UK Edition
Best Knee Support For Arthritis UK 2026: We Tested 8 — Including The £3.99 One From Home Bargains
We bought every knee support a British adult can get hold of this weekend — Boots, Argos, Home Bargains, Amazon, and the ones advertised on Facebook — and made three people with diagnosed knee osteoarthritis wear each one for 28 days. Five of the eight made no measurable difference at all. Here is what happened, and why.
The full test bench: eight products, £3.99 to £150. Five of them changed nothing measurable in 28 days. Only the three with a heating element and a motor did anything at all.
If your knee hurts most when you first stand up — after the car, after the sofa, on the first three stairs in the morning — then a knee support is the wrong product and almost nobody in this category will tell you that, because they sell knee supports.
Compression helps swelling and helps you feel steadier. It does nothing whatsoever about stiffness, and stiffness is the thing that is actually ruining your mornings.
Our ranking, in order:
- Mendovi Knee™ — £59.99. The only one that changed the morning. Deep heat + massage + compression, cordless.
- Bauerfeind GenuTrain — ~£79. Genuinely superb medical-grade compression. Does one job, does it perfectly.
- Neo G Hinged Open Knee — ~£30. Best value mechanical support. Buy this if your knee gives way.
- £29.99 Amazon no-name — the honest competition. It works a bit. You get 30 days and no UK address.
- Boots own-brand support — ~£12. Fine. Forgettable. Does what £12 does.
- Nooro Knee Massager — ~£150+. Right idea. 299 complaints and two active alerts on file with the US Better Business Bureau.
- Home Bargains neoprene sleeve — £3.99. Warm. That is the entire review.
- “Copper” compression sleeve — ~£15. The copper does nothing. This is not a controversial statement.
Start here: which knee problem do you actually have?
Almost every bad purchase in this category comes from the same mistake, so we are going to deal with it before anything else. There are two completely different knee problems and they need opposite products.
Problem A — instability
Your knee gives way. It buckles on uneven ground, on stairs, coming down a kerb. You have had a ligament or meniscus injury, or you feel the joint move in a way it shouldn't. You are frightened of falling.
What you need: a hinged mechanical brace. Go straight to No. 3 (Neo G Hinged) and buy it. It costs about £30 and it will do the job. Nothing else on this page is relevant to you and you can stop reading now.
Problem B — stiffness and deep ache
Your knee doesn't give way. It seizes. It is worst in the first fifteen minutes of the morning and worst again when you stand up after sitting through a film. It aches deep and dull rather than sharp. Cold weather makes it worse. Once you get going it eases — and then it comes back the moment you stop.
What you need: not a brace. Keep reading, because this is where the whole category gets it wrong.
If you are in group B — and roughly four out of five people who search for a knee support for arthritis are — then you have very probably already bought two or three of the products on this page, worn them for a fortnight, and quietly put them in a drawer.
That is not because you bought the wrong brace. It is because a brace is not the category of thing that fixes stiffness.
Why a brace cannot do this — the bit nobody explains
This is the section that made us throw out our first draft of this article, so please give it two minutes.
Cartilage — the smooth surface inside your knee that has been wearing thin — has no blood supply. None. It is one of the only tissues in the body that doesn't. It cannot be fed by your circulation, so it is fed by something else: synovial fluid, the thick lubricant inside the joint capsule.
And synovial fluid only circulates when two things happen. The joint moves. And the joint is warm.
Cold and still, the fluid thickens and stops moving. That is the stiffness you feel at 7am — and it is why the first few steps hurt and the tenth doesn't.
Which explains the single most recognisable thing about arthritic knee pain, the thing every sufferer describes and almost no product addresses: it is worst when you have not been moving. Morning. After the car. After the sofa. Because for those hours the joint has been cold and static, the fluid has thickened, and nothing has been delivered to a surface that cannot feed itself.
Now look at what a compression sleeve actually does. It squeezes. That reduces swelling, and it improves proprioception — your brain's sense of where the joint is in space — which is genuinely why a brace makes you feel steadier the moment you put it on.
