Sleep Loss — A Debt Built Night by Night, Reminders That Lag Behind

The Day After an All-Nighter

I pulled an all-nighter for a deadline. Sitting in the morning meeting, I cannot remember a sentence I just heard, and replying to a single email takes me much longer than usual. When you are deprived of sleep, your mood, thinking power, and physical movement all drop across the board. Among these, the abilities that weaken the most are the power to hold onto attention and the power to briefly keep information in your head (PMID 15798944). The front part of the brain that regulates attention is sensitive to a lack of sleep. When you scan the brains of sleep-deprived people, activity in this area is reduced (PMID 42547208).

There is also a pattern to which functions fall apart the most. The losses were most pronounced when the all-nighter lasted 45 hours or more, extending into a second night. Looking at the type of task, the loss was clearer in speed than in accuracy. Moreover, performance dropped more sharply in tasks that required keeping to a set beat rather than tasks you could do at your own pace (PMID 1533178).

The desire to do things also shrinks. On days when you have not slept, the same task feels harder, and your motivation to push through and accomplish it weakens. If you dangle an external reward like money at times like this, your effort increases and part of the loss is made up (PMID 31072557). This means that the slump on the day after an all-nighter is a mix of not being able to and not wanting to.

Even when pulling the same all-nighter, the degree of breakdown varies from person to person. Some people's frontal brain areas, which are responsible for judgment and planning, are heavily shaken by a lack of sleep, while others are less shaken (PMID 15798944). Therefore, you cannot gauge your own state just because the person next to you looks fine.

Attention That Keeps Cutting Out

When you are short on sleep, your waking state becomes unstable. You feel fine, and then experience repeated, momentary cutouts (PMID 15798944). An attention lapse is a brief gap where your response to a moment that requires action is delayed by a long time or missing entirely. The less sleep you get, the more frequent these lapses become (PMID 15798944).

A microsleep is an extremely short sleep that sneaks into waking hours. When you are short on sleep, microsleeps increase (PMID 15798944, PMID 31072557). There is a hypothesis that staying awake for a long time causes these microsleeps to slip in between moments and create attention lapses, and the results of gathering all-nighter studies matched this hypothesis (PMID 1533178).

Because microsleeps cause attention lapses, in tasks like driving they can lead to accidents. However, the standard test for measuring sleepiness in hospitals does not separately count these brief intrusions (PMID 40482398). This means that these short moments are missing from the sleepiness measured by tests. The day after an all-nighter, those momentary switch-offs are where the danger lies.

When Days of Sleeping Just a Little Less Add Up

More common than pulling an all-nighter is living with a pattern of sleeping a little less over several days. Chronic sleep restriction is a state where, instead of pulling an all-nighter, a person continuously sleeps slightly less than needed over multiple days. Researchers view the sleep deprivation that actually happens in society as closer to this pattern than to all-nighters (PMID 15798944).

The result of spending two weeks sleeping 6 hours or less a day was a loss equivalent to staying awake for up to two full nights (PMID 12683469). This result came from a single laboratory experiment. Among 48 healthy adults, some were randomly assigned to spend either 4, 6, or 8 hours in bed each day for 14 days. For the 4-hour and 6-hour groups, performance on all tasks kept getting worse as the days went on, and the less they slept, the worse it got. For comparison, other participants went three nights without a wink of sleep.

Based on this, the researchers concluded that even moderate-looking sleep restriction can severely impair daytime cognitive function in healthy adults (PMID 12683469). Under the condition of staying awake for three straight nights, the wakeful performance and sleep EEG responses were disproportionately large relative to the amount of sleep lost (PMID 12683469). This means the toll of pulling an all-nighter was greater than what would be estimated from the amount of lost sleep alone.

