Alcohol and sleep
Read the review →- Overstated
“A nightcap helps you sleep.”
One or two drinks barely change sleep onset in pooled data; faster onset appears at about 4–5 drinks, alongside dose-dependent REM loss and second-half disruption. The sedative benefit also fades within about a week of nightly use.
- Overstated
“Alcohol shortens sleep onset at every dose.”
This was the 2013 qualitative review's conclusion; the 2025 meta-analysis found a significant effect only at high doses (~0.85 g/kg).
- Partly true
“Alcohol gives you deeper sleep.”
Slow-wave sleep rises in the first half of the night, but it is offset later and whole-night N3 is not significantly changed in pooled data.
- False
“REM rebound later in the night makes up for what alcohol takes.”
Rebound is inconsistent, and total REM is reduced overall.
- Overstated
“One drink has no effect on your night.”
Sleep-stage changes at low blood alcohol are small, but HRV-based recovery still drops ~9 percentage units and sleeping heart rate rises ~1.4 bpm after about one drink.
- Overstated
“Your body burns off one drink per hour.”
The average is about 7 g/h for a 70-kg adult, half of a 14-g US standard drink, with 3–4-fold variation between people.
Cederbaum 2012Jones 2010National Institute on Alcohol Abuse and Alcoholism 2025
- Overstated
“If you stop drinking 3 hours before bed, alcohol can't affect your sleep.”
A moderate dose 6 h before bed still disturbed sleep with zero breath alcohol at bedtime, and the timing question is largely unstudied.
- False
“Coffee sobers you up or cancels the drink's effect on sleep.”
Caffeine does not speed alcohol clearance, and the two impair sleep in different ways that can mask each other.
- False
“A drink helps a sleeping pill work better.”
Alcohol adds CNS and respiratory depression and raises the risk of complex sleep behaviours; the zolpidem label says not to take it after drinking that evening.
Sanofi-Aventis U.S. 2025National Institute on Alcohol Abuse and Alcoholism 2014US Food and Drug Administration 2019
Pre-sleep arousal and wind-down techniques
Read the review →- Unproven
“4-7-8 breathing will put you to sleep in 60 seconds.”
No trial has tested 4-7-8 breathing for sleep onset in the general population or insomnia. Slow breathing in general has only low-quality evidence for sleep, and no technique reliably produces sleep within a minute.
- Overstated
“Cognitive shuffling is a scientifically proven sleep hack.”
It is a plausible idea from a researcher with a commercial stake, supported so far only by conference abstracts. The nearest tested relative, imagery distraction, helped in one small trial.
- Partly true
“A hot shower right before bed knocks you out.”
Warm water bathing does help onset, but the evidence points to 1–2 hours before bed so that the body has time to shed heat; immediately before bed was not shown to work as well.
- False
“If you can't sleep, stay in bed and keep trying — at least you're resting.”
Trying harder raises arousal, and lying awake in bed conditions the bed as a place of wakefulness. Stimulus control (getting up when unable to sleep) and reducing sleep effort are evidence-based.
- Overstated
“Blue light is the main reason phones in bed keep you awake.”
Large blue-light effects came from hours of bright reading in the lab. Real-world harm from phones in bed seems to come largely from delayed lights-out and lost sleep time; 30 minutes of social media in the lab barely changed arousal.
Chang et al. 2015Combertaldi et al. 2021Hjetland et al. 2025
- False
“Writing worries down before bed just makes you dwell on them.”
Structured writing (a specific to-do list, worries with next steps) reduced pre-sleep arousal or sleep onset in small trials. What you write matters: listing completed tasks did not help.
- Overstated
“Good sleep hygiene and relaxation are enough to fix insomnia.”
Sleep hygiene alone is not recommended for chronic insomnia, and relaxation added little inside CBT-I packages. CBT-I (cognitive work, stimulus control, sleep restriction) is first-line.
- Unproven
“Meditation apps work as well as therapy for insomnia.”
