What New Sleep-Timing Research Means for Your Bedtime

On August 5, 2026, a Flinders University-led team published an unusually long look at irregular sleep: about 20 months of under-mattress sensor data from 95,819 people, reported in the Journal of Sleep Research. People whose sleep length and timing swung widely in at least three of every four months had 46% higher odds of high blood pressure than people whose sleep was irregular in a quarter of months or fewer. For you, it means when you sleep is a second target alongside how long you sleep, not a replacement: the evidence does not show steady timing beats total hours for everyone, and adults are still advised to get at least seven.

Key takeaways

In 60,977 UK adults averaging about 63 years old, sleep regularity predicted death from any cause slightly better than sleep duration: about 30% lower risk for the most regular fifth versus 24% for the best duration group.
Regularity does not replace hours. Adults aged 18 to 60 are advised to sleep 7 or more hours a night, and about 1 in 3 US adults reported less in 2020.
In UK Biobank studies, the least regular sleepers had higher risk of type 2 diabetes and of heart attack, stroke or heart failure even when they met sleep-duration recommendations.
Weekend catch-up sleep eases sleep debt but did not prevent the metabolic effects of short weeknights in a lab trial.
In the 2024 mortality study, the most regular fifth fell asleep and woke up within roughly 1-hour windows on most days; the least regular fifth used roughly 3-hour windows.

What it means for you

Weekend sleep-ins ("social jet lag"): a 2026 UK Biobank analysis linked a 2-hour or larger weekday-to-weekend shift to about 30% higher risk of cardiovascular disease. See "Social jet lag and catch-up sleep".
Shift workers: a sleep block shared by work days and days off keeps part of the schedule steady. See "How much variation is fine".
Parents of newborns: the studies here did not look at new parents. A practical suggestion, not a research finding: protect total sleep and naps first; regularity can wait.
Teens: the AASM recommends 8 to 10 hours for ages 13 to 18. More weekday sleep leaves less to recover on weekends.
Travelers: CDC adjustment rates and short-trip advice are in the FAQ.
People with insomnia: a fixed schedule alone is not the recommended treatment. See the FAQ on insomnia.

What the studies actually found

Most of the evidence comes from observational studies that measured sleep with a device, then followed people for years. They show associations, not proof of cause, and most participants were middle-aged or older UK adults.

Study	Who	Main finding	What it says about hours
Windred et al., SLEEP, 2024	60,977 UK adults, sleep measured for 7 days, about 6 years of follow-up	The most regular fifth had 30% lower all-cause mortality than the least regular (hazard ratio 0.70, fully adjusted)	Duration also predicted death (best group 0.76), but less strongly than regularity
Chaput et al., Diabetes Care, 2024	73,630 UK adults aged 40–79, sleep measured for 7 days, 8 years	Irregular sleepers had 38% higher risk of type 2 diabetes than regular sleepers	Risk stayed 35% higher among irregular sleepers who slept 7 hours or more
Chaput et al., J Epidemiol Community Health, 2025	72,269 UK adults aged 40–79, sleep measured for 7 days, 8 years	Irregular sleepers had 26% higher risk of heart attack, stroke or heart failure	Enough sleep offset the risk only for moderately irregular sleepers
Nauha et al., BMC Cardiovascular Disorders, March 2026	3,231 Finnish adults aged 46, 10 years	Irregular bedtimes linked to about twice the risk of major cardiac events (95% CI 1.00–4.01, a borderline result)	The link appeared only in people sleeping less than the group median of 7 h 56 min
Jones et al., J Sleep Res, August 2026	95,819 users of an under-mattress sensor worldwide, about 20 months	Chronic irregular sleepers had 46% higher odds of high blood pressure and 40 more minutes awake at night	Irregularity counted swings in both duration and timing

Together, these studies support a narrower claim than "timing matters more than hours". Regularity is an independent risk marker: the stronger predictor in the mortality study, and not fully cancelled by seven-plus hours in the diabetes and heart studies. But the Finnish study found irregular timing mattered only among shorter sleepers, and the two measures travel together: in the 2024 mortality study, regularity rose with sleep duration up to about 7.8 hours.