The losses that accumulate this way are often compared to a debt. Sleep debt is a term comparing the sleep missed relative to what the body needs, piling up day by day, to a debt. The same experiment also shows how this debt is counted. The baseline is about 15.84 hours of wakefulness per day. When adding up all the hours spent awake past this line each day, the number of attentional lapses increased almost in proportion to that accumulated amount. It was the same whether from sleep restriction or an all-nighter.

Researchers noted that it is best to understand sleep debt as a cost accumulating in the brain each time someone stays awake past that line (PMID 12683469). However, this baseline value is from a statistical model fitted to laboratory data from 48 individuals aged 21 to 38. It cannot be transposed as the exact amount of sleep needed by everyone. What this experiment measured was the shape of how debt accumulates, and whether sleeping extra by the amount missed pays it off precisely is not in the results reported by this experiment.

Debt usually comes with collection notices. The equivalent of a collection notice for sleep debt is sleepiness. However, for debt accumulated by sleeping a little less over multiple days, sleepiness does not grow as large as the debt. In the same experiment, the sleepiness scores self-rated by participants rose on the first day of reduced sleep, and only edged up slightly afterward. They could not even clearly tell apart the 4-hour and 6-hour groups (PMID 12683469).

Judging by the sleepiness scores, participants appeared largely unaware of their growing cognitive losses (PMID 12683469). This is an interpretation drawn by researchers from the sleepiness scores. It is not the result of having participants directly evaluate their own performance. Also, this result belongs to multi-day sleep restriction, and under other conditions, sleepiness can remain high (the banking-sleep experiment coming up later).

A review summarizing chronic sleep restriction experiments describes the same picture. Even amid the feeling of having adapted to sleepiness, the losses pile up as time goes on (PMID 15798944). Researchers noted that this may be why chronic sleep deprivation is often thought of as harmless (PMID 12683469). The feeling of having adapted to a life of sleeping less is no proof that performance has recovered.

Can the Debt Be Paid Back?

On Friday night, you resolve to sleep in over the weekend to pay back the sleep you missed during the week. The evidence on whether this actually pays it back is partial. Weekend catch-up sleep and naps show some protective effects for cardiometabolic health. However, those effects appear only under specific conditions and cannot be considered equivalent to regular sleep, as one review concludes (PMID 42788298 — English abstract of a review written in a non-English language). What those conditions are is not stated in this abstract. How much weekend sleep reverses the attention and performance deficits of the accumulated week is not directly addressed by the evidence gathered in this article.

Banking sleep in advance, in other words the method of saving up sleep, has also been studied. If you increase your sleep ahead of an all-nighter or sleep restriction, attentional lapses decrease and reactions speed up during the subsequent sleep deprivation. However, subjective sleepiness in this case often remained high, which differs from the chronic sleep restriction in the previous section. The experimental studies numbered seven papers with about 140 participants, mostly healthy young adults (PMID 41892386). The authors evaluated it as a useful supplemental tool for shortly scheduled sleep deprivation, and noted that it does not completely eliminate subjective fatigue (PMID 41892386).

As means to soothe immediate sleepiness, caffeine and naps have been studied. Gathering seven randomized trials that gave caffeine to sleep-deprived soldiers, caffeine distinctly increased alertness and reaction compared to a sham pill. The authors evaluated this as a short-term supplemental tool and noted that the dosage and timing require further study (PMID 42761483).

Naps showed the greatest cognitive enhancement when they were under 60 minutes. Naps exceeding 60 minutes were associated with the risk of heart disease and diabetes (PMID 41710553). An association means they appear together, so it does not go as far as meaning that long naps cause disease. What the evidence says about caffeine and naps is about immediate alertness and performance. Whether the entire accumulated debt is paid back is not addressed by these pieces of evidence.

What Sleep Is Believed To Do

The four functions commonly attributed to sleep are memory, brain clearance, metabolism, and immunity. The strength of the evidence varies for each role. To discuss the first two, we need the names of three sleep stages.