Mindfulness programmes beat attention-matched controls modestly, but have not been shown to beat evidence-based sleep treatments, and most trials used 6–8-week courses, not bedtime app sessions.
Behaviour change: what improves sleep and how to make it stick
Read the review →- Overstated
“Good sleep hygiene is enough to fix insomnia.”
Hygiene education helps a little but less than CBT-I; the AASM suggests not using it as the only treatment, and it adds nothing measurable on top of other components.
- False
“It takes 21 days to form a habit.”
The real-world modelling study found a median of 66 days with a range of 18 to about 254; a meta-analysis found means of 106-154 days.
- False
“Miss one day and the habit is ruined.”
A single missed opportunity did not materially affect habit formation; repeated misses lower the plateau.
- Unproven
“The more sleep rules you follow, the better you will sleep.”
A few components (sleep restriction, stimulus control, cognitive work) carry most of the effect, and brief one- or two-component treatments work.
Furukawa et al. 2024Steinmetz et al. 2024Epstein et al. 2012Buysse et al. 2011
- False
“If good nights line up with a habit, the habit caused them.”
Expectancy, regression to the mean and choosing which nights to do the step can all produce the same pattern; within-person tests need design discipline.
- False
“If you can't sleep, try harder.”
Sleep effort and focused attention on sleep are thought to inhibit sleep and are higher in people with insomnia.
- Overstated
“CBT-I works mainly by making you sleep longer.”
Its largest effects are on insomnia severity, onset, night waking and efficiency; total sleep time changes least.
- False
“An app can't deliver real sleep therapy.”
Fully automated digital CBT-I programs have large RCTs and one received FDA clearance.
Espie et al. 2019Ritterband et al. 2017Vedaa Ø 2020US Food and Drug Administration 2020
- Overstated
“Sleep restriction therapy is just harmful sleep deprivation.”
It causes real but temporary sleepiness in the first weeks and is among the most effective components; it needs screening and care.
Caffeine and sleep
Read the review →- Partly true
“Caffeine doesn't affect me — I can have an espresso and still fall asleep.”
Sensitivity really does differ, partly for genetic reasons, and falling asleep is only delayed by about 9 min on average. Deep sleep, night-time wake and sleep efficiency can still be affected, and people often don't notice.
Gardiner et al. 2023Drake et al. 2013Gardiner et al. 2025Rétey et al. 2007
- False
“Caffeine wears off after a few hours.”
With a half-life of about 4–6 h, roughly a fifth to a third of a coffee drunk 9 h before bed is still in the body at bedtime.
- Overstated
“A fixed noon or 2 pm cutoff works for everyone.”
The safe gap depends on dose (tea little, regular coffee about 9 h, pre-workout about 13 h) and on the person (smoking, oral contraceptives, pregnancy, age).
Gardiner et al. 2023Gardiner et al. 2025Parsons et al. 1978Abernethy et al. 1985Robillard et al. 2015
- Overstated
“Six hours between caffeine and bed is enough.”
This idea comes from Drake 2013, where 400 mg 6 h before bed still cut objective sleep by more than an hour. Pooled data point to about 9 h for a regular coffee and longer for high doses.
- Overstated
“Coffee gives regular drinkers extra energy.”
In habitual users much of the lift reverses overnight withdrawal; net gains in mental alertness are small, though caffeine helps when you're sleep-deprived.
- False
“You can run on caffeine instead of sleep.”
On 5-h nights, caffeine's benefit faded within about 3 days and recovery was slower afterwards.
- Unproven
“Delay your first coffee 90–120 minutes after waking to prevent the afternoon crash.”
A popular podcast claim; we found no controlled trial testing it. The afternoon dip has a circadian basis, and in regular users a crash is more plausibly caffeine wearing off.
- Partly true
“The afternoon slump is caused by eating lunch.”
The dip occurs even without lunch and is tied to the body clock; a high-carbohydrate lunch makes it worse.
