Is Sleeping In on Weekends Bad for You? What Studies Say

Is Sleeping In on Weekends Actually Bad for You?

Picture two schedules. On work or school days your alarm goes off at 6:30. On days off there's no alarm, and you wake up somewhere around 9:30 or 10. Is that gap costing you something, or is your body just collecting what it was owed? Three ideas hide inside that question: catch-up sleep, social jetlag and sleep regularity, and the studies on them do not fully agree.

Confident answers are easy to find: sleeping in repays your sleep debt, wrecks your body clock, or is linked to about a fifth lower heart-disease risk (a headline from one observational conference abstract, not a peer-reviewed article). Each of them can point to some research, which is exactly why none of them works as a flat rule.

"Weekend" in the title is search shorthand, and this article uses it mostly when a study or a search phrase does. Plenty of people get days off midweek, work rotating shifts, or follow school calendars that look nothing like a Monday-to-Friday grid. The ideas below apply to any stretch of days when you can sleep on your own clock, though most of the studies measured calendar weekends as a stand-in for free days.

A note on health information: This is general information, not medical advice. The studies described here report patterns in groups of people. They can't show what will happen to you, and most can't show cause and effect. If you have ongoing sleep problems, heavy daytime sleepiness, or symptoms that worry you, talk to a healthcare professional.

The 30-Second Answer

Whether extra sleep on days off protects health is unsettled. Large studies disagree, and a lab study suggests it doesn't fully undo severe sleep loss. What the evidence more consistently supports is that big, recurring swings in sleep timing are associated with poorer outcomes, and that sleep duration and sleep regularity each carry information. Treat any figure of 1 to 2 extra hours as a rough guide drawn from studies and one guidance panel, not as a limit.

• Catch-up sleep: large UK analyses using wrist-accelerometer data disagree: one was a peer-reviewed study and the other a conference abstract, and they reached different conclusions about whether weekend catch-up sleep was linked to lower heart-disease risk
• Social jetlag: linked to some measures, such as body mass index and blood sugar, but the evidence is small and inconsistent
• Sleep regularity: irregular sleep is associated with higher mortality and cardiovascular risk in observational data
• The lab evidence: in a 36-person randomized lab study of severe sleep restriction, the 14 people given two recovery nights still showed metabolic changes once 5-hour nights resumed
• Worth a conversation with a professional: persistent very short sleep, loud snoring or breathing pauses, heavy daytime sleepiness, or needing unusually long sleep

Three Things People Mean by "Weekend Sleep"

Researchers measure three different things here, and headlines tend to treat them as one.

Weekend catch-up sleep

This is the extra sleep people get on free days. Studies usually sort people by the difference in average sleep duration into bands: none, under an hour, one to two hours, two hours or more. Most of the catch-up studies below used calendar weekends and weekdays as their measure, so that is a stand-in for the broader idea of free days versus work days.

Social jetlag

This is a shift in timing. Researchers usually describe it using midsleep, the halfway point between falling asleep and waking up. Social jetlag is midsleep on free days minus midsleep on work or school days. The term comes from a 2006 paper by Wittmann and colleagues on how social schedules pull against people's natural sleep timing.

Sleep regularity

This is how consistent your sleep and wake times are from one day to the next, across the whole week. The best-known measure is the Sleep Regularity Index (SRI), introduced in 2017, which scores how often you're in the same state, asleep or awake, at the same time on consecutive days. It isn't the only measure. Others look at the spread of bedtimes, wake times, or nightly sleep duration.

Here's how the three come apart. Alex and Jordan both sleep 11 p.m. to 7 a.m. on work days, so midsleep is 3 a.m. On free days Alex sleeps 11 p.m. to 9 a.m. and Jordan sleeps 1 a.m. to 9 a.m. Alex gets two hours of catch-up sleep, and midsleep moves one hour later. Jordan gets the same eight hours every night, so no catch-up sleep, yet midsleep moves two hours later, to 5 a.m. Alex has more catch-up sleep but less social jetlag than Jordan. Neither example says which schedule is healthier.