REM sleep is sleep in which the eyes move rapidly beneath closed eyelids and most body muscles relax and do not move, and during which many vivid dreams occur. Non-REM sleep is the rest of sleep that is not REM sleep, divided into several stages ranging from light sleep to deep sleep. While it is true that there are many vivid dreams in REM sleep, dreams also occur in non-REM sleep. Deep sleep is the deepest stage of non-REM sleep, where brain waves change into large, slow waves, which is also called slow-wave sleep. How the stages cycle back and forth throughout the night is covered by another piece in this series, "Sleep Stages and Sleep Cycles".

The first is memory. Memory consolidation is the process by which newly learned memories settle into a form that lasts a long time. That sleep helps hold onto memories has been established by over a century of research (PMID 23589831). As for how it helps, this explanation is the current leading theory: memories learned during the day are replayed during deep sleep and transferred to a long-lasting store, and the subsequent REM sleep can stabilize them (PMID 23589831).

There is experimental evidence as well. This is an experiment where learning cues, such as sounds, are given during study, and those cues are played back to the sleeping person. When 91 studies and 2,004 people are pooled, memory improved by a small amount but clearly. The effect appeared only during non-REM sleep, and did not appear during REM sleep or while awake (PMID 32027149).

However, not all memories are the same. Learning skills that involve moving in a set order is called motor learning. In motor learning, there was no evidence that sleep further boosts learning, and the role of sleep in stabilizing those memories could not be concluded from studies so far (PMID 25822130). The hypothesis that sleep generally leaves behind better memories of emotionally charged scenes was also not supported by the pooled results. The possibility that REM sleep and deep sleep support different kinds of memory remains open (PMID 32179376).

The second is brain clearance. The brain and spinal cord are bathed in a clear fluid called cerebrospinal fluid, and the spaces between cells in brain tissue are filled with a fluid called interstitial fluid. The glymphatic system is a proposed flow in which cerebrospinal fluid enters the brain along microscopic pathways around arteries (perivascular spaces), mixes with interstitial fluid, and carries away waste products. True to its name, it is a proposed flow. Within brain tissue, substances can be carried along by fluid flowing like a river, or they can spread on their own, the way ink diffuses in water. The latter is called diffusion. Debate continues over how much fluid flow and diffusion respectively contribute to moving substances within brain tissue (PMID 42134763, PMID 42571787).

There is an explanation that this flow is most efficient during sleep and also clears out proteins associated with Alzheimer's disease. However, that evidence has come mostly from rodents such as mice, and imaging indicators showing this flow in humans still have limitations, which is cited as a point of debate (PMID 42571787).

The human-side evidence is mostly indirect observation. Human studies looking at how sleep and lifestyle factors like blood pressure connect to indicators related to this flow number 105, with sleep-related studies being the most numerous at 45. When pooled, an association is visible. However, most of these were studies that merely observed people and used indicators that indirectly estimate the flow. Out of 105 papers, only five measured clearance in a way close to direct (PMID 42723212).

The authors of one review consider that evidence is growing that a breakdown of clearance function is a cause of brain disease. At the same time, they point out that whether that breakdown is a cause or a byproduct of disease remains an unresolved question (PMID 42430835). The explanation that the brain is washed during sleep comes mainly from animal evidence, and in humans it is still at the stage of observing an association.

The third is metabolism. In experiments, reducing sleep alters the body's energy balance. Insulin, the hormone that lowers blood sugar, becomes less effective, and the function of cells that produce insulin also changes (PMID 34779405). Changes in hormones that control appetite are also reported. However, the evidence for this part relies on data observing people (epidemiological data) and animal experiments (PMID 36558355). This differs from the evidence in the preceding sentence, which came from experiments where sleep was reduced in humans.

The fourth is immunity. Healthy adults who slept less around the time of vaccination produced fewer antibodies. This association was clear only when sleep was measured by devices (304 people), and self-reported sleep duration (504 people) did not cross the statistical threshold predetermined by the researchers. This means it was not clear enough to separate from a difference arising by chance. When broken down by sex, the effect was large in men but not clear in women (PMID 36917932). The authors suggested that sleeping sufficiently before and after vaccination may boost the antibody response, while writing that large-scale, well-controlled studies are urgently needed (PMID 36917932).