- Overstated
“Tea is as bad as coffee for sleep.”
A typical cup of tea has roughly half the caffeine of coffee or less, and the meta-regression found no cutoff time for black tea.
- Overstated
“A CYP1A2 gene test tells you your personal caffeine cutoff.”
The common CYP1A2 variant mainly affects how far the enzyme can be induced (e.g. in smokers). Hormones, drugs, liver function and adenosine-receptor genes also matter.
Bedroom environment
Read the review →- Overstated
“The ideal bedroom temperature is 18 °C / 65 °F (or 16–19 °C).”
Heat clearly harms sleep, but no trial has identified a single best temperature; bedding shifts the comfortable range and the only home-sensor study (older adults) found 20–25 °C best.
Baniassadi et al. 2023Okamoto-Mizuno et al. 2012Chevance et al. 2024
- False
“The colder the bedroom, the better you sleep.”
Very cold rooms alter cardiac autonomic activity during sleep and sleep efficiency also dips at cooler temperatures.
- Unproven
“White noise machines improve sleep.”
Systematic reviews rate the evidence as very low quality, with mixed results.
- Partly true
“You get used to traffic noise, so it stops affecting your sleep.”
Subjective and sleep-stage effects habituate, but heart-rate arousals to noise events do not.
- Overstated
“Any light in the bedroom shuts down your melatonin.”
In the 100-lux study melatonin did not differ, though heart rate and insulin resistance worsened; eyelids filter most light. The concern is real but the mechanism is not simply melatonin.
- Overstated
“Sleeping with your dog ruins your sleep.”
Objective sleep efficiency is slightly lower with a dog on the bed, but differences are small and owners rarely perceive disruption.
- Unproven
“Keeping your phone by the bed harms sleep through its radiation.”
Bedroom phone research points to night-time use, checking and notifications; the main meta-analysis deliberately excluded radiation studies.
Light, the circadian clock and sleep timing
Read the review →- Unproven
“Blue-light-blocking glasses fix your sleep.”
A 2023 Cochrane review found very low certainty evidence for sleep effects, with trials evenly split between benefit and no difference.
- Overstated
“Night Shift or dark mode makes phone use before bed harmless.”
An RCT found no sleep benefit from Night Shift, and changing colour alone did not change melatonin suppression in the lab; brightness and engagement still matter.
- Overstated
“Screens hurt sleep mainly because of their blue light.”
Large lab effects come from hours of bright exposure; an hour of typical use had minimal light effects, while interactive use and time displacement better explain real-world links.
Chang et al. 2015Heath et al. 2014Brosnan et al. 2024Reichenberger et al. 2024
- False
“Screens before bed make no difference to sleep.”
Associations with shorter and later sleep are consistent, and interactive use predicts later sleep onset night by night.
- Partly true
“Switching to warm bulbs is enough to protect evening melatonin.”
Warm light carries less melanopic light per lux, but bright warm light still suppresses melatonin; dimming does more.
- Unproven
“You must be asleep by 10 pm; hours before midnight count double.”
Bedtime-health data are observational and U-shaped; regularity and alignment with your clock predict outcomes better than clock time.
Nikbakhtian et al. 2021Windred et al. 2024Murray et al. 2017
- False
“Night owls can simply choose to become morning people.”
Chronotype is partly genetic, peaks late around age 19, and late types can be more sensitive to evening light; light and schedules shift it only within limits.
- Partly true
“You need a specific number of minutes of sunlight within minutes of waking.”
Morning light does advance the clock, but no trial sets a precise minimum dose or window for healthy people; consensus asks for at least 250 lux melanopic EDI during the day.
- Overstated
“Any tiny light in the bedroom ruins sleep.”
Harm has been shown with 5-10 lux and 100 lux all night; tiny indicator lights are untested, though darker is better.
Measuring sleep, and where an app's job ends
Read the review →- Overstated
“Everyone needs exactly 8 hours of sleep.”