Construct	What it measures	Typical calculation or definition	What it does not measure
Weekend catch-up sleep	Extra sleep duration on free days compared with work or school days	Average free-day sleep duration minus average work-day sleep duration, often sorted into bands	When you sleep, how consistent your schedule is, or whether you were short on sleep
Social jetlag	How far your sleep timing shifts between work or school days and free days	Midsleep on free days (MSF) minus midsleep on work days (MSW)	How long you sleep, or how much your timing varies within each kind of day
Sleep regularity	Day-to-day consistency of sleep and wake times	The Sleep Regularity Index, plus other measures such as the spread of bedtimes	How much you sleep, or whether your timing suits your body

These are researchers' measures. A score you work out yourself, or see in an app, isn't a validated clinical reading.

What the Research Says About Catch-Up Sleep

The largest studies come from the UK Biobank, a research resource with health records for around half a million UK adults, where a subset wore wrist accelerometers for a week.

Two big UK Biobank analyses, two different answers

The first is Chaput and colleagues' 2024 paper in Sleep, which was peer-reviewed and used device-measured sleep. It followed 73,513 UK adults (mortality) and 70,518 (cardiovascular disease, or CVD) for about eight years. Catch-up sleep was defined as longer average sleep on weekends than weekdays. In adjusted models, compared with people who had no catch-up sleep, those with 2 hours or more had a hazard ratio of 1.17 for death (95% confidence interval 0.97 to 1.41) and 1.05 for new CVD (0.94 to 1.18). Both intervals include 1, so neither result is statistically significant. The authors concluded that weekend catch-up sleep "was not associated with mortality or CVD incidence," and the finding held when they looked only at people sleeping under six hours on weekdays. This is observational data, and such wide intervals can't rule out modest effects either way.

The second came from a group at Fuwai Hospital in Beijing. It was presented at the 2024 European Society of Cardiology congress and printed as a conference abstract in the European Heart Journal's supplement. It used accelerometer data from the same resource, on 90,903 participants, and split weekend compensation into quartiles. The abstract reports a hazard ratio of 0.809 (95% CI 0.766 to 0.854) for heart disease in the top quartile, a combined outcome that covers ischemic heart disease, heart failure, atrial fibrillation and stroke, over a median follow-up of up to 13.8 years. Media coverage relied on a press release (reprinted by ScienceDaily), which described the comparison as top quartile against bottom.

This needs a clear label. It is a conference abstract published in the journal's supplement, not a peer-reviewed full research article, and it is observational. It doesn't list the variables adjusted for, and as of late September 2026 we couldn't find a full peer-reviewed version. It shows that some analyses of this resource find an association, but it isn't settled evidence.