The Observation That Too Little or Too Long Sleep Brings Many Diseases

Compared to people who sleep 7 to 8 hours a day, the risk of death was 14% higher in those who sleep less than 7 hours, and 34% higher in those who sleep 9 hours or more. This is the result of combining 79 studies, all of which are observational studies (PMID 40072785). An observational study is research that does not interfere with people's lives, leaves them to sleep as they naturally would, and watches what happens afterward. Therefore, observational studies only show what appears together and cannot determine on their own whether one thing caused another.

Looking at the numbers again, it is not just a story about the short end. A U-shaped association is a shape where the risk is high at both ends of short and long sleep times, making the graph curve like the letter U. Death, cardiovascular disease, and stroke all drew this shape, and the spot with the lowest risk was about 7 hours a day (PMID 28889101).

The shape is slightly different for each disease. People who sleep short developed more diabetes, high blood pressure, cardiovascular disease, and obesity (PMID 27743803). People who sleep long also developed more diabetes, cardiovascular disease, stroke, and obesity, but showed no association with high blood pressure (PMID 28890167). For high blood pressure, the association with the short end is consistent across multiple analyses. For the long end, there is either no association or results vary across analyses (PMID 39008468, PMID 32587332, PMID 25902823).

There are also analyses that grade how reliable these associations are. The association that received the highest grade here was a single one between long sleep and death. The association between short sleep and overweight or obesity, along with a few associations between long sleep and stroke, was one grade lower (PMID 35127769). This grading assesses how reliable the association is. It does not assess how large the risk is, so it cannot be read to mean that long sleep is the most harmful.

Does short sleep then cause disease? Observational studies alone cannot determine the direction. Whether the association between short sleep and obesity or diabetes is causal, and which one triggers which, still requires evaluation, and the possibility of a bidirectional relationship is also being examined (PMID 37156196). Researchers noted that for long sleep as well, it must be clarified in the future whether that association is causal and modifiable (PMID 28890167).

There are also methods trying to look closer at causality. Mendelian randomization is an analytical method that estimates the tendency to sleep short or long using genotypes determined at birth, in order to reduce the problem of other differences like lifestyle habits stepping in. Genotypes are determined like drawing lots when inherited from parents, and do not change depending on lifestyle habits chosen during life. Therefore, other lifestyle differences are less mixed among groups divided by genotype.

The authors of an analysis combining 51 such studies concluded that genetic evidence confirms causal relationships between sleep characteristics and various diseases (PMID 41517697). This analysis pooled genetic predispositions for short sleep, long sleep, total sleep duration, and insomnia. It also reported connections with various diseases such as cardiovascular disease, diabetes-related markers, neurological disorders, and psychiatric disorders. However, because the results are not broken down by predisposition in the abstract, the abstract alone cannot determine which side, short or long sleep, is linked to which disease (PMID 41517697).

Whether increasing sleep reduces disease is another question. Emerging trials suggest that increasing sleep may slightly improve weight management, blood sugar, and blood pressure (PMID 41035532).

Activity levels also change the picture. An analysis that examined twelve papers and pooled five of them used people with high activity and intermediate sleep as the comparison baseline. Compared to this baseline, the association between short sleep and mortality risk was not clear in highly active people. In people with low activity, the risk was high whether their sleep was intermediate or short. People who slept long had a high risk regardless of their activity level. The authors noted that activity may offset the risks of short sleep (PMID 40217556). Both short and long sleep appear alongside disease, but those numbers are associations from observation, and how much things change when sleep is modified has only been confirmed to a small extent so far.

How Many Hours Should We Sleep?