Consensus sets a floor of 7 or more hours for adults (7–9 h per the NSF), and need varies between individuals.
- False
“Taking more than 15 minutes to fall asleep is a sign of insomnia.”
Up to 30 minutes is within good-quality norms. Research criteria for insomnia start at 31 minutes or more on at least 3 nights a week over months.
- Overstated
“Waking up during the night means you slept badly.”
One awakening and up to 20 minutes awake are within good-quality norms for adults.
- Overstated
“Your tracker knows how well you slept better than you do.”
Devices detect sleep well but miss much of the time awake and are inconsistent on stages. Self-report is the clinical standard for sleep quality.
- False
“Self-reported sleep is unreliable, so it doesn't count.”
Sleep diaries are the recommended clinical and research tool. Misperception is a recognised feature of insomnia, not a reason to dismiss the complaint.
- Overstated
“Good sleep hygiene cures insomnia.”
The AASM suggests not using sleep hygiene as a stand-alone therapy for chronic insomnia. CBT-I is first-line.
- Partly true
“Snoring is just harmless noise.”
Simple snorers often have normal sleepiness scores, but loud snoring with witnessed pauses, gasping or daytime sleepiness can signal obstructive sleep apnoea, which needs a sleep test.
- Unproven
“A higher sleep score means you are healthier.”
No nightly composite sleep score has been validated against health outcomes, and negative score feedback can itself worsen how the day feels.
- False
“Adding 'not a medical device' keeps an app out of medical-device regulation.”
Regulators look at the claims, labelling and functions (intended purpose), not the disclaimer.
U.S. Food and Drug Administration 2026Medical Device Coordination Group (European Commission) 2025
Sleep pressure, naps, exercise and daytime behaviour
Read the review →- Overstated
“Any nap will ruin your night's sleep.”
Short early-afternoon naps rarely change night sleep in healthy people, and in older adults naps often add to 24-h sleep. Late, long or frequent naps, and naps in people with insomnia, are the exceptions.
Campbell et al. 2005Boukhris et al. 2026Werth et al. 1996Mograss et al. 2022
- Unproven
“3 p.m. is the scientifically established cutoff for naps.”
It is a round-number public-health convention. Nap timing relative to bedtime matters more, and the classic brief-nap study itself scheduled naps at 15:00.
- False
“You should never exercise in the evening.”
Meta-analyses find evening exercise does not harm sleep on average; only very hard sessions ending close to bedtime may delay sleep.
- Overstated
“Napping shortens your life.”
The association is observational and likely reflects underlying illness or poor night sleep rather than harm from naps themselves.
- False
“If you sleep badly, spend more time in bed to catch up.”
For insomnia, extra time in bed dilutes sleep drive; the effective treatment restricts time in bed to actual sleep.
- Overstated
“Everyone needs exactly 8 hours.”
The consensus floor is 7 or more hours for adults aged 18-60; individual need varies and maximal sleep capacity declines with age.
- Partly true
“The afternoon slump is just caused by lunch.”
The dip appears even without lunch because it is partly circadian; a heavy high-carbohydrate lunch and short sleep can deepen it.
- Unproven
“Never eat within 3 hours of bedtime.”
Links between late meals and poor sleep are observational and inconsistent; one small trial found a carbohydrate meal 4 h before bed shortened sleep onset.
Sleep regularity and schedule consistency
Read the review →- False
“You can fully catch up on a week of short sleep by sleeping in at the weekend.”
In a randomised lab study, weekend recovery sleep added only about 1.1 h and did not prevent metabolic harm once short sleep resumed; one long night does not restore attention after a short-sleep week. Moderate catch-up is better than none, but it is a partial fix.
- Overstated
“Any weekend lie-in is bad for you.”
For people who are short on sleep during the week, a moderate lie-in (up to about 2 h) is associated with lower mortality, depression and BMI than none at all, and the NSF panel judged catch-up sleep may be beneficial. The ideal is enough sleep every night at consistent times.