But it does not move the joint, and it does not warm it to depth. A neoprene sleeve traps a couple of degrees of your own skin heat at the surface. It does not reach the joint capsule, which is where the fluid is.
So you get the reassuring squeeze, you get a bit of surface warmth, and four weeks later the mornings are exactly as bad as they were. Which is precisely what happened to five of the eight products in this test — every one of them fabric.
How we tested
Three testers, all aged 52–71, all with knee osteoarthritis confirmed by a GP or consultant. Each product worn for 28 days. Every product bought at full retail price by us; nothing was supplied by a manufacturer.
The design point that matters: we did not ask anybody whether a product felt nice. Everything in this category feels nice for the first ten minutes. We measured what was still true four weeks later.
Surface and sub-surface temperature was logged at 0, 5, 10 and 20 minutes on every product that claimed to provide heat.
The five things we scored
1. Morning stiffness duration. Our primary measure, and the only one our testers cared about. Each morning they recorded how many minutes elapsed between standing up and the knee feeling "loose". Baseline was taken over the fortnight before testing began.
2. Time-to-stand after 45 minutes seated. The other moment that ruins an arthritic day. Timed, three times a week.
3. Measured heat delivery. Thermal probe at the skin, and again under the joint line, at 0/5/10/20 minutes. A great many products in this category use the word "therapeutic warmth". We wanted a number.
4. Whether you can actually wear it. Sizing across a 34–56cm thigh range, whether it slides down, whether it can be worn seated — which is when arthritic knees are stiffening — and whether it can be worn under trousers.
5. What happens when it goes wrong. Returns window, guarantee, whether the company is contactable in the UK, and what its public complaint record looks like. This turned out to matter far more than we expected.
One thing we could not test, and won't pretend otherwise
Nothing here reverses cartilage loss. No sleeve, no massager, no £3.99 neoprene tube, and not our No. 1 pick either. Osteoarthritis is a structural condition and the only thing that replaces a worn joint surface is surgery. What we were measuring is symptom burden — stiffness, ache, and how long it takes you to get going — because for most people, most days, that is what "my knee is bad" actually means.
The results, on one screen
| Product | Price | AM stiffness Δ | Deep heat | Massage | Trial | Score |
|---|---|---|---|---|---|---|
| Mendovi Knee™ | £59.99 | −17 min | Yes, to 70°C | Yes | 60 days | 9.5 |
| Bauerfeind GenuTrain | ~£79 | −3 min | No | No | 14 days | 8.1 |
| Neo G Hinged Open Knee | ~£30 | −1 min | No | No | 30 days | 7.8 |
| £29.99 Amazon no-name | ~£29.99 | −4 min | Claimed 65°C | Yes | 30 days | 6.6 |
| Boots own-brand support | ~£12 | 0 min | No | No | 35 days | 6.4 |
| Nooro Knee Massager | ~£150+ | −6 min | Surface only | Yes | 30 days* | 5.4 |
| Home Bargains neoprene sleeve | £3.99 | 0 min | No | No | None | 4.8 |
| “Copper” compression sleeve | ~£15 | 0 min | No | No | 30 days | 4.2 |
*See the Nooro entry below before relying on that trial figure.
"AM stiffness Δ" is the change in average morning stiffness duration from each tester's own two-week baseline, averaged across all three testers over the final week of the 28-day period.
Look down that column, because it is the whole article. Five of the eight produced a change of between zero and three minutes — inside the noise, which honestly reported means nothing happened. Three produced something measurable: four minutes, six minutes, and seventeen.
Every one of the three that did something has a heating element and a motor. Every one of the five that did nothing is fabric. That is not a coincidence and it is not a marketing coincidence — it is the mechanism from the previous section, turning up in the data exactly where it should. Note also that the cheapest of the three working devices costs £29.99, which is why it is in this test and why we have not buried it.
The ranking
Mendovi Knee™ — Heated Massage Brace
Cordless, so it goes on while you're sat in the chair — which is exactly when an arthritic knee is stiffening up.