The recommended amount of sleep for healthy adults is a range. A panel of experts from the National Sleep Foundation recommended 7 to 9 hours for young adults and adults, and 7 to 8 hours for older adults, while also discussing the limitations and caveats of this recommendation (PMID 29073398). From 7 hours to 9 hours, it is a range. The panel was composed of experts from multiple fields, evaluated the evidence, and reached a consensus through voting (PMID 29073398). The same year, the American Academy of Sleep Medicine and the Sleep Research Society also reviewed literature evidence and issued consensus recommendations through expert voting, documenting the assumptions and difficulties they encountered along the way (PMID 26039963, PMID 26235159). The recommended hours are numbers agreed upon by experts after gathering evidence.

The commonly heard "everyone needs 8 hours" does not match these pieces of evidence. In observational studies, the point where the risk was lowest was about 7 hours (PMID 28889101). Another analysis pooling 35 papers used 7 hours of nighttime sleep as the baseline. Compared to this, both people sleeping 4 hours and people sleeping 8 hours at night had a similarly elevated risk of death of about 7% (PMID 26900147). The authors of that analysis concluded that 7 hours a day should be recommended, but the evidence comes from observational studies (PMID 26900147).

Because this is an association from observational studies, it does not mean that reducing sleep for someone who slept 8 hours lowers their risk. Whether the association on the longer sleep side is causal and modifiable has not yet been revealed (PMID 28890167). Conversely, the fact that the range has a width does not mean it is fine to sleep less, and the reason lies in the experiments we saw earlier. Experimental participants who slept slightly less for multiple days mostly appeared not to notice their own loss when measured by sleepiness scores (PMID 12683469).

What invites sleep and when we feel sleepy is covered by another part of this series, "Sleep Pressure and the Body Clock", and lifestyles where the clock is misaligned, such as shift work, are also looked at there. The problems of being unable to fall asleep every night or stopping breathing during sleep are covered in "Insomnia and Sleep Apnea".

Drowsy Driving

A person who has been sleeping 6 hours a day in bed for two weeks gets behind the wheel on their commute home. In the same experiment, conditions with 6 hours of sleep or less per day showed progressively worse performance as the days went on, reaching impairments comparable to up to two all-nighters. The less they slept, the worse they got, while sleepiness only crept up a little bit over that time (PMID 12683469). When sleep falls short, microsleeps increase (PMID 15798944), and the attentional lapses caused by microsleeps can lead to road accidents (PMID 40482398).

Drowsy driving is a real danger. It is estimated that sleepiness is involved in 10 to 20% of car crashes (PMID 34461582). Pooling studies that mostly snap a picture of a single time point, sleepiness behind the wheel and sleep deprivation were linked to traffic accidents. The main factor bringing on sleepiness while driving was a lack of sleep (PMID 36205209).

The same pattern appears after long shifts. When resident physicians—doctors in training at hospitals—drive after finishing long shifts exceeding 16 hours, dangerous driving events such as collisions or falling asleep at the wheel increase substantially. In one study, the value comparing the likelihood of such events occurring after a night shift to that after a day shift (the odds ratio) was 3.9 (PMID 39697913).

The signal of sleepiness has two directions. As seen earlier, when people slept a little less each day over several days, sleepiness did not grow in proportion to the accumulated impairment (PMID 12683469). Therefore, the feeling that you are not very sleepy is not proof that you are okay. Conversely, when sleepiness is felt severely, that sleepiness was a signal predicting accidents in driving studies. In an analysis pooling 34 simulation and real-road driving studies, drivers did recognize sleepiness, and the more severe the sleepiness, the more clearly it predicted accidents and lane departures (PMID 34461582).

There is a caveat to this analysis. How accurately drivers perceive their own sleepiness has long been a matter of debate, and this analysis started from that debate. Sleepiness was tracked more accurately in driving simulators than on real roads. How this relationship changes depending on the degree of sleep deprivation remains a blank space (PMID 34461582). The researchers concluded that stopping to drive when feeling sleepy can reduce the risk of accidents (PMID 34461582).