Other studies, other outcomes

• Coronary artery calcium. In the Brazilian ELSA-Brasil cohort (average age 49), more than 90 minutes of catch-up sleep went with lower odds of new artery calcium over about five years than 90 minutes or less (odds ratio 0.62, 95% CI 0.52 to 0.74), mostly among people averaging 6.55 hours of sleep or less.
• Brain health. Two more UK Biobank accelerometer analyses disagreed on dementia. In one analysis, only 1 to 1.5 hours of catch-up sleep was associated with lower dementia risk among adults over 50 (HR 0.64, 0.49 to 0.86, against 0.5 hours or less). The other analysis found no link with dementia, stroke or Parkinson's, though short and medium amounts went with lower depression risk.
• Day-to-day rebound. A 2026 analysis in Nature Communications looked at daily patterns in UK Biobank rather than weekdays versus weekends. In this observational analysis, short sleep with no rebound was linked to higher mortality than regular sleep, while short sleep followed by a rebound wasn't significantly different. A US survey sample gave similar results.
• Earlier cohorts. In the Sleep Heart Health Study (3,128 adults aged 40 to 64), among people with normal lab-measured sleep, short catch-up sleep went with lower mortality, while under stricter cutoffs, short catch-up plus very short measured sleep went with higher mortality. In a cohort of about 44,000 adults followed 13 years, people under 65 who slept little on weekdays and long on weekends had mortality in line with steady 6-to-7-hour sleepers (the study's reference group, not a recommendation).
• Blood pressure, 2026 journal article. A US cross-sectional survey analysis of 16,268 adults aged 20 and older, published in International Journal of Hypertension in August 2026, found that catch-up sleep was not robustly associated with hypertension overall. Because it was cross-sectional, it cannot show which came first. The only significant category was 0 to 1 hour, which had lower odds of hypertension than none or less in the fully adjusted model (odds ratio 0.814, 95% CI 0.666 to 0.995). The 1 to 2, 2 to 3 and 3 hours or more categories were not significant, and the authors saw no dose-response relationship. They said the isolated result needs confirmation and shouldn't be read as evidence for a preventive strategy.
• Blood pressure, 2026 conference abstract. An abstract presented at ESC Congress 2026 analyzed 40,727 Korean adults from a national health survey. It reported a small overall association with lower odds of hypertension (adjusted odds ratio 0.94, 95% CI 0.88 to 0.997), with the lowest odds at about 1 to 2 hours among physically inactive adults. The accessible abstract text does not specify the odds-ratio comparison group. Its own interaction test between catch-up sleep and physical activity was not significant (p = 0.75). Because this was an observational analysis of hypertension prevalence, it cannot establish which came first. It is preliminary, and not a peer-reviewed article.

Findings vary by outcome, effects aren't steady as catch-up amounts rise, short sleepers show stronger links in some analyses and none in others, and no single amount emerges as a universal health threshold.

Why the disagreement isn't settled

The two main analyses differ in several ways. Chaput sorted people into bands of weekend-minus-weekday sleep, with "none" as the reference, while the Fuwai analysis used quartiles of weekend compensation (its press release shows the bottom quartile included people who slept less on weekends). The outcomes differ (mortality and CVD incidence versus a combined heart-disease outcome), so does follow-up (about eight years versus up to 13.8), and adjustment can't be compared because the abstract doesn't list its variables. According to a commentary in Sleep, Chaput and colleagues adjusted for factors including shift-work history and chronotype, meaning whether someone is naturally an early or late sleeper.

Any of those could matter, and none has been shown to explain the gap. The same commentary floated the idea that catch-up sleep's benefits might be offset by the irregularity it creates, but its authors noted that regularity wasn't measured in the Chaput study, so the idea remains untested. The summary is that the cardiovascular evidence is mixed, and the reason for the disagreement is not settled.

