How much sleep is enough? For years, public-health advice has given a population-level answer: most adults need around seven to nine hours, and less than six hours counts as short sleep. But a new study from the National University of Singapore (NUS) suggests that fixed cut-offs miss something important — because “short” for you may not be short for someone else. The researchers propose a personalized way of defining insufficient sleep, based on your own long-term sleep pattern rather than a single number applied to everyone.
Published in the journal Sleep Medicine and announced on October 6, 2026, the study followed 462 working adults who wore Oura Rings and completed daily smartphone assessments for up to one year. Instead of asking only whether someone slept less than six hours, the researchers asked a different question: was this night substantially shorter than what this person usually achieves when they have greater opportunity to sleep? The answer, they argue, reveals sleep shortfall that fixed nightly cut-offs can miss — and that has practical implications for anyone tracking their sleep with a wearable.
This article is general health information, not medical advice. If you have persistent sleep problems, talk to a clinician — especially if you snore loudly, stop breathing during sleep, feel excessively sleepy during the day, or have symptoms that interfere with daily life.
The Study: 462 Adults, One Year of Wearable Data
The study comes from the Centre for Sleep and Cognition at the NUS Yong Loo Lin School of Medicine. Researchers followed 462 working adults using Oura Rings — wearable devices that track sleep duration and timing night after night — combined with daily smartphone assessments, for up to one year. That long observation window is the study’s key strength: most sleep research captures a few nights in a lab or a short survey, but this study captured hundreds of nights from the same people in real life, with real work schedules, weekends, holidays, and disruptions.
With that much data per person, the researchers could build a picture of each individual’s long-term sleep distribution — not just their average, but the range of nights they actually achieve, including the longer nights that happen when work and social constraints ease. That upper part of the distribution matters, because it provides a practical estimate of the sleep a person gets when they have greater opportunity to sleep — closer to what their body would choose if life did not get in the way.
What Is Personalized Short Sleep?
The researchers’ personalized short sleep (pSS) framework works like this: instead of labeling every night under six hours as “short sleep,” it compares each night with the individual’s own long-term pattern. The personalized threshold is set at one hour below the 75th percentile of that person’s sleep duration. In plain terms: look at all of someone’s nights, find the duration they reach or exceed on their better quarter of nights, and subtract one hour. A night shorter than that threshold counts as personally short.
The one-hour reduction was informed by experimental studies showing functional effects of sleep restriction — in other words, losing about an hour relative to your own well-rested pattern is enough to matter for how you function. The 75th percentile is used because it reflects nights when constraints were reduced, giving a realistic picture of a person’s sleep need rather than their sleep-deprived average.
Why does this matter? Because two people can sleep the same number of hours and be in completely different situations. A person who usually sleeps eight hours and gets six is two hours below their well-rested pattern — a real shortfall. A person who usually sleeps six and a half hours and gets six is close to their own pattern and may be fine. A fixed six-hour cut-off treats both nights identically. The personalized framework does not.
Fixed Cut-Offs vs Personalized Short Sleep
| Approach | How it works | Strengths | Limitations |
|---|---|---|---|
| Fixed cut-off (under 6 hours) | Any night below six hours counts as short sleep, for everyone | Simple, easy to apply to populations, useful for public-health guidance | Misses individual differences; can label healthy short sleepers as short and miss shortfall in naturally long sleepers |
| Population recommendation (7–9 hours) | Adults should aim for seven to nine hours per night | Good general target; backed by large-scale evidence | A target is not a diagnosis; individual needs vary within and beyond the range |
| Personalized short sleep (pSS) | Threshold = one hour below the 75th percentile of your own long-term sleep | Captures personal shortfall; uses months of real-life wearable data | Needs long-term tracking data; not a clinical diagnosis on its own |
The point is not that fixed cut-offs are wrong — they remain useful for population guidance and for flagging obviously short sleep. The point is that they are incomplete. A personalized measure adds a layer that fixed numbers cannot provide: whether this night was short for this person.
How to Find Your Own Sleep Pattern
You do not need a research lab to apply the study’s idea. If you wear a sleep tracker — a ring, watch, or phone-based tracker — you already have most of what you need. Here is a practical, low-effort approach:
Step 1: Collect at least a few weeks of data
The study used up to a year, but a few weeks of consistent tracking is enough to start seeing your pattern. Wear the tracker nightly, including weekends, so the data reflects both constrained work nights and freer nights. Consistency matters more than precision: the same device, worn the same way, gives comparable night-to-night data even if its absolute accuracy is imperfect.
Step 2: Find your own “good night” level
Look at your longer nights — roughly the top quarter of your recorded nights, the ones you get when you have time to sleep in or fewer morning obligations. That level approximates your well-rested pattern. Many tracker apps show sleep duration distributions or trends that make this easy to eyeball: your typical long night is a better personal reference than any population average.
Step 3: Watch for nights an hour or more below that level
Following the study’s logic, a night about an hour or more below your good-night level is a personally short night. One such night after a late event is normal life. A pattern of them — several per week, week after week — is the signal worth acting on, because that is what chronic shortfall looks like in your own data.