Countermeasures to reduce drowsy driving have also been studied. For resident physicians after long shifts, providing taxi and ride-hailing expense support, scheduled rest, drowsy driving risk education, and caffeine have been suggested as ways that may lower risk. Their effectiveness still requires further study (PMID 39697913). Caffeine improved alertness and reaction in sleep-deprived soldiers, but it was evaluated as a short-term supplemental tool and was not measured for whether it reduces driving accidents (PMID 42761483). The action pointed to in the conclusion by the authors of the driving study pooling sleepiness self-awareness is to stop driving when sleepy. Measures like rest and caffeine are being studied as supplements alongside that.

Where to Look Next

• You can read on about how sleep helps shelve memories in "The Three Steps of Memory — Putting It In, Keeping It, Taking It Out".
• To look deeper into the brain-cleaning hypothesis, there is "The Glymphatic System". It is good to bring the evidence grades seen in this article along with you.
• The pathways by which social media reduces sleep in adolescents are covered in "Is Social Media Harming Adolescent Mental Health?".
• Other parts of this series: "Sleep Stages and Sleep Cycles", "Sleep Pressure and the Body Clock", and "Insomnia and Sleep Apnea".
• The unanswered questions converge into one: how much of the disease risk attached to short and long sleep is due to sleep itself, and how much does it decrease if you change your sleep?