Study	Population	Outcome	Result	What it can and can't tell us
Chaput 2024 (Sleep)	73,513 UK adults	Death and CVD, about 8 years	2 hours or more vs none: death HR 1.17 (0.97 to 1.41), CVD HR 1.05 (0.94 to 1.18)	Observational. Neither result significant; wide intervals
Fuwai abstract 2024 (conference)	90,903 UK adults	Heart disease, up to 13.8 years	Top quartile: HR 0.809 (0.766 to 0.854) as printed; comparison group per press release	Conference abstract, not a full peer-reviewed article. Methods and adjustment unclear
ELSA-Brasil 2025	1,832 Brazilian adults (actigraphy)	New coronary calcium; about 5 years	More than 90 minutes vs 90 or less: OR 0.62 (0.52 to 0.74)	Early artery changes, not events. Concentrated in short sleepers
Chen 2025 and Zheng 2025	83,776 and 91,665 UK adults	Dementia; depression, anxiety	Dementia lower only at 1 to 1.5 hours (HR 0.64) in one; no link in the other	Same resource, different answers
Li 2026 (Nature Communications)	85,618 UK adults; 4,586 US survey adults	Death, about 8 years	Vs regular sleep: short sleep without rebound, higher risk; with rebound, not significant	One week of data. Daily patterns, not weekday vs weekend
Windred 2024 (Sleep)	60,977 UK adults	Death; up to 7.8 years	Top vs bottom regularity quintile: HR 0.70 (0.59 to 0.83) fully adjusted	Observational. Irregularity may be a cause, a marker of illness, or both
Chaput 2025 (JECH)	72,269 UK adults	Heart attack, heart failure, stroke; about 8 years	Irregular vs regular (SRI): HR 1.26 (1.16 to 1.37)	Can't show that improving regularity lowers risk
Bouman 2023 and Lin 2024	68 studies; 3 for one Lin result	BMI, waist, HbA1c; metabolic syndrome	Small differences in some measures; no link with existing metabolic syndrome or diabetes in the larger review	Social jetlag as defined in each study. Mostly cross-sectional; can't show cause
Depner 2019 (Current Biology)	36 healthy adults, average age 25; randomized lab study (14 in the recovery group)	Insulin sensitivity, weight, eating	Insulin sensitivity fell about 9% to 27% once 5-hour nights resumed	Severe restriction, small sample. Not ordinary sleeping in

UK Biobank rows use one week of wrist accelerometer data. ELSA-Brasil used one week of wrist actigraphy.

What the Research Says About Sleep Regularity

Windred 2024: Sleep Regularity and Mortality

Windred and colleagues analyzed 60,977 UK Biobank adults (average age about 63) who wore an accelerometer for a week and were followed for up to 7.8 years, during which 1,859 died. They split people into five groups by Sleep Regularity Index score. In this observational cohort, people in the highest regularity quintile had a mortality rate about 30% lower than those in the lowest after full adjustment (hazard ratio 0.70, 95% CI 0.59 to 0.83). The fully adjusted model included the 46,721 participants with complete covariate data. With minimal adjustment for age, sex and ethnicity only, the figure was 0.52 (0.45 to 0.60). The authors noted the truth likely sits between the two, because the fully adjusted model includes factors such as smoking that may partly lie on the causal path.

Regularity and duration together

When both went into one fully adjusted model, regularity stayed associated with lower mortality (top versus bottom quintile, HR 0.74, 0.62 to 0.89), and duration also stayed associated in two of its longer-sleep bands (HR 0.84, 0.71 to 0.99 for each, against the shortest), though not in the others. In Windred's observational UK Biobank cohort, model comparisons found sleep regularity to be a stronger predictor of mortality than sleep duration. Regularity-based models fit better than equivalent duration-based ones, and adding duration to a regularity model did not significantly improve its fit (p = 0.14 to 0.20). Those are statements about model fit in one observational cohort, separate from the band-by-band duration results above. They don't show that regularity causes better outcomes, or that it matters more than duration for any one person.

Regular sleepers also tend to get enough sleep: in Chaput's 2025 analysis (below), 60.9% of regular sleepers met sleep-duration recommendations, against 47.8% of irregular sleepers. Because regularity and duration overlap, their separate contributions in Windred's analysis are hard to pull apart.

Chaput 2025 and cardiovascular events

In Chaput and colleagues' 2025 analysis, an observational UK Biobank study of 72,269 adults aged 40 to 79, irregular sleepers had an adjusted hazard ratio of 1.26 (95% CI 1.16 to 1.37) for major cardiovascular events (heart attack, heart failure or stroke) compared with regular sleepers, over about eight years. When the researchers combined regularity with sleep duration, the higher risk was no longer evident among moderately irregular sleepers who met the age-specific duration recommendation, but remained in the most irregular group.

What these studies can't tell you

All of this is observational. Sleep was tracked for one week. The Windred authors describe their group as older and mostly homogeneous in ethnicity, and say the findings should be replicated in other cohorts, including across cultures. Only about 5.5% of the people invited to join UK Biobank took part, according to the Chaput 2025 paper. Irregular sleep may be a cause, an early sign of illness, or both. The Chaput team excluded early cardiovascular events to reduce that last problem, which helps but doesn't remove it.

