For how we live.
The science of rest, and the industry built on measuring it.
Sleeping apart is increasingly common and can genuinely cut disturbance. But the lab evidence is double-edged: sharing a bed measures as slightly worse sleep yet feels better, and may even boost REM.
Probably not ideal. Even moderate light during sleep measurably raises night-time heart rate and next-morning insulin resistance, and in one large study women who slept with a light or TV on gained more weight.
A warm bedroom measurably degrades sleep, and the best-studied comfortable band sits at roughly 20–25°C. The effect is real but personal, and heat matters more than most people assume.
Knowing you are dreaming and sometimes steering it is real and surprisingly common, and it can be trained. But deliberately inducing it may carry a mental-health cost for some people.
The viral drink pairs tart cherry juice with a magnesium supplement. Both have some human trial data for sleep, but it is thin, small and low-quality — far short of what the trend implies.
The viral countdown says no caffeine 10 hours before bed, no food at 3, no work at 2, no screens at 1. The caffeine and food steps have real evidence behind them; the rest is thinner.
Yes, within limits. Short sleep is linked to more colds and weaker vaccine responses, but the strongest effects turn up when sleep is measured objectively, not from what people report.
When researchers tested people with insomnia, those told to picture an engaging scene fell asleep faster. A vaguer instruction to just distract yourself, the counting-sheep approach, did not.
For most healthy adults the measured difference is small. But side-sleeping has real evidence in three specific situations: nocturnal reflux, obstructive sleep apnoea and late pregnancy.
In the phase 2 Vibrance-1 trial, alixorexton lifted mean sleep latency from about 2 minutes on placebo to 24 or more on the drug. The six-week study was funded by the manufacturer.
That heavy, foggy feeling on waking is often sleep inertia — a normal, temporary state, not a sign you slept badly. It bites hardest when you wake from deep sleep or during the biological night.
A large public randomised trial found reading a book in bed nudged sleep quality up. But the format matters: paper helps, while a glowing screen can quietly work against you.
Grinding and clenching in sleep is common, and in healthy people experts now class it as a behaviour rather than a disorder. But the evidence on how common it is, and what causes it, is thinner than you would expect.
For a few seconds to a couple of minutes you are awake but frozen, sometimes with a sense of a presence in the room. It is REM sleep intruding into wakefulness — common, frightening and, on its own, harmless.
Nightmare disorder affects about 4% of adults. The treatment with the clearest backing is not a drug but a rehearsal technique that rewrites the dream while you are awake.
Childhood sleepwalking is usually harmless and outgrown. In adults it is rarer but a different proposition — one clinic series found most had done something violent during an episode and one in six had been injured.
A short nap does not wreck most people's night sleep, and can help offset sleep debt. The evidence that gives pause is about long naps: naps of 60 minutes or more track with worse cardiovascular outcomes.
When you eat may shape how you sleep, but the science is young. The clearest signal is that a heavy meal close to bedtime disrupts sleep, while simply shrinking your eating window neither reliably helps nor harms it.
The 2024 American Academy of Sleep Medicine guideline strongly backs intravenous iron and gabapentinoids, and now suggests against the dopamine agonists that were first-line for two decades.
A systematic review of 38 studies rated the evidence that continuous white noise improves sleep as very low, and warned it may worsen sleep for some. A few small studies do find a benefit.
Across the trials, the best-supported remedy is melatonin taken near destination bedtime after crossing several time zones. Light exposure is the other real lever; sleeping pills chase the symptom, not the clock.
Physically acting out dreams is one of the strongest known early signs of Parkinson's and related diseases. In a 1,280-patient cohort, most converted within 12 years — but it is a risk marker, not a verdict.
Waking up drenched is common and, the best review found, usually nonspecific — not a sign of shortened life. In women the leading identifiable driver is the menopause, which can last far longer than most expect.
In a 304-veteran trial, prazosin matched placebo on trauma nightmares and sleep quality. Sleep medicine's one recommended treatment for nightmare disorder is a behavioural therapy, not a drug.
The only systematic review to pool the trials found a warm shower or bath in the evening does speed sleep onset — but the effect is modest and depends on timing the heat one to two hours before bed.
Magnesium is one of the most popular over-the-counter sleep aids, but a 2026 review of the randomised trials concluded the evidence does not support it as a routine treatment for insomnia.
The most-cited meta-analysis found a faint signal that valerian might improve sleep, then flagged publication bias and poor-quality trials. Sleep-medicine guidelines suggest against using it.
The 2024 AASM guideline now advises against routinely using the dopamine agonists that were once standard, because over time they often make the condition worse — and puts anticonvulsants and iron first.
The evidence is real but modest, and stronger for anxiety than for sleep itself. One 120-person trial found a large benefit; pooled analyses across studies find a smaller, more uncertain effect.