Åkerstedt et al. 2019Zhou et al. 2025Im et al. 2017Sletten et al. 2023
- Overstated
“Sleeping in 2 hours on Saturday gives you 2 hours of jet lag on Monday.”
Measured clock delays over a sleep-in weekend are about 30-45 min, much smaller than the extra time in bed. The jet lag analogy fits the direction of the effect, not its size.
- Overstated
“Sleep regularity matters more than sleep duration.”
One large UK Biobank study found regularity a stronger statistical predictor of mortality, and other studies find the two similar in size. Both matter, they overlap, and no trial shows that fixing regularity beats fixing duration.
- Unproven
“Irregular sleep has been proven to cause heart disease and early death.”
The links are consistent but observational; illness, depression and life circumstances can cause irregular sleep. Randomised trials of schedule stabilisation are still needed.
- False
“Good sleep hygiene tips, including a regular schedule, are enough to treat insomnia.”
The AASM advises against sleep hygiene as a stand-alone insomnia treatment; a fixed rise time works as part of stimulus control, sleep restriction or full CBT-I.
Sleep supplements
Read the review →- Overstated
“Melatonin is a natural sleeping pill that knocks you out.”
On average it shortens sleep onset by about 7 minutes and adds about 8 minutes of sleep. Its clearer role is as a timing signal for a late body clock.
Ferracioli-Oda et al. 2013Burgess et al. 2010Sateia et al. 2017
- Overstated
“If a low dose of melatonin doesn't work, take a bigger one.”
Clock shifts are no larger at 3 mg than at 0.5 mg, and high doses can spill into the wrong part of the phase-response curve. Side effects rise at 10 mg or more, and the sleep-promoting effect plateaus around 4 mg.
Burgess et al. 2010Lewy et al. 2002Menczel Schrire et al. 2022Cruz-Sanabria et al. 2024
- False
“The dose on a melatonin label is what you get.”
Tested products ranged from −83% to +478% of the label, and only 3 of 25 US gummies were within ±10%.
- False
“Melatonin is natural, so it's safe to give children.”
Paediatric ingestions rose 530% in a decade, with ICU admissions. Paediatric dosing and long-term effects are uncertain, and the AASM advises asking a paediatric clinician first.
Lelak et al. 2022Freeman et al. 2024American Academy of Sleep Medicine 2022
- Overstated
“Most poor sleepers are magnesium-deficient, and magnesium fixes insomnia.”
RCTs are few, small and inconsistent, with low to very low certainty. Reviews do not support routine use for insomnia.
- Unproven
“Magnesium glycinate is the best form of magnesium for sleep.”
There is one RCT with a small effect (d = 0.2) and no head-to-head comparison between forms.
- Unproven
“Magnesium L-threonate boosts deep sleep.”
The positive trial used consumer-ring sleep-stage scores and was industry-linked. A second funded trial found no ring-measured sleep difference.
- Overstated
“Valerian is a proven herbal sleeping pill.”
Meta-analysis found no effect on sleep latency or quantified sleep quality, and the AASM recommends against it.
Fernández-San-Martín et al. 2010Leach et al. 2015Sateia et al. 2017
- Unproven
“CBD helps you sleep.”
There is insufficient evidence in any sleep disorder, and a pilot insomnia RCT found it similar to placebo on most outcomes.
- Overstated
“Tart cherry juice is a natural melatonin source that fixes sleep.”
The evidence is a handful of small, heterogeneous studies.
- False
“Over-the-counter sleep aids are harmless because you can buy them freely.”
Tolerance to diphenhydramine's sedation develops within about 3–4 days. First-generation antihistamines are on the Beers list for older adults because they are anticholinergic, and guidelines recommend against them for insomnia.
Richardson et al. 20022023 American Geriatrics Society Beers Criteria Update Expert Panel 2023Sateia et al. 2017Riemann et al. 2023