Source Papers

• "Neurocognitive consequences of sleep deprivation." (2005) — PMID 15798944 — Sleep deprivation particularly impairs attention and working memory, increases microsleeps and attentional lapses, accumulates deficits under chronic restriction even when adaptation to sleepiness occurs, and depends heavily on individual vulnerability.
• "Sleep, attention, and executive function in health and disease." (2026) — PMID 42547208 — Attention is vulnerable to sleep deprivation, and prefrontal activity decreases during sleep deprivation.
• "Meta-analysis of the relationship between total sleep deprivation and performance." (1992) — PMID 1533178 — 27 studies, strongest associations found at 45+ hours of all-nighters, speed metrics, and fixed-pacing tasks, consistent with the attentional lapse (microsleep) hypothesis.
• "Sleep deprivation, effort allocation and performance." (2019) — PMID 31072557 — Sleep deprivation reduces the willingness to exert effort, while external rewards partially offset the loss.
• "Shedding light on microsleep episodes for comprehensive sleepiness assessment: a narrative review." (2025) — PMID 40482398 — Microsleeps represent brief intrusions of sleep missed by standard sleepiness tests, potentially linked to road accidents via attentional lapses.
• "The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral functions and sleep physiology from chronic sleep restriction and total sleep deprivation." (2003) — PMID 12683469 — 48-participant randomized experiment, 14 days of 6 hours or less equals the loss of up to two all-nighters, sleepiness ratings barely increase so deficits go unrecognized, proportional to the accumulation of wakefulness exceeding 15.84 hours.
• "[Sleep alterations as a risk factor for cardiometabolic health: emerging evidence and clinical implications]." (2026) — PMID 42788298 — Non-English review (English abstract), weekend catch-up sleep and naps offer partial protection under specific conditions but are not equivalent to regular sleep.
• "The Role of Sleep Banking in Reducing Cognitive and Motor Impairments from Subsequent Sleep Restriction: A Narrative Review." (2026) — PMID 41892386 — Pre-sleep extension raises objective margins during subsequent sleep deprivation, but subjective sleepiness remains, 7 studies, about 140 participants.
• "Acute caffeine supplementation as a nutrition-based strategy to mitigate sleep-loss-related cognitive and operational performance impairments in military personnel: a systematic review and meta-analysis." (2026) — PMID 42761483 — 7 military randomized trials, caffeine improves alertness and reaction during sleep loss, a short-term aid.
• "Multiple Health Outcomes of Daytime Napping: A Comprehensive Umbrella Review." (2026) — PMID 41710553 — Naps under 60 minutes maximize cognitive enhancement, while naps exceeding 60 minutes are linked to heart disease and diabetes risks.
• "About sleep's role in memory." (2013) — PMID 23589831 — Sleep's benefit in memory maintenance is established, active system consolidation theory via reactivation during deep sleep, REM sleep can stabilize.
• "Promoting memory consolidation during sleep: A meta-analysis of targeted memory reactivation." (2020) — PMID 32027149 — 91 experiments, 2,004 participants, cue re-presentation during sleep improves memory (g=0.29), effective only in non-REM sleep.
• "Sleep and motor learning: Is there room for consolidation?" (2015) — PMID 25822130 — Accounting for confounding variables, no evidence shows sleep enhances motor learning, and a stabilization role cannot be concluded either.
• "Sleep's impact on emotional recognition memory: A meta-analysis of whole-night, nap, and REM sleep effects." (2020) — PMID 32179376 — Hypothesis that sleep generally better enhances emotional memory is not supported.
• "Alzheimer's disease as a disorder of time-gated clearance: Circadian regulation of Glymphatic function in neurodegeneration." (2026) — PMID 42571787 — Explanation that glymphatic flow is most efficient during sleep, rodent data dominate, human imaging metric limitations are a point of debate.
• "The Glymphatic system: A dynamic regulator of brain health and therapeutic target." (2026) — PMID 42134763 — Ongoing debate surrounds the contributions of flow versus diffusion.
• "Modifiable Physiological Factors Associated With Human Glymphatic and Interstitial Fluid Dynamics-Related Measures: A Scoping Review." (2026) — PMID 42723212 — 105 human studies across general lifestyle physiological factors (45 related to sleep), associations exist but rely heavily on observational studies and surrogate metrics, direct measurements in 5 studies.
• "Glymphatic dysfunction in neurodegeneration: From impaired clearance to mechanism-driven therapeutic innovation." (2026) — PMID 42430835 — While acknowledging accumulating evidence as a cause, whether it is a cause or a byproduct remains an unresolved issue.
• "The links between sleep duration, obesity and type 2 diabetes mellitus." (2021) — PMID 34779405 — Experimental sleep restriction is linked to changes in energy balance, insulin resistance, and beta-cell function.
• "Sleep Deprivation and Central Appetite Regulation." (2022) — PMID 36558355 — Appetite hormone changes, based on epidemiological data and animal models.
• "A meta-analysis of the associations between insufficient sleep duration and antibody response to vaccination." (2023) — PMID 36917932 — Device-measured short sleep reduces antibody response (304 participants), self-report falls short of criteria (504 participants), pronounced in males.