A 2025 systematic review of 59 studies found moderate-certainty, consistent links between irregular timing and outcomes such as depression symptoms, blood pressure and cardiovascular events, but couldn't pool results because the metrics differed, and it called for randomized trials. The nearest thing we found is a small before-and-after pilot: 11 adults with high blood pressure kept a fixed bedtime for two weeks and saw modest drops in blood pressure, which isn't enough to suggest it as a treatment.

Social Jetlag: What It Measures and What It Doesn't

Take Jordan from the example above: midsleep is 3 a.m. on work days and 5 a.m. on free days, so social jetlag (MSF minus MSW) is two hours. That describes a schedule. It isn't a health score.

In a 2019 self-critical review, the researcher who coined the term, together with colleagues, pointed out that social jetlag is tangled up with sleep debt, since sleeping longer on free days pushes free-day midsleep later, and that the original idea "may be too simplistic." Travel jet lag is a different thing, since it involves crossing time zones, and our jet lag guide covers that version.

A 2023 review of 68 studies found that, among high-quality studies, people with social jetlag had slightly higher body mass index (0.49 points), waist circumference (1.11 cm) and HbA1c (0.42 percentage points) than people without, with high variation between studies for some of those measures. It found no significant association with existing metabolic syndrome or type 2 diabetes, and the authors called the evidence low-level and mostly cross-sectional. A smaller 2024 pooled analysis combined just three studies for metabolic syndrome and reported an odds ratio of 1.52 (95% CI 1.12 to 2.06) for social jetlag of 2 hours or more compared with under 1 hour. That's one small analysis set against a larger, less consistent picture.

Social jetlag has been linked to some metabolic measures in observational studies, but the evidence is small and inconsistent and does not establish that social jetlag causes disease. The "2 hours" line comes from how researchers grouped people. Nothing in these studies establishes it as the point where health effects begin.

Can Sleeping In Make Up for Lost Sleep?

The best experimental evidence comes from Depner and colleagues in 2019. They randomized 36 healthy adults (18 women, 18 men, average age about 25) in a lab to one of three groups: 9-hour sleep opportunities (8 people), 5 hours for nine nights (14), or 5 hours for five nights, then two nights of sleeping as much as they liked, then two more restricted nights (14).

Across the two recovery nights, the recovery group slept about 1.1 hours more in total than at baseline. When short nights resumed, their insulin sensitivity dropped by about 9% to 27% from baseline (whole-body, liver and muscle), late-evening eating and weight went up, and body-clock timing shifted later.

In a small lab study using severe 5-hour restriction, two recovery nights did not prevent metabolic changes once short nights resumed. The participants were young and healthy, the restriction was severe by design, and the restricted stretch lasted about nine nights. It didn't test an ordinary one-hour sleep-in, ordinary life on days off, or long-term outcomes. The authors also noted how little sleep people recovered, so it's unclear what more recovery would have done.

So How Much Is Reasonable?

Figure	Where it comes from	What it means	What it does not mean
1 to 2 extra hours on days off	National Sleep Foundation panel guidance (2023), for people who don't sleep enough on work or school days; wording from its news release	A panel's rough advice	A tested or medically safe limit
About 1 hour	Windred and colleagues: the most regular quintile went to sleep and woke within roughly 1-hour windows on most days	A description of how the most regular sleepers behaved	A cutoff for healthy regularity
None, under 1, 1 to 2, 2 or more hours	Bands Chaput and colleagues used to group catch-up sleep	An analysis choice	Thresholds for safe or unsafe sleeping in
More than 90 minutes	The split point in the ELSA-Brasil analysis	A study-specific comparison	A target
Under 1 vs 2 or more hours	Groups compared in pooled social-jetlag analyses	A comparison of groups	The point where social jetlag becomes harmful

These are study categories or rough guidance figures, not safe or unsafe limits.