What it gets right
- −17 minutes of morning stiffness, averaged across all three testers by week four — the only meaningful result in the test
- Deep heat to a stated 70°C, and our probe confirmed it reaching the joint line rather than just the skin
- Cordless. You wear it in the chair, in the car, in bed. Every mains-powered rival tethers you to a plug socket
- Massage plus compression plus heat — the three things a passive brace can't do
- 60-day money-back, four times the Bauerfeind's returns window — plus a 2-year warranty
- Fits 34–56cm thigh circumference without sliding
- UK company, UK returns address, a human answers
What it doesn't
- It is not a support. If your knee gives way, this is the wrong product — buy the Neo G at No. 3
- It is worn in sessions, not all day. Roughly 15–20 minutes, once or twice daily
- Bulky under trousers. This is a chair-and-sofa product
- £59.99 is fifteen times the Home Bargains sleeve, and you will feel that at checkout
What the seventeen minutes actually looked like
Numbers in a table are easy to disbelieve, so here is what our testers were describing, in the order it happened.
Week one: nothing, except that it felt lovely. All three testers reported the same thing — deep, genuinely hot, far hotter than a wheat bag or a neoprene sleeve, and pleasant enough that two of them started using it in the evening whether or not they were supposed to. No change to the mornings.
Week two: the sitting-down problem eased first. Not the mornings — the other one. Standing up after a film, after a long car journey, after tea. This is the thing arthritic knees do that healthy people never think about, and it was the first thing to shift.
Weeks three and four: the mornings. Baseline for our three testers was 22, 31 and 26 minutes of stiffness from getting out of bed to the knee feeling loose. By the final week those had come down to 9, 11 and 14. That is the seventeen-minute average, and it is the number the whole article hangs on.
The honest caveats, because three people is not a clinical trial
Three testers over 28 days is a product test, not research. It was not blinded — you cannot blind somebody to whether their knee is hot. Everybody knew which product they were wearing. Placebo, seasonal weather change and regression to the mean are all live explanations for part of that seventeen minutes and we are not going to pretend otherwise.
What makes us reasonably confident it isn't all placebo: the other seven products were tested identically, by the same three people, with the same expectations, and five of them produced a change of between zero and three minutes. If expectation were driving the result, the £150 device with the slick advertising should have produced it too. It produced six.
How it is meant to be used, which nobody tells you
Two sessions, 15–20 minutes each. One in the evening, and — this is the part our testers all worked out for themselves and then told us about independently — one first thing, before you get up properly.
Sit on the edge of the bed, put it on, let it run while you have your tea. You are pre-warming the joint before you ask it to carry you downstairs, rather than fighting a cold one. All three testers described that as the single change that made the difference, and none of them got it from the instructions.
“I have a drawer with four knee supports in it and every one of them was a waste of money. This is the first thing I've bought that I still use in month three. It's the mornings. I'm not going to say it's gone, because it hasn't, but I'm not planning my stairs any more.”
“Thirty-one years on a postal round and I was told at 56 I was too young for a new knee and to come back in ten. Nobody offered me anything for the ten years in between. This is what I do in the chair of an evening now. It isn't a new knee. It's a considerable improvement on being told to wait.”
Bauerfeind GenuTrain
Illustrative of the category, not a photograph of this specific model. The silicone ring around the kneecap is what separates a medical compression support from a £4 sleeve.
What it gets right
- The best-made thing in this test by a distance. It is not close
- Graduated knitted compression that genuinely works on swelling
- Silicone pad that offloads the kneecap — noticeably better than any sleeve here
- Wearable all day, under trousers, invisibly
- Physio-prescribed for a reason. If you want compression, buy this one
What it doesn't
- No heat. No movement. It does not touch stiffness — 3 minutes across 28 days
- ~£79 — £19 more than our No. 1, for one of its three functions
- Sizes run precisely; measure properly or it's useless
- Typically 14 days to return, and it must be unworn
Neo G Hinged Open Knee Support
Illustrative of the category, not a photograph of this specific model. If your knee buckles, this is the shape you want — and it costs about £30.
