⚡Key Takeaways

10 min read

The strongest positive trial (120 psychiatric outpatients with insomnia) found large improvements with an 8 kg chain blanket — but people can feel which blanket is heavy, so it could not be blinded
In autistic children, a careful trial found no gain in objectively measured sleep, although children and parents preferred the weighted blanket
Pooled reviews point to a small-to-moderate reduction in anxiety; effects on sleep are less consistent
The “10% of body weight” rule is a manufacturer convention, not a tested dose — the best trial used 6–8 kg
Never for babies, and not for anyone who cannot push it off unaided

Weighted blankets went from occupational-therapy tool to mass-market bedding in a few years, carried by a simple and appealing idea: firm, even pressure calms the nervous system, the way a hug or being tucked in tightly does. Some people find them genuinely helpful. Others find them hot, heavy and claustrophobic.

The research sits between the marketing and the scepticism. There is one striking positive trial in adults with psychiatric conditions, one well-run negative trial in autistic children, a handful of small studies, and a basic problem that runs through all of them. This guide goes through what was found, what weight to choose, the heat question, and the safety rules that are not optional.

What a weighted blanket is

Most weighted blankets are quilted into small pockets filled with glass beads or plastic pellets, so the weight stays spread evenly. Others use metal chains sewn into the fabric (the design used in the best-known trial), or thick knitted yarn with no filler at all. Adult blankets usually weigh between about 4 and 13 kg (roughly 9 to 30 lb).

The proposed mechanism is “deep pressure stimulation”: steady pressure on the body is thought to shift the nervous system away from a fight-or-flight state. It is plausible, and it is the basis of long-standing occupational-therapy practice, but it has not been directly demonstrated during sleep. One small laboratory study in 26 healthy young adults found a larger rise in saliva melatonin in the hour before lights-out under a blanket weighing about 12% of body weight than under a light one, with no difference in sleepiness or sleep duration. A co-author worked for a weighted-blanket company.

The trials

Insomnia in adults with psychiatric conditions: a large effect

The trial most often quoted was run at the Karolinska Institute in Sweden. It randomised 120 outpatients with major depression, bipolar disorder, generalised anxiety disorder or ADHD, all with insomnia, to four weeks with a metal-chain blanket of 8 kg (6 kg for those who found 8 too heavy) or a light plastic-chain blanket of about 1.5 kg. On the Insomnia Severity Index, 59% of the weighted-blanket group halved their score, against 5% of the control group, and 42% versus 4% reached remission. Daytime fatigue, depression and anxiety scores also fell, and the benefit held over a 12-month open follow-up.

That is a large effect, and the study was publicly funded with no declared industry ties. The catch is that the main outcome was a questionnaire, and participants could obviously tell whether they had been given an 8 kg blanket or a light one. When people know they are in the active group and rate their own symptoms, expectation inflates the result. Tellingly, the wrist activity monitors the participants wore showed no significant change in total sleep time or time awake during the night; the better sleep maintenance came from people’s own reports.

Autistic children: no measurable gain

A UK trial took the opposite approach and measured sleep objectively. Seventy-three children aged 5 to 16 with autism and severe sleep problems used a weighted blanket and an otherwise identical normal blanket for two weeks each, in random order, with sleep recorded by actigraphy. The weighted blanket did not increase total sleep time, shorten the time to fall asleep, or reduce night waking, and there was no difference in behaviour. Children and parents nonetheless preferred it.

Children with ADHD: small effects

A Swedish crossover trial in 94 children with ADHD and sleep problems found statistically significant but small improvements in total sleep time and time awake after falling asleep (a few minutes), and no change in how long it took to fall asleep.

What the reviews conclude

Two 2024 meta-analyses of the psychiatric studies came to similar conclusions: a small-to-moderate reduction in anxiety, an insomnia effect that depends on which studies are included, few serious side effects, and a literature that is small, varied and often at high risk of bias because participants could not be blinded. An earlier systematic review put it plainly: weighted blankets may help with anxiety, but there was not enough evidence that they help with insomnia.

📊Why the blinding problem matters

You cannot hide an 8 kg blanket. Every trial that relies on people rating their own sleep or anxiety is exposed to expectation effects, and the trial that measured sleep objectively (in autistic children) found nothing. That does not mean weighted blankets do nothing — feeling calmer and preferring your bed are real outcomes — but it means the size of the benefit in the headline trial is probably an overestimate.

Who might reasonably try one

Adults whose insomnia is tangled up with anxiety or low mood, as an add-on to treatment, not a replacement. This is the group with the most encouraging data.
People who already know they like firm pressure— being tucked in tightly, a heavy duvet, a pet on the bed.
Older children and teenagers who like the feel and can remove the blanket easily themselves, with parental supervision and realistic expectations.

It is not a treatment for chronic insomnia on its own. The first-line treatment for that is cognitive behavioural therapy for insomnia (CBT-I), covered in our sleep environment guide. If you are weighing supplements too, see our sleep supplement protocol.