In 2023 the US-based National Sleep Foundation's expert panel, using a modified Delphi process across 63 publications, agreed that regular timing matters for health and that catch-up sleep may be beneficial when weekday sleep is short. The foundation's news release put it as "sleeping an additional 1-2 hours a day" on non-work days for most people who are short on sleep. We couldn't access the panel's full text, so that wording comes from the release. It's guidance from experts weighing mixed evidence, not a trial result. The 2024 Chaput paper noted that its findings "do not align with" recommendations of this kind.

Two 2026 blood-pressure analyses show why a single amount is hard to defend. The US survey's only significant band was 0 to 1 hour, with no dose-response. The Korean conference abstract placed its lowest odds around 1 to 2 hours in its results but at approximately 2 hours in its conclusion, and described that curve for physically inactive adults only. Neither is a recommended amount.

Instead of an amount, three questions are more useful:

1. Are you getting enough sleep on work or school days? If not, the extra sleep on days off is telling you something about the other days.
2. Is your free-day shift large and recurring? Big, repeated swings are the pattern most consistently linked with poorer outcomes in observational data.
3. Any red flags? Loud snoring, breathing pauses, heavy daytime sleepiness or needing unusually long sleep are worth raising with a professional, whatever the numbers say.

Why Sleep-App "Regularity" Scores May Be Hard to Compare

The SRI's 0-to-100 score (random to perfectly repeating) came from that 2017 study of 61 undergraduate students, and more than one calculator has since been published to work it out from sleep and wake data.

A 2026 paper in Sleep compared two widely used open-source ones, called sleepreg and GGIR, on the same accelerometer data from more than 70,000 adults. The scores differed markedly. The choice of calculator alone meaningfully changed results in models of mortality, type 2 diabetes and atrial fibrillation.

This is a measurement problem, and it doesn't mean wearables or sleep apps are wrong. A consumer app's "sleep consistency" score is that company's own calculation and isn't automatically the SRI. Regularity scores are only comparable when you know they were calculated with the same method. Heart rate variability has the same trap, with numbers from different devices that don't line up. Comparing your score with your own past scores in the same app is a safer use than comparing across apps or against a study.

What to Do With This Information

• Work-day sleep is the part you may be able to change. If you're short on sleep because of your schedule rather than trouble sleeping, that's the direct route, since these studies don't show that catch-up sleep fully makes up for a run of short nights. For short sleepers the findings are mixed, and the research offers no catch-up amount to aim for.
• Don't cut your sleep to improve consistency. Setting an early alarm after a short night to protect a steady clock isn't what these studies tested, and none of them recommends shortening sleep.
• Narrow big recurring swings where it's practical. You don't need clock-perfect timing. Many guides suggest holding a steady wake time, and our guide to everyday sleep habits covers the broader basics.
• Night owls: people with later natural sleep timing tend to show bigger gaps between work and free days, since early schedules cut into their sleep on work days. That's context, not a recommendation.
• Shift workers: most of these studies weren't designed around shift work, and several only adjusted for it statistically, so don't apply their findings to your schedule automatically. Individual advice from a clinician or, where available, an occupational health service fits better.

If you have persistent very short sleep, or any of the red flags listed earlier, a healthcare professional can help sort out why. An article can't do that.

Bottom Line

None of these studies tests a single late morning, so none of them can say whether yours matters. The more useful thing to look at is the pattern behind it: how much you sleep on work or school days, and how large and recurring your timing swings are. Most of this research is observational, and none of it proves cause.

---

Related reads:

• How to Sleep Better at Night Naturally: A 30-Day Reset
• How to Beat Jet Lag: The Science-Backed Protocol
• Normal HRV by Age: Research vs. Wearable Numbers
• Does a Cortisol Detox Work? What the Science Says