97 hours per night, yet 23 percent sleep less than six hours. Furthermore, only 26 percent of respondents maintain a consistent sleep schedule, and sleep disturbances have been increasing. The study utilized objective accelerometer data to overcome the limitations of self-reported sleep measures, which often correlate poorly with actual sleep patterns.
While the study identified 156 significant associations between sleep patterns and incident diseases after statistical corrections, the authors noted that as an observational study, it cannot definitively prove that sleep patterns directly cause changes in disease risk. Nonetheless, the results provide substantial evidence for the role of balanced sleep stages and adequate duration as safeguards against a wide array of chronic illnesses.
I slept five hours last night and now I’m reading this. Guess I’m joining the 37-condition club until I fix my routine.
My doctor said six to eight hours was optimal, but this adds real depth to that advice. Quality really does matter as much as quantity.
6.97 hours average in China is concerning. If we can’t even hit seven hours, how do we optimize REM stages?
Correlation isn’t causation, folks. We don’t know if REM prevents disease or if healthier people just sleep better. Don’t oversell it.
Wait, so more REM equals less dementia risk? I might start dreaming harder about walking without pain. Seriously though, fascinating study.
Finally, someone proves deep sleep isn’t the only thing that matters. I’ve been ignoring my REM cycles for too long.
This is huge news! I always knew REM sleep was important, but 83 diseases? I’m setting my bedtime earlier tonight.