What weight to choose

You will see “about 10% of your body weight” everywhere. It is a manufacturer convention, not a number that has been tested against alternatives. What the research actually used:

The positive adult trial used 8 kg, and let people who found that too heavy drop to 6 kg (ten of them did).
The melatonin study used about 12% of body weight.

A practical approach: start at or slightly below 10% of your body weight, and choose lighter rather than heavier if you are between sizes. The blanket should feel like firm pressure, not like being pinned. The non-negotiable test is that you can push it off and get out from under it easily, half asleep, without help. Buy a size that covers you rather than the whole bed — a weighted blanket that drapes over the edges tends to slide off during the night.

Heat

Weight and warmth tend to come together. Dense bead-filled blankets with polyester covers trap a lot of heat, and heat is one of the most reliable ways to make sleep lighter and more broken. If you sleep hot:

Choose an open knitted design or a cotton, linen or lyocell cover rather than minky or fleece.
Glass beads pack more weight into less volume than plastic pellets, so the blanket is thinner for the same weight.
Use it instead of a duvet, not on top of one, and keep the room cool.
Try it for the first part of the evening or while falling asleep, and push it off later if you wake warm.

“Cooling” weighted blankets vary a lot. The fabric and construction tell you more than the label.

Trying one: a simple four-week test

Because expectation plays a large part in how people rate weighted blankets, it helps to judge one a little more carefully than “it feels nice”:

Buy where returns are allowed. Many sellers offer a trial period, and weighted blankets are expensive to get wrong.
Write down a baseline for a week first: roughly how long it takes you to fall asleep, how many times you wake, and how rested you feel, on a 1–10 scale.
Use it nightly for four weeks, the length of the main adult trial, and keep the same notes.
Compare the weeks, not single nights. If nothing has changed except that you feel warmer, return it or keep it as a sofa blanket for winding down.

Safety

⚠️Never for infants

The American Academy of Pediatrics advises that weighted blankets, weighted sleep sacks, weighted swaddles and other weighted objects should not be used on or near a sleeping infant. Babies cannot move a heavy covering off their face or roll back if they get stuck, and overheating is itself a risk factor for sudden infant death.

⚠️Children, and anyone who can't remove it

A weighted blanket is only safe for someone who can lift it off themselves. Many makers do not sell them for children under about 3 years or below a minimum body weight. Never use one to hold a child still, wrap a child in one, or cover a child’s head: in 2008 a Quebec coroner found that a 9-year-old autistic boy suffocated after being rolled tightly in a weighted blanket at school. The same logic applies to adults who are frail, have limited strength or mobility, have dementia, or are heavily sedated by alcohol or medication.

⚠️Talk to a clinician first if you have

Obstructive sleep apnoea or any breathing or lung condition (weight on the chest and abdomen can make breathing harder); heart failure or circulation problems; low blood pressure; a condition affecting skin sensation or temperature regulation; or significant claustrophobia. Pregnancy is not well studied either; ask your midwife or doctor. If you snore loudly or wake gasping, get checked for sleep apnoea before adding weight to your bed.

Two more practical points: check seams regularly on bead-filled blankets, because loose beads are a choking hazard for small children and pets; and follow the washing instructions, since a wet 10 kg blanket can damage a domestic washing machine. Many people use a removable cover for this reason.

Frequently Asked Questions

Do weighted blankets actually work for sleep?

For some people, probably. The best adult trial found large improvements in insomnia in people with depression, anxiety, bipolar disorder or ADHD, but it relied on self-report without blinding. The trial that measured sleep objectively in autistic children found no effect. Expect a modest benefit if any, and judge it on how you sleep over a few weeks.

How heavy should my weighted blanket be?

Around 10% of body weight is the usual starting point, but it is a convention. The main adult trial used 6–8 kg. Pick the lighter option if you are unsure, and make sure you can push it off easily.

Can I use a weighted blanket every night?

In the Swedish trial, 112 of the 120 participants carried on into a year-long open follow-up, and the authors reported no side effects beyond one person who found the blanket made them anxious. If it helps and you are not overheating, nightly use is reasonable for healthy adults.

Are weighted blankets safe for children?

Never for babies. For older children, only if the child can easily remove it without help, it is sized and weighted for them, and an adult supervises. The trial evidence in children is weak: no objective sleep benefit in autism and small effects in ADHD.

Can I use one if I have sleep apnoea?

Check with your sleep physician first. Extra weight on the chest and abdomen is a reasonable concern in anyone whose breathing is already compromised during sleep, and weighted blankets have not been tested in people with sleep apnoea.

The Bottom Line

Weighted blankets are low-risk for healthy adults and may help with anxiety and anxious insomnia, but the most impressive result comes from a trial whose participants knew which blanket they had, and the most rigorous objective test found nothing. Buy one because you like the feel, choose the lighter end of the weight range and a breathable construction, and keep them away from infants and anyone who cannot remove them. For more on the rest of the bedroom, see our guides to choosing a mattress and the sleep environment, the weighted blanket reference, and the sleep learning track.

Compare weighted blankets on the Formulate sleep shelf →

Weighted Blankets: What the Trials Show for Sleep and Anxiety