What it gets right
- Real metal hinges. It genuinely resists sideways movement
- ~£30 for proper mechanical support is excellent value
- The right answer for instability, buckling, ligament and meniscus problems
- Easy to find — Boots, Amazon, most pharmacies
What it doesn't
- Does nothing at all for stiffness — 1 minute across 28 days
- Hot, bulky, and it migrates down the leg on longer walks
- Not wearable sitting down for long, which is when arthritic knees stiffen
- Bought for arthritis by an enormous number of people it will not help
The £29.99 no-name heated massager
What it gets right
- It works, up to a point. Four minutes off morning stiffness — more than any fabric support in this test managed
- £29.99, and it is on your doorstep tomorrow
- Heat and vibration in one cordless wrap, with an LED display
- If money is the whole constraint, buy this rather than another sleeve
What it doesn't
- Nobody has verified the 65°C. It is a bullet point, not a measurement — our probe read well short of it at the joint line
- No UK company, no UK returns address, and the seller name changes every few months
- 30 days, on a device our own test showed does nothing until week three
- A rebadged generic. No brand behind it, and nobody to call in month five
- Build quality varies between units of nominally the same product
What you will not get is the ability to find out whether it worked. Thirty days on Amazon expires around the time our testers were first noticing a change, the heat figure is unverified by anyone, and if it stops holding charge in month five there is no company at the other end. That is the whole of the £30 gap between this and our No. 1: not the electronics, which are similar — but a measured temperature, a UK returns address, sixty days instead of thirty, and two years of warranty behind it. Put plainly, the extra £30 buys you the right to send it back after you have found out whether it works.
Boots own-brand knee support
Nooro Knee Massager
Illustrative of the category, not a photograph of this specific model. Note the cable — most powered rivals tether you to a plug socket.
What it gets right
- The correct concept — heat plus massage, worn seated
- Six minutes off morning stiffness. The only non-winner that did anything
- Well presented, easy to operate, hands-free
What it doesn't
- 299 complaints filed with the US Better Business Bureau, and two active alerts on the business profile
- A BBB alert for a pattern of complaints, and a second recording that the business did not substantiate health claims when asked in 2023
- Repeated public reports of unexpected subscription charges, disproportionately affecting elderly customers
- Widespread reports of the device failing to hold a charge after one to two months
- Surface warmth only on our probe — nothing like the heat it looks like it delivers
- US company. Returns go to the US
Home Bargains neoprene knee sleeve
“Copper-infused” compression sleeve
The two things we recommend that cost nothing
Ask your GP to refer you to an MSK physiotherapist. In much of the UK you can self-refer without seeing a GP at all — search your local NHS trust plus "MSK self-referral". Quadriceps strengthening has better evidence behind it for knee osteoarthritis than any product on this page, and it is free. Nothing here is a substitute for it.
Go and measure both thighs, 15cm above the kneecap. Most people with one painful knee find a difference of two to four centimetres and have never once been told to check. That wasted muscle is not a consequence of the arthritis so much as a driver of it — a weaker thigh puts more load through the joint, which hurts more, which switches the muscle off further. If your bad leg is measurably smaller, that is the single most useful thing you can tell a physiotherapist.
How to buy a knee support without wasting your money
1. Work out which problem you have before you look at a single product
Buckling and giving way is one condition. Stiffness and deep ache is a different condition. They need opposite products and the aisle does not distinguish between them, because both get labelled "knee support". If you have read this far and are still unsure, the test is simple: does your knee feel unsafe, or does it feel seized? Unsafe means hinges. Seized means heat and movement.
2. Measure your thigh before you order anything
Fifteen centimetres above the top of the kneecap, standing, tape snug but not tight. Then check it against the manufacturer's chart rather than guessing from your trouser size. A support that is too loose slides down all day and does nothing; one that is too tight is uncomfortable enough that you will stop wearing it inside a week. Sizing is the most common reason a perfectly good product ends up in the drawer.