Expert bodies agree. A 2023 National Sleep Foundation consensus panel concluded that consistent sleep onset and wake timing is important for health, safety and performance. The AASM and Sleep Research Society recommendation of at least 7 hours for adults, published in SLEEP in 2015, still stands; the AASM lists "daily regularity" as one part of healthy sleep, not a replacement for duration. Short sleep is common: a CDC analysis of 2020 survey data found 33.2% of US adults reported less than 7 hours, and 2022 state rates ranged from 30% in Vermont to 46% in Hawaii.

How the regularity score works

The Sleep Regularity Index (SRI) asks one question: if you pick any moment of the day, how likely is it that you were in the same state, asleep or awake, at that same clock time 24 hours earlier? The 2024 mortality paper describes a score of 100 as perfectly regular and 0 as random.

SRI = 200 × P(same) − 100

P(same) = share of all minutes in the recording where your
          sleep/wake state matches the state 24 hours later

If every minute matches, the SRI is 100; if only half match, as chance would give, it is 0. Because it compares minute by minute, the score drops when bedtimes or wake times move, and also with long night awakenings or daytime naps.

Illustrative example: same hours, different scores

These idealized weekly schedules have no night waking and are scored with the formula above. Real-world scores run lower because devices pick up awakenings and naps.

Weekly pattern	Average sleep	SRI
11 p.m. to 7 a.m. every night	8.0 h	100
Same, but midnight to 8 a.m. on Friday and Saturday nights	8.0 h	95.2
Same, but 1 a.m. to 9 a.m. on Friday and Saturday nights	8.0 h	90.5
Bedtime alternates 10:30 p.m. and 11:30 p.m. every night, always 8 hours	8.0 h	83.3
11:30 p.m. to 5:30 a.m. on weeknights, 2 a.m. to noon on Friday and Saturday nights	7.1 h	78.6
Midnight to 5 a.m. every night	5.0 h	100

In this example, shifting bedtime and wake time by 1 hour on the two weekend nights costs about 5 points, and a 2-hour shift about 10, with the same 8 hours of sleep. Yet the last row scores a perfect 100 while sleeping 2 hours less than the guideline every night: the SRI measures only consistency, so neither it nor hours alone describes healthy sleep. For scale, the UK Biobank group had a median SRI of 81, and the diabetes and heart studies classed scores above 87.3 as regular and below 71.6 as irregular.

Social jet lag and catch-up sleep

Social jet lag is the gap between the body clock and the schedule that work or school imposes, a term from chronobiologist Till Roenneberg's group. Their 2012 study in Current Biology linked it to higher body mass index beyond the effect of sleep duration. It is usually measured as the shift in mid-sleep, the clock time halfway through the night, between workdays and free days. In the third row of the example table, mid-sleep moves from 3 a.m. on weeknights to 5 a.m. on the weekend, so social jet lag is 2 hours.

That 2-hour line matters. In a 2026 Journal of Internal Medicine analysis of 51,562 UK Biobank participants, people with 2 hours or more of device-measured social jet lag had 30% higher risk of cardiovascular disease, and severe social jet lag stayed linked to risk even among people with normal sleep duration. In that analysis, sleep duration itself was not independently linked to risk.

What sleeping in can and cannot fix

The evidence points in two directions, and both can be true:

It does not reverse the metabolic effects of short weeknights. In a University of Colorado lab trial (Depner et al., 2019), young adults given 5-hour weeknights and a free weekend slept about 1.1 hours extra in total, but once short nights resumed, their body clocks had shifted later and insulin sensitivity fell by about 9% to 27%.
It may be better than not recovering at all. The National Sleep Foundation says that when workday sleep is short, an extra 1 to 2 hours a day on days off can help most people recover sleep debt. A study presented at the European Society of Cardiology Congress 2024 in London found that, among about 91,000 UK Biobank participants, those with the most catch-up sleep were 19% less likely to develop heart disease than those with the least. It is not a full peer-reviewed paper, and the British Heart Foundation cautioned that a lie-in is no substitute for regular good sleep.