The old antidepressant is prescribed at low doses for insomnia far more than its evidence supports. A Cochrane review found only a modest, low-certainty effect on subjective sleep and none on measured sleep efficiency.
In the largest pooled cohort studies, the lowest-risk amount of sleep sits closer to seven hours, and both too little and too much track higher mortality. But the curve is a warning sign, not a prescription.
Two meta-analyses find evening exercise leaves sleep largely unchanged or slightly improved. The real exception is vigorous exercise finishing within about an hour of bedtime.
A prescribed home apnea test can diagnose clear-cut cases, and one under-mattress sensor flagged severe apnea with 86% sensitivity — but both miss milder disease and events a lab would catch.
OSPREY implanted every participant, then randomly decided whose device was switched on. That design answers a question most implant studies leave open, and it has limits the authors state plainly.
AD109 cut the apnoea-hypopnoea index by four events an hour more than placebo over 26 weeks. It did not improve fatigue, and one in five participants stopped because of side effects.
In a 12-person crossover trial, blue-enriched daytime light with reduced evening blue did not cut time spent awake at night, though it nudged a brainwave marker of sleep.
Two small pilot studies in 18-to-25-year-olds found morning bright light on its own shifted circadian timing little, while adding a timed schedule and evening blue-blockers advanced it about 35 minutes.
Both brands sell drug-free external nasal dilators with much the same evidence behind them. The differences are range, materials, and the fact that only one also sells a mouth tape.
In 7,530 people from two long-running cohorts, central sleep apnea with high oxygen overshoot carried elevated cardiovascular risk. The same apnea with low overshoot did not.
It is a strip of tape over the lips at night, meant to force nose breathing. The idea is simple; the safety caveat is the part that matters most.
In 1,238 people from two community cohorts, apnea severity showed no link to depressive symptoms. Sleep fragmentation did, and that association disappeared once insomnia symptoms entered the model.
The remedies that work are the ones matched to the cause — nose, mouth, tongue or palate. Here is what the trials support, and the sign that means see a doctor first.
A systematic review of longitudinal MRI after CPAP for sleep apnoea finds changes reported in every domain and reproducible effects in none — a useful map of what the evidence does not yet show.
In 152 adults tested against two nights of polysomnography, the Galaxy Watch reached an AUROC of 0.94 for moderate-to-severe sleep apnea — with 66.7% specificity at the setting it ships with.
SleepCARE randomised 219 Australian women receiving chemotherapy for breast cancer to CBT-I, bright light therapy, both, or sleep hygiene. Only CBT-I moved insomnia — and nothing moved fatigue.
A meta-analysis of 27 studies found a dose-response pattern that runs against the nightcap logic: the sedative effect needs a high dose, and the REM disruption starts at a low one.
A 4,066-person survey series across Funafuti and the outer islands links second-hand smoke exposure — reported by 61% of participants — to obesity and elevated STOP-Bang scores.
The company's second major product launch since its July 2024 founding pairs nasal strips with the mouth tape it built its business on.
The company says it was built on a specific complaint — that existing sleep products on the market didn't hold up — and shipped its first product, a mouth tape, one month after founding.
Tirzepatide has been approved for obstructive sleep apnea since December 2024. A January 2026 reanalysis of the trial behind that approval separates what weight loss fixed from what treating the breathing fixed.
Nigerian adults scored well below US and Norwegian comparison samples on a six-dimension sleep health measure. In South Africa, students with backup power were far less likely to report worse sleep during outages.
Stanford's SleepFM, published in Nature Medicine in January 2026, is the most consequential sleep paper of the year. Its own limitations section explains why it is not yet a clinical test.
Only three of the 15 studies pooled in a Chilean meta-analysis used sleep testing at all. A Lima clinic analysis shows why that shortcut matters.
External nasal strips have real, repeatable evidence behind mild snoring relief. They have never been shown to treat sleep apnea, and the two problems are not the same problem.
The practice has real theory behind it and a genuine safety warning attached to it. The studies are small, mostly low quality, and say less than either side of the debate claims.
A National Sleep Foundation position statement and a new national survey land on the same problem: a third of US adults track their sleep, and the numbers they trust most are the ones devices measure worst.
Sleep genuinely gets lighter and more broken with age, though only its efficiency keeps declining after 60. Insomnia itself is not normal ageing, and its first-line treatment is not a pill.
About a third of adults on two continents wake at least three nights a week. The 7.7% who cannot resume sleep carry almost all of the daytime damage — and the cause is often not what the sleeper thinks.
The trend stacks mouth tape, magnesium, trackers and a 'sleepy girl mocktail'. Sleep physicians say one unglamorous treatment carries the evidence; most of the rest is theatre.