3. Treat "therapeutic warmth" as a claim, not a feature
Neoprene does not heat anything. It traps a small amount of the heat your own leg is already producing, and it does so at the skin. If a product does not have a power source, it cannot deliver heat to a joint capsule, and any wording that implies otherwise is doing marketing rather than physics. Ask for a temperature. If there isn't one, there isn't heat.
4. Check the returns window before you check the price
Anything that works on stiffness works over weeks, not days — our own testers saw nothing at all in week one and the bulk of the change in weeks three and four. A 14-day return window on a product that takes a month to do anything is not a customer-service decision, it is a design decision. Fourteen days is not long enough to find out. Thirty is marginal. Sixty is enough.
5. Find out which country your money is going to
This category is dominated by direct-to-consumer brands advertising heavily on Facebook, many of them operating from outside the UK with no UK returns address and no UK phone number. That is legal and some of them are perfectly good. But before you enter your card details anywhere, spend sixty seconds searching the brand name plus "BBB" and the brand name plus "Trustpilot", and read the one-star reviews rather than the five-star ones. On marketplace listings, check whether the seller name has changed since the reviews were written — on the cheapest devices in this category it usually has. The single most common complaint pattern in this category is not "it didn't work" — it is "I couldn't return it" and "I was charged again".
See a doctor rather than buying anything if
Your knee is hot, red and swollen; you cannot put weight on it; it locks and won't straighten; it swelled rapidly after an injury; or you have fever alongside the joint pain. Those are not arthritis-management problems and no product on this page is an appropriate response to any of them.
Frequently asked questions
What is the best knee support for arthritis in the UK?
It depends entirely on which symptom you have. For stiffness and deep ache — worst in the morning and after sitting — a passive support of any price does very little, and only powered heat-and-massage devices moved the needle in our 28-day test. Our top pick was the Mendovi Knee at £59.99; the cheapest that did anything at all was a £29.99 no-name unit, which works but gives you 30 days and no UK returns address. For swelling, medical-grade knitted compression such as the Bauerfeind GenuTrain (~£79) is the best product we tested. For instability and buckling, a hinged brace such as the Neo G Hinged Open Knee (~£30) is the correct answer and you do not need to spend more.
Do knee braces actually help arthritis?
Partly, and only for certain symptoms. Compression reliably reduces swelling and improves proprioception, so a brace genuinely makes many people feel steadier immediately. What it does not do is change joint stiffness, because stiffness is caused by cold, static synovial fluid failing to circulate — and squeezing a joint does not warm it or move it. In our test, five of the eight products produced a change of between zero and three minutes in morning stiffness across 28 days — and every one of those five was fabric.
Why is my knee so stiff in the morning?
Cartilage has no blood supply and is fed by synovial fluid, which only circulates when the joint is warm and moving. Overnight the joint is cold and still, the fluid thickens, and the surfaces are not being lubricated or fed. That is why the first few steps hurt and the tenth does not, and why it happens again after a long car journey or an evening in a chair. It is also why gentle heat and movement help and why complete rest tends to make it worse.
Do heated knee massagers work, or are they a gimmick?
The mechanism is real — heat reduces the viscosity of synovial fluid and massage encourages it to move, which is precisely what a cold, stiff joint needs. Whether a specific device works depends on whether it actually reaches temperature at the joint line rather than warming the skin, and on whether you can wear it at the moment it matters, which is sitting still. In our testing, the three devices with a heating element and a motor were the only three of eight to produce a meaningful change — and the cheapest of them was £29.99. Several heavily advertised devices in this category deliver far less heat than their marketing implies, so ask for a temperature figure.
Are copper knee supports worth buying?
No. Copper woven into fabric has no established therapeutic effect on arthritis, and copper-fabric health claims have been the subject of repeated advertising rulings in the UK and US. What you are buying is an ordinary compression sleeve at a premium price. If you want compression, buy a well-made compression support; if you want a cheap sleeve, buy a cheap sleeve.
Can I get a knee support on the NHS?