• "Imbalanced sleep increases mortality risk by 14-34%: a meta-analysis." (2025) — PMID 40072785 — 79 cohort studies, compared to 7–8 hours, under 7 hours HR 1.14, 9+ hours HR 1.34.
• "Relationship of Sleep Duration With All-Cause Mortality and Cardiovascular Events: A Systematic Review and Dose-Response Meta-Analysis of Prospective Cohort Studies." (2017) — PMID 28889101 — Mortality, cardiovascular events, coronary events, and stroke all show a U-shape, lowest risk around 7 hours.
• "Short sleep duration and health outcomes: a systematic review, meta-analysis, and meta-regression." (2017) — PMID 27743803 — 153 studies, 5.17 million people, short sleep linked to mortality, diabetes, hypertension, cardiovascular disease, and obesity.
• "Long sleep duration and health outcomes: A systematic review, meta-analysis and meta-regression." (2018) — PMID 28890167 — 137 studies, long sleep linked to mortality, diabetes, cardiovascular disease, stroke, and obesity, unrelated to hypertension, causality and variability remain future tasks.
• "Association between sleep duration and hypertension incidence: Systematic review and meta-analysis of cohort studies." (2024) — PMID 39008468 — Short sleep is linked to hypertension incidence (HR 1.07), long sleep is unrelated.
• "The association between sleep duration and hypertension: a meta and study sequential analysis." (2021) — PMID 32587332 — Short sleep is linked to hypertension (RR 1.161), 8+ hours is not significant due to insufficient samples.
• "Relationship between Duration of Sleep and Hypertension in Adults: A Meta-Analysis." (2015) — PMID 25902823 — Hypertension odds ratio is U-shaped, confidence interval on the 9+ hour side spans 1, authors conclude both extremes may be risk factors.
• "Sleep Duration/Quality With Health Outcomes: An Umbrella Review of Meta-Analyses of Prospective Studies." (2021) — PMID 35127769 — Confidence grading of 85 meta-analyses, long sleep-mortality is 'highly suggestive', short sleep-obesity is 'suggestive'.
• "The interrelationship between sleep, diet, and glucose metabolism." (2023) — PMID 37156196 — Causality and direction of short sleep and obesity/diabetes association need evaluation, bidirectional possibility.
• "Evaluating the causal connections between sleep duration and disease prevalence: A comprehensive systematic review and meta-analysis of Mendelian randomization studies." (2026) — PMID 41517697 — 51 Mendelian randomization studies, authors conclude genetic evidence confirms causal links between sleep traits and multiple diseases.
• "Sleep and Cardiometabolic Health: A Narrative Review of Epidemiological Evidence, Mechanisms, and Interventions." (2025) — PMID 41035532 — Sleep extension trials slightly improve body weight, blood glucose, and blood pressure.
• "The associations between physical activity, sedentary behaviour, and sleep with mortality and incident cardiovascular disease, cancer, diabetes and mental health in adults: a systematic review and meta-analysis of prospective cohort studies." (2023) — PMID 40217556 — Meta-analysis of 5 out of 12 studies, short sleep-mortality link not pronounced with high activity (may offset), risk increases with low activity and short sleep, or inactivity regardless of long sleep.
• "Joint Consensus Statement of the American Academy of Sleep Medicine and Sleep Research Society on the Recommended Amount of Sleep for a Healthy Adult: Methodology and Discussion." (2015) — PMID 26235159 — Methodology, literature, voting process, assumptions, and challenges of the consensus statement on recommended sleep duration.
• "Recommended Amount of Sleep for a Healthy Adult: A Joint Consensus Statement of the American Academy of Sleep Medicine and Sleep Research Society." (2015) — PMID 26039963 — AASM and SRS issued consensus recommendations for adult sleep duration using a modified RAND appropriateness method (no numbers in abstract).
• "National Sleep Foundation's updated sleep duration recommendations: final report." (2015) — PMID 29073398 — Multidisciplinary expert panel consensus, recommends 7–9 hours for young adults and adults, 7–8 hours for older adults, discussing limitations and caveats.
• "Nighttime sleep duration, 24-hour sleep duration and risk of all-cause mortality among adults: a meta-analysis of prospective cohort studies." (2016) — PMID 26900147 — 35 studies, compared to 7 hours, nighttime sleep of 8 hours has an RR of 1.07, risk increases toward both extremes.
• "I think I'm sleepy, therefore I am - Awareness of sleepiness while driving: A systematic review." (2021) — PMID 34461582 — Sleepiness is involved in 10-20% of accidents, drivers are aware of sleepiness, sleepiness predicts accidents, stopping when sleepy can reduce risk.
• "Risk assessment of road traffic accidents related to sleepiness during driving: a systematic review." (2022) — PMID 36205209 — Sleepiness and sleep deprivation while driving are linked to traffic accidents, with sleep deprivation as the primary factor for sleepiness.
• "Resident Physicians Have an Increased Risk of Adverse Driving Events Following Extended-Duration Work Shifts: A Systematic Review." (2024) — PMID 39697913 — Odds of adverse driving events are 3.90 times higher after shifts exceeding 16 hours, countermeasures lack verified effectiveness.