The best fix is fewer short weeknights, so less recovery is needed.

How much variation is fine

No guideline sets an official limit in minutes. The closest evidence-based markers come from the studies themselves:

In the 2024 mortality study, the most regular fifth fell asleep and woke within roughly 1-hour windows on most days, and the authors suggest aiming for 1-hour windows as a feasible strategy. The least regular fifth used windows of about 3 hours.
The 2026 blood pressure study classed a month as irregular when the standard deviation of sleep length was 90 minutes or more and the standard deviation of mid-sleep was 60 minutes or more, thresholds earlier linked to poorer heart health.

A practical reading of those numbers, not an official rule: pick a target window that allows at least 7 hours in bed asleep, then try to start and end sleep within about 30 minutes either side of it most nights, and keep the weekend shift in mid-sleep under 2 hours, the level linked to higher risk in the 2026 social jet lag study. Among people sleeping under about 8 hours, the Finnish study linked bedtime and mid-sleep variability, but not wake-time variability, to cardiac events, so a steady wake time alone may not be enough.

Shift work

NIOSH training for nurses warns that switching between night work and night sleep once or more a week can cause "a perpetual state of jet lag". It describes a compromise schedule: sleep right after night shifts, and on days off stay up until about 3 to 4 a.m. and sleep until noon or 1 p.m., so that a block such as 8 a.m. to noon is sleep time on both work and rest days.

Frequently asked questions

Does a steady bedtime or a steady wake time matter more?

Studies rarely separate them, because moving one usually moves the other. The 2026 Finnish study is one exception, and it was observational with borderline statistics, so it supports watching both ends of the night rather than picking one.

Is it harmful to sleep more than nine hours on weekends?

Not necessarily. The AASM and Sleep Research Society set no upper limit and said more than nine hours may be appropriate for young adults, people recovering from sleep debt and people who are ill. Timing still counts: two extra hours taken entirely in the morning move mid-sleep 1 hour later, while one hour earlier to bed plus one hour later up leaves it unchanged.

How quickly does the body clock adjust after a flight?

The CDC Yellow Book gives average adaptation rates of about 1.5 hours a day after flying west and 1 hour a day after flying east. It suggests shifting sleep 1 hour a day in the travel direction for 2 to 3 days before departure. For trips shorter than about 2 days, it advises staying on home-time sleep and meals instead of adapting.

Does irregular sleep cause health problems, or only signal them?

Probably some of both. The authors of the 2024 mortality study say regularity may be both a cause and a marker of risk, and observational studies cannot fully separate the two. Lab trials such as the Colorado study show that shifting and shortening sleep changes metabolism within days, which supports a causal role for part of the link, though that trial shortened and shifted sleep together, so it cannot isolate timing.

Will a regular sleep schedule cure insomnia?

Usually not on its own. The AASM's 2021 guideline strongly recommends multicomponent cognitive behavioral therapy for insomnia (CBT-I) for chronic insomnia in adults and suggests, as a conditional recommendation, that clinicians not use sleep hygiene tips alone as treatment. CBT-I includes structured schedule techniques such as stimulus control and sleep restriction.

A two-week plan for steadier sleep

These steps are practical suggestions built from the studies above, not a clinically tested program.

For one week, log when you fall asleep and wake up, weekends included, and note your weekday and weekend mid-sleep.
Pick a target window that fits your fixed commitments and gives at least 7 hours of sleep (8 to 10 for teens).
If your weekend mid-sleep is 2 hours or more later than on weekdays, move weekend bedtime and rising time 30 to 60 minutes closer to your weekday times in week two.
If you are short on weeknight sleep, move bedtime earlier rather than relying on weekend lie-ins. If you do sleep in, keep it to 1 to 2 extra hours.
After two weeks, compare the width of your windows with the 1-hour benchmark from the mortality study.

This article is general education, not medical advice. The CDC advises talking to a healthcare provider if you regularly have problems sleeping, and lists feeling sleepy or tired even after enough sleep as a sign of poor sleep quality. The CDC names insomnia, sleep apnea, restless legs syndrome and narcolepsy as common sleep disorders.