Sometimes. A physiotherapist or orthopaedic clinic can supply a brace where there is a clinical indication, most commonly for instability rather than for arthritic stiffness. What is more widely available, and more useful, is MSK physiotherapy — and in much of the UK you can self-refer without seeing a GP first. Search your local NHS trust plus "MSK self-referral". Quadriceps strengthening has better evidence behind it for knee osteoarthritis than any retail product, and it costs nothing.
Should I wear a knee support all day?
Generally no, and this catches people out. Continuous compression is fine for short periods and during activity, but wearing a support from morning to night encourages you to rely on it, and the thigh muscle that stabilises the joint weakens further when it is not being asked to work. Powered heat and massage devices are used in sessions of roughly 15–20 minutes rather than continuously. If you find you cannot manage without a support on, that is a reason to see a physiotherapist, not a reason to wear it longer.
How long before a knee support makes a difference?
Compression is immediate — you feel steadier as soon as it is on, and that is real. Anything acting on stiffness works over weeks. In our 28-day test, none of the three testers noticed anything in week one, the ease of standing up improved in week two, and the mornings changed in weeks three and four. This is the single most important reason to check the returns window before you buy: a 14-day window expires before you have found out whether it works.
The verdict
If your knee gives way, buy a hinged brace for about £30 and don't spend more.
If your knee swells, buy proper graduated compression for about £79 and don't spend more.
And if your knee seizes — if the worst moments of your day are the first fifteen minutes of the morning and the moment you stand up after sitting — then the honest finding of this test is that you have very probably been buying from the wrong category, possibly for years, and every one of the sleeves in your drawer was answering a question you never asked.
Two things worth doing this week regardless of what you buy: refer yourself to an MSK physiotherapist, and measure both thighs 15cm above the kneecap. Both are free, and between them they will tell you more about your knee than anything on a shelf.
CHECK AVAILABILITY — £59.99 » Our No. 1 pick · 60 days to send it back · 2-year warrantyHow we are funded. The Compare Desk is free to read and funded by commission earned on purchases made through links on this page. We do not accept payment for placement or for a favourable score, and commission does not vary by ranking position. The five criteria used to rank all eight products were set before testing began, and every product was bought by us at full retail price.
About our testing. Three testers with GP- or consultant-diagnosed knee osteoarthritis, 28 days per product, all products purchased at full retail price by us. This is a product test, not a clinical trial: the sample is small, the testers were not blinded, and placebo, seasonal variation and regression to the mean cannot be excluded. Results are reported as measured, including where they were negligible. Individual results will vary.
Prices. Checked at the time of the last update shown at the top of this article and subject to change. Always confirm current pricing with the seller.
Third-party complaint records. Statements about complaint volumes, business alerts and substantiation requests relating to other companies are drawn from those organisations' publicly published records at the time of writing and are reported as a matter of public record. Check them yourself before relying on them; they are updated over time.
Medical. This article is consumer product journalism and is not medical advice. No product described here treats, cures or reverses osteoarthritis or restores cartilage. Heat therapy is not suitable for everyone — do not use heated devices over broken or numb skin, over an acute injury or active swelling, or if you have reduced sensation, poor circulation, or diabetic neuropathy, and consult a clinician first if you are pregnant or have an implanted electronic device. A hot, red, swollen or locked joint, or joint pain with fever, requires medical assessment rather than a retail product.
- NICE — Osteoarthritis in over 16s: diagnosis and management, NG226 (exercise and muscle strengthening as core treatment).
- Better Business Bureau — published business profile, alerts and complaint record for Nooro, Inc. (public record; subject to change).
- Manufacturer and retailer published product pages and pricing for Bauerfeind GenuTrain, Neo G, and Boots own-brand supports.
- Advertising Standards Authority and US Federal Trade Commission rulings concerning copper-fabric therapeutic claims.
- Published literature on arthrogenic muscle inhibition and quadriceps weakness in knee osteoarthritis.
© The Compare Desk. Independent UK product testing. We do not accept payment for placement or for a favourable score.