Twenty months of under-mattress sensor data from 95,819 people found that chronic irregularity was common, persistent, and tracked with 46% higher odds of hypertension.
In-lab trials that restricted sleep on weekdays and allowed recovery at the weekend found insulin sensitivity and glucose tolerance still worsened. Population surveys point the other way, and the designs explain why.
A Cochrane review found they may not reduce eye strain and could not determine whether they help sleep. A separate meta-analysis found small effects. Neither found the evidence anyone assumes exists.
The Pittsburgh Sleep Quality Index global score is the field's default outcome. Tested in a large South African cohort, its single-factor structure fit poorly and internal consistency was modest.
A laboratory trial found a light-emitting e-reader really does suppress melatonin and delay the body clock. Two 2026 field studies found the day-to-day effect runs almost entirely through when people go to bed.
Two American guideline bodies grade cognitive behavioural therapy for insomnia as a strong recommendation. Every drug in the sleep-medicine guideline — the ones it endorses and the ones it rejects — came out weak.
A consortium reduced overnight sleep studies to nine composites. Two tracked with better cognition. None tracked with whether a person later developed dementia.
The pooled effect is about seven minutes off the time it takes to fall asleep. The American Academy of Sleep Medicine recommends against it — on evidence it grades as weak in both directions.
Two national survey analyses find schedule unpredictability — not shift work itself — associated with insomnia symptoms, with a dose-response by how much advance notice a worker gets.
The number most people quote is not the number the American Academy of Sleep Medicine and Sleep Research Society agreed on. Their recommendation is a floor, set by expert vote rather than by trial.
After four weeks, the comorbid-insomnia group's blood pressure fell 20 points more than the apnea-only group. The finding flips the usual framing of COMISA as simply a worse-outcomes phenotype.
Both morning and evening chronotypes showed elevated risk against the middle ground, and the effect nearly doubled in people who also had sleep apnea.
A home-monitoring trial found some patients don't sleep on their backs at home the way they did during their diagnostic sleep study, while others simply failed the vibrating device meant to keep them off their backs.
A meta-analysis of six studies found a moderate effect on sleep duration and a small one on depressive symptoms — with high heterogeneity on both.
Stress peaked during the fasting month and fell afterwards, but depression and anxiety stayed flat. The consistent finding across every phase was that sleep quality tracked mental health.
In 377 patients, hypoxic burden — a metric combining how often, how deep, and how long oxygen drops — tracked with abnormal nighttime blood pressure patterns where conventional measures didn't.
The sinuses really do produce nitric oxide at concentrations far above the lower airway. Whether routing more of it through your lungs at night does anything measurable is a separate, unanswered question.
A meta-analysis of fully automated programmes found a large effect on insomnia severity. A review of 68 studies found the field has barely measured whether recipients find them acceptable.
Pooling eight trials, oral appliances left nearly 12 more breathing events per hour than CPAP. But the trials disagreed with each other so much that the researchers say that number itself should be read with caution.
Nearly twice as many CPAP-treated patients saw their HFpEF-like markers improve at one year compared with those who went untreated — in a population sleep specialists have historically deprioritized for treatment.
Pooled against the sleep lab, finger-worn devices call sleep versus wake at 87% but fall to the mid-60s on light sleep. An Oura meta-analysis found no significant gap on total sleep time.
Orexin antagonists were long avoided in sleep apnea over fears they'd suppress the arousals that end dangerous breathing pauses. Sixteen patients on daridorexant showed no meaningful change in breathing severity.
Women who reported vigorous physical activity had a lower arousal index, less time awake after falling asleep, and more total sleep time. The same association didn't hold for arousal index in men.
The best-validated sleep apnea screener works by being wrong in a specific direction. A 47-study meta-analysis of 26,547 people shows what a questionnaire can and cannot conclude about you.
The comparison group matters: against veterans with unspecified narcolepsy diagnoses, the mortality gap widened to 140% higher odds. The study can't yet say what's driving the excess deaths.
A 327-person survey of economically distressed Eastern Kentucky counties found insomnia prevalence ranged from 44% to 83% depending on income alone.
A 20-point rise in apnea severity tracked with about three times more blood pressure increase in adults with a BMI of 20 than in those with a BMI of 30, flipping a common assumption.
Moderate-to-severe sleep apnea alone roughly doubled all-cause mortality risk. Adding comorbid insomnia to the same patients raised it further — evidence for treating the combination as its own risk category.
Both machine types cut breathing interruptions equally well in a 108-patient study. But auto-adjusting PAP's constantly shifting pressure came with more time awake and less total sleep on night one.
Reanalyzed EEG data from a phase 3 trial finds lemborexant and zolpidem leave distinct signatures on sleep architecture. Only the slow-oscillation increase predicted less insomnia severity.