Step 4: Look at the pattern, not single nights
Sleep varies. Travel, illness, stress, and celebrations all disrupt individual nights, and that is normal. The useful signal is the trend over weeks: are your weeknights consistently an hour below your good-night level? Is your weekend catch-up sleep getting longer, suggesting a weekday deficit? Trends, not single nights, should drive changes.
What to Do If Your Pattern Shows a Shortfall
If your own data shows a persistent personal shortfall, the fixes are mostly unglamorous — which is why they work. Keep a consistent wake time, even on weekends, because a regular wake time anchors your sleep rhythm more reliably than a regular bedtime alone. Get bright light in the morning and dim the lights in the evening, giving your body clear day and night signals. Keep the bedroom cool, dark, and quiet, and reserve the bed for sleep rather than work or scrolling.
Cutting back on late caffeine and alcohol helps more people than any supplement: caffeine late in the day delays sleep onset, and alcohol fragments the second half of the night even when it seems to help you fall asleep. Screens deserve an honest look too — not because of blue light alone, but because engaging content keeps you awake past your intended bedtime. A wind-down routine that starts 30 to 60 minutes before bed, with the phone charging outside the bedroom if possible, is one of the highest-yield changes most people can make.
If shortfall persists despite good habits, or if you have symptoms like loud snoring, witnessed breathing pauses, morning headaches, or overwhelming daytime sleepiness, see a clinician or a sleep specialist. Wearable data is a useful conversation starter with a doctor, but it does not diagnose sleep disorders — and some, like sleep apnea, need proper testing and treatment.
Why Wearables Changed Sleep Research
This study would have been impractical a decade ago. Lab sleep studies are precise but capture only a night or two in an artificial environment. Surveys are cheap but unreliable — people misremember their sleep by large margins. Wearables sit in between: less precise than a lab, but capturing hundreds of nights in real life, which is exactly what a personalized framework needs.
The trade-off is worth understanding. Consumer wearables estimate sleep from movement and heart rate, and they can misclassify quiet wakefulness as sleep or miss brief awakenings. That means your tracker’s numbers are estimates, not medical measurements. But for the personalized approach, that limitation matters less than it seems: because the framework compares your nights with your own other nights, measured the same way, consistent bias cancels out. What matters is the pattern, and wearables are very good at patterns.
Personalized Sleep vs Other Sleep Trends
The personalized short sleep idea sits alongside other shifts in how we think about sleep. Sleep banking — sleeping extra before a period of expected shortfall — recognizes that sleep need is dynamic, not fixed. Chronotype research recognizes that morning people and evening people have genuinely different rhythms, so the same bedtime is not right for everyone. Sleep regularity research shows that consistent timing matters as much as total duration for health outcomes.
What the NUS study adds is a concrete, data-driven way to define shortfall for an individual. It does not replace population guidelines — seven to nine hours remains a sound general target — but it refines them, giving people with tracking data a personal benchmark to complement the public one. For the growing number of adults who already wear a sleep tracker, that is a practical upgrade to advice they can actually use.
Limitations of the Study
Honest caveats matter. The study’s participants were working adults in Singapore, and sleep patterns vary by culture, climate, work norms, and age — so the framework needs testing in other populations. Wearable estimates, as noted, are not lab measurements. And the personalized threshold itself — one hour below the 75th percentile — is a research definition, not a clinical diagnostic cut-off; it identifies shortfall for research and self-monitoring, not a disease.
None of that undermines the core insight. It simply means the framework should be used as a personal monitoring tool and a research direction, not as a medical test. Future studies with longer follow-up, clinical outcomes, and diverse populations will show how well personalized shortfall predicts health outcomes compared with fixed cut-offs — and whether interventions guided by personal thresholds work better than generic advice.
Frequently Asked Questions
What is personalized short sleep?
Personalized short sleep is a framework from NUS researchers that defines a short night relative to your own long-term sleep pattern, rather than a fixed cut-off. The threshold is one hour below the 75th percentile of your own sleep duration — in other words, a night substantially shorter than your better nights.
How is this different from sleeping less than six hours?
Six hours is a population cut-off applied to everyone. Personalized short sleep asks whether a night was short for you specifically. A six-hour night may be a real shortfall for someone who usually sleeps eight hours, but close to normal for someone who usually sleeps six and a half.
How did the NUS study measure sleep?
Researchers followed 462 working adults for up to one year using Oura Rings and daily smartphone assessments, capturing hundreds of nights per person in real life. The study was published in the journal Sleep Medicine.
Can I use my sleep tracker to find my personal threshold?
Yes, approximately. Collect a few weeks of consistent nightly data, identify your longer nights (roughly your top quarter), and treat nights about an hour or more below that level as personally short. Look at trends over weeks rather than single nights, and remember tracker data is an estimate, not a medical measurement.
When should I see a doctor about sleep?
See a clinician if sleep problems persist despite good sleep habits, or if you have loud snoring, breathing pauses during sleep, morning headaches, or excessive daytime sleepiness. This article is general information, not medical advice, and wearable data does not diagnose sleep disorders.
