Is a Regular Bedtime Better Than Eight Hours?

So there is this video — late August, a Straits Times production, titled simply Sleep earlier — in which a National University of Singapore professor named Dean Ho, a man who has been fasting and dieting and logging his 90 minutes of daily exercise while algorithms keep a running tab on his ‘biological age,’ explains that he improved his sleep, and the whole thing gets compressed, as these things do, into an invitation so tidy it could fit on a clock face: move bedtime forward, move death farther away (it drew nearly 17,000 views in four days, according to an Apify capture, which tells you something about how hungry we are for instructions that simple).

Here is the honest question that compression raises without quite answering: is a regular bedtime actually better than eight hours — and is earlier better at all? It matters because modern sleep science increasingly treats sleep as a pattern, not merely a nightly total, and because advice this simple gets acted on by millions of people who cannot all be one self-experimenting professor. The most defensible current answer: sleep regularity predicts mortality more strongly than duration in large observational cohorts, but it has not been shown to extend anyone’s life. The evidence supports addition, not substitution.

What ‘regular’ actually measured

The headline study followed 60,977 UK Biobank participants whose wrist accelerometers recorded more than 10 million hours across seven days. Researchers calculated a Sleep Regularity Index: the probability that a person was in the same state—awake or asleep—at any two moments 24 hours apart. A high score therefore reflects much more than hitting one bedtime. It captures relatively stable sleep onset, waking and duration across the week.

During an average 6.3 years of follow-up, 1,859 participants died. Compared with the least regular fifth, each of the other four fifths had a 20% to 48% lower adjusted risk of death from any cause; regularity predicted mortality more strongly than average duration in comparable models. A separate, diverse US cohort found that a cluster averaging seven hours with regular timing had a 39% lower mortality hazard than a cluster averaging 5.8 hours with roughly twice the irregularity.

Those numbers are associations, not minutes added to a life. The British participants averaged about 63 years old and volunteered for a biobank; their single measured week may not represent decades of sleep. Irregularity can be a consequence of illness, pain, caregiving, unstable work, depression, medication or social disadvantage. Statistical adjustment helps, but it cannot make a randomized longevity trial. Even the US comparison combined two advantages—regularity and sufficient duration—so it cannot tell us which one produced the lower hazard.

Eight hours did not become obsolete

‘Regularity was a stronger predictor’ does not mean duration is unimportant, or that six tidy hours beat eight variable ones. The American Academy of Sleep Medicine and Sleep Research Society recommend that adults sleep seven or more hours per night on a regular basis; individual need varies, and some people need more. Eight hours is a familiar center of a range, not a biological pass-fail line.

A large 2026 Nature analysis made that nuance newly visible. Using self-reported duration and 23 biological-aging clocks across brain and body systems, investigators found U-shaped associations: the lowest age gaps occurred between about 6.4 and 7.8 hours, depending on organ and sex, while less than six and more than eight hours were associated with older-looking clocks and more disease. Yet the clocks are surrogate measures, and long sleep can signal underlying illness rather than cause it. The study did not prove that trimming an eight-and-a-half-hour sleeper to seven reverses aging.

Intervention evidence also refuses the either-or framing. In 2026, 30 healthy habitual short sleepers were asked to add two hours of time in bed for four weeks. They slept about 47 minutes longer, moved sleep onset about 52 minutes earlier and reported better alertness, while regularity barely changed. Increasing opportunity helped duration without fixing consistency. Conversely, a small older experiment that kept undergraduates above 7.5 hours found that adding a regular schedule reduced daytime sleepiness. These studies measured sleep and function, not survival, but together they suggest distinct levers.

Earlier is not the same as aligned

The civil clock does not contain a universal longevity hour. In two observational cohorts, mortality risk traced a U-shaped curve around sleep timing, with approximately 11 p.m. bedtime and 7 a.m. waking at the statistical low point. Late bedtimes after midnight were associated with higher mortality. But in the 21-country PURE study, both going to bed at 10 p.m. or earlier and after midnight carried modestly higher adjusted risks than sleeping between 10 and midnight. Early sleep can reflect frailty, illness or an early chronotype just as late sleep can reflect work and social constraints.

What biology notices is alignment: sleep scheduled in a repeatable relationship with the internal circadian clock, light exposure, meals and activity. Morning types and evening types differ. Forcing a natural night owl into bed at 9 p.m. may simply create two wakeful hours and anxiety about failing at sleep. A schedule can be regular yet too short, and eight hours can be mistimed for a night worker. ‘Sleep earlier’ is therefore a poor universal prescription; ‘make enough sleep more predictable and better aligned’ is closer to the evidence.

The shift-work exception is not a personal failure

Regularity advice can sound moralizing to nurses, factory workers, emergency crews and parents whose schedules are not theirs to choose. A meta-analysis of 16 cohorts found that night work, compared with day work or never working nights, was associated with 6% higher all-cause mortality and 15% higher cardiovascular mortality, with considerable variation between studies. That does not prove circadian disruption alone caused the difference; occupation, stress, income and baseline health travel with the schedule.

Small experiments point toward harm reduction rather than perfection. In 18 older adults undergoing simulated night shifts, scheduled afternoon-evening sleep plus bright light later in the shift preserved six to seven hours of sleep and improved alertness over successive nights; controls slept three to five hours. It did not test long-term disease or lifespan. For real shift workers, the practical target is a protected sleep opportunity, a repeatable plan across runs of similar shifts, strategic workplace light and darkness during the trip home and sleep period—not pretending a 10 p.m. bedtime is available.

A protocol with no immortality promise

Start with wake time because it is often easier to anchor than the exact moment sleep begins. Keep it within roughly the same one-hour neighborhood when life allows, including weekends, then count backward to reserve at least seven hours of actual sleep opportunity. If the current schedule is short, expand the window gradually rather than sacrificing duration for a perfect streak. Seek outdoor light after waking; lower bright light and stimulating activity near the intended bedtime. Track weekly patterns, not a wearable’s verdict on one bad night.

Do not chase regularity through prolonged wakefulness in bed. Persistent insomnia is treated most effectively with cognitive behavioral therapy for insomnia, and loud snoring, gasping, morning headaches or disabling sleepiness deserve evaluation for sleep apnea or another disorder. People with bipolar disorder, eye disease, medications affected by light, or rotating shifts should get individualized guidance before using bright-light or melatonin protocols.

The August video’s best lesson is not that everyone should copy one professor’s 9 p.m. bedtime. It is that behavior can be measured, tested and revised. The population evidence makes regular sleep a promising health signal; it does not yet make a regular bedtime a life-extension treatment. Protect enough sleep, make the pattern steadier where you can, and treat the barriers that a clock alarm cannot solve. Educational, not medical advice.