For years the disorder was treated off-label. A phase 3 randomised-withdrawal trial won lower-sodium oxybate the first US approval for it — but its design measures what is lost on stopping.
Researchers deliberately withdrew CPAP for three nights to simulate untreated sleep apnea, then tested propranolol against placebo. It slowed the heart-rate surges tied to apnea events, at the price of lighter sleep.
An analysis of nearly 139,000 Americans finds sedative misuse for sleep clusters among both the socially advantaged and the chronically unwell — pointing to two distinct drivers behind the same behavior.
In 2,286 patients with sleep apnea undergoing major surgery, a measure of how deeply and how long oxygen fell — not the apnea-hypopnea index — tracked 30-day cardiovascular events and death.
A four-arm randomised trial in 112 young women tested whether high-intensity circuit training and sleep-health counselling add up. On objective measures they did — by minutes.
The clearance story about sleep has been brain-centric. A Nature paper reports that in Drosophila, immune cells from the circulation do part of the job — and that blocking them shortens sleep.
Breathing through the mouth in sleep tracks closely with sleep apnea severity. The direction of the arrow is the part the research has not settled, and it changes what any intervention can claim.
A systematic review located 19 articles on melatonin in children aged 0 to 6. Trials covered 167 children, all with neurologic conditions. There was no efficacy data for typically developing children.
In a Pennsylvania cohort followed 7.5 years, the combination carried the highest risk when insomnia came with objectively short sleep. Insomnia with normal sleep duration showed no significant association.
Among 16-to-24-year-olds in NHANES, those who slept longer on weekends had 41% lower odds of daily depressive symptoms. The survey measured both at the same moment, which limits what can be concluded.
Standard questionnaires found no daytime difference between suvorexant and placebo. Four-times-daily smartphone prompts found fatigue worse in the morning and better later. The trial had 40 people.
Matched against a comparator diabetes drug, incretin agonists were associated with an 8% lower rate of CPAP procedures — alongside a 32% lower mortality that should make readers cautious.
A UK Biobank analysis of 96,365 people found a dose-response relationship between lifetime night-shift exposure and cardiovascular-kidney-metabolic disease — amplified in those who also slept short.
A small longitudinal study found reduced spindle activity over the temporal lobes in early Alzheimer's — and that lower temporal spindle density tracked with faster decline over up to seven years.
In a US cohort followed eight years, average weekday sleep rose from 8.1 to 8.3 hours while reports of fatigue and trouble sleeping increased. The socioeconomic gap in duration closed; the gap in symptoms did not.
In a US cohort wearing wrist actigraphs, self-reported sleep exceeded measured sleep by about half an hour, with the gap largest on weekends and among Black adolescents and those from lower-income households.
Age, sex, employment and socioeconomic level together explained under 1% of variance in insomnia severity. Perceived stress and generalised anxiety explained close to a third.
Among 1,000 adults in Mwanza, overnight oximetry found sleep apnea in 17% of people with HIV and 19% without. Excessive daytime sleepiness, oddly, was more common in those without HIV.
An Australian laboratory trial played 20-second noise bursts to 68 sleeping adults. Arousal probabilities were low for both, and lower for wind farm noise at 40 and 50 decibels.
A national insurance-claims analysis found long-term adherence far below what trial populations suggest, despite universal coverage and strict usage monitoring. The monitoring may be part of the problem.
A meta-analysis of 22 crossover trials with overnight polysomnography found consistent losses in total sleep time and efficiency. The age and dose subgroup differences everyone wants were not statistically significant.
In 24,223 Swedish workers followed to 2018, insomnia and long sleep raised cardiovascular risk only when accompanied by daytime impairment — and rising sleep duration preceded disease.
Fourteen Swedish primary care patients were interviewed alongside a trial of nurse-led group sleep restriction therapy. Depression and life circumstances were what got in the way.
Across 20,121 adults and 25 years of registry data, morningness markers predicted lower educational attainment but also lower odds of the bottom income quintile. Effects were absent in women.
In 99 men hospitalised for stress-related disorders, roughly four in five met probable criteria for trauma-associated sleep disorder — a condition that is not yet a recognised diagnosis.
Seltorexant held its effect over two weeks while zolpidem's faded, in a 364-person study run in 2017-2019 and published only now. The delay is disclosed in the paper itself.
In 311 adults aged 55 to 95, remission rates were 40% and 51% in the two intervention arms at one year, against 16% for online patient education. Adding human support helped a little.
Applying prevalence figures to UN population data yields 852 million adults with insomnia. Only 31 of 237 countries had a usable national estimate of their own.
FDA-cleared in September 2024, the feature caught 89% of severe obstructive sleep apnea in Apple's validation but 43% of moderate cases. It is a nudge toward testing, not a test.