How much sleep is actually optimal for healthy aging?

Modern lives, for many of us are spent behind screens. If you’re curious to quantify the number of hours, check your weekly screen time. Screens constantly demand more of our time and attention to keep up with overflowing to-do lists. To counteract this, we’re being told we should aim for 10,000 steps every day, clock at least 150 minutes of moderate activity every week (that’s about 4 days at the gym assuming your workouts average 40 minutes), eat healthily, and try to reduce stress. This can easily read as: fit more into an already packed schedule. It’s no wonder that sleep itself seems to fill whatever hours have remained unscheduled in our busy, and short, 24 hours.

The focus of this article is a recent and highly regarded study published in Nature that’s been making headlines, and for good reason. Rather than simply asking whether sleep matters for health (we already know it does), an international team of researchers built a detailed map of how sleep duration relates to the biological age of multiple organs and systems in the body, offering some of the most fine-grained evidence yet on what an “optimal” amount of sleep might actually look like for different people, but also for different parts of our bodies (not all organs age at the same rate!).

What is known and what does this new study contribute?

Most sleep guidance boils down to a single recommended range, albeit tailored to age groups. The National Sleep Foundation updated their 10-year landmark review in 2015, and found the same results; that adults aged 18 to 65 years need between 7 and 9 hours of sleep, whilst adults aged 65+ require 7 to 8 hours. They also found no significant differences in the number of hours of sleep by gender. To get to these results, they meta-reviewed 133 reviews, which is a substantial body of scientific literature to sift through. However, a question addressed by this recent Nature article, is how does sleep influence aging, and does this have the same impact on different organs and systems in the body? Using health data from over 500,000 UK Biobank participants (ages 37 to 84), they set out to identify what the ideal sleep range is.

How did the researchers measure biological aging?

The concept behind a “biological aging clock” is fairly intuitive. Using brain scans (MRI) and blood tests measuring thousands of proteins and metabolites, researchers trained statistical models to predict a person’s age from the state of a specific organ or system. When a model’s age estimate for someone’s brain, liver, or immune system, for example, is higher than their actual (chronological) age, that gap is used as a marker of accelerated biological aging in that organ. When it is lower, that organ appears to be aging more slowly than expected. In this study, the researchers built 23 such clocks: 7 derived from MRI scans of organs (including the brain, heart, liver, pancreas, kidney, spleen and fat tissue), 11 derived from blood proteins, and 5 derived from blood metabolites. These gaps cannot be interpreted on their own (i.e., they cannot be used for diagnostic purposes), but higher than expected biological age has previously been linked to greater risk of disease and death. Though worth remembering that association does not imply causation!

What did they find?

Sleep and biological age follow a U-shape, not a straight line. Across 9 of the 23 clocks, (spanning the brain, lungs, liver, immune system, skin, hormones, fat tissue, and pancreas) both too little and too much sleep were associated with an older than expected biological age for that organ or system. In other words, more sleep isn’t automatically better; there appears to be a middle zone that looks best.
The ideal duration clustered around 6.4 to 7.8 hours. The exact ideal point varied organ and by sex, but across the clocks it mostly fell between roughly 6.5 and 8 hours, which is reasonably close to general public health guidance (by the National Sleep Foundation mentioned above).
Short and long sleep were both linked to a higher risk of death and disease. Compared with people sleeping 6 to 8 hours, those sleeping under 6 hours had about 50% higher risk of death from any cause, and those sleeping more than 8 hours had about 40% higher risk. Both patterns were also linked to a higher rate of over 150 different health conditions spanning nearly every organ system, including heart disease, type 2 diabetes, depression, anxiety, back pain, digestive issues, and respiratory disease.
Short sleep looked more like a direct risk to aging, whilst long sleep seemed to be more predictive of specific conditions. Short sleep was tied to a broader, more consistent set of health problems across the body. Long sleep, by contrast, was more narrowly and strongly tied to brain-related and psychiatric conditions. Further analysis suggested that long sleep’s link to late-life depression tends to run indirectly, through changes in brain and other organ aging clocks, rather than long sleep itself being the cause. This raises the possibility that, in at least some cases, long sleep is likely a marker of an underlying issue.
The pattern looks more environmental than genetic (encouraging!). When the researchers looked at genetic risk scores for these same aging clocks, they did not see the same U-shaped pattern that shows up with actual sleep duration. This suggests the sleep-aging relationship is driven more by habits and environment (i.e. things we have power to change to an extent) than by genes.

Why might sleep duration affect biological aging?

Sleep is a maintenance window for the body. Adequate sleep supports immune regulation, hormone balance, and tissue repair processes across multiple organs, all of which can be disrupted when sleep is chronically short.
Long sleep may be a sign of poor(er) health rather than a cause. Needing noticeably more sleep than usual can be an early symptom of an underlying condition or of fatigue driven by something else.

As always, there are some limitations we need to consider

Sleep duration was self-reported through a single questionnaire item, which introduces recall error and doesn’t capture sleep quality, timing, or fragmentation.
The findings are observational: they show associations, not proof of cause and effect.
Participants were predominantly of European ancestry, which may limit how well the findings generalise to other populations.
The blood-based measurements were single snapshots, which can be influenced by short-term factors such as illness, medication, or diet at the time of testing.

Practical takeaways

As with previous newsletters where we’ve deep dived into recent research, the evidence here is observational and shouldn’t be taken as a rule. That said, below are a few sensible takeaways:

Treat 6 to 8 hours as the target range, with roughly 7 to 7.5 hours a reasonable aim for most people, rather than chasing the lowest amount you can function on or assuming more is always better.
Pay closer attention to persistent short sleep. Its associations in this study were broader and more consistent across the body.

Notice sustained changes in your own sleep need, rather than a single long night. If you find yourself needing meaningfully more sleep than usual for a prolonged period, it may be worth mentioning to your GP, since it could reflect something else worth checking on, rather than being something to correct through sleep alone.

For those who want to explore the data behind these findings in more depth, the research team behind this paper has made an interactive version available at labs,laboratory.com/sleepchart.

External content is for information only. If you have any health concerns connected to this article please contact your GPP