• The knee is the most commonly affected joint of all, and six in ten of the people living with osteoarthritis are women.
• Between 10 and 12 per cent of all osteoarthritis follows an injury, and it can surface ten years or more afterwards.
• Body weight contributes through inflammation and signalling from fat tissue, not only through load on the joint.
• Black seed oil has the most knee-specific evidence of the topical herbs, and the trials using it on the skin did better than those using it orally.
• Boswellia has the best evidence of the herbs taken orally, though the trials behind it are small and uneven.
• With turmeric, absorption is the limiting factor, so the preparation matters more than the milligrams on the label.
• Finding what started the osteoarthritis comes before choosing a herb, and that is the part a blog post cannot do for you.

Osteoarthritis of the knee makes stairs difficult, and standing up from a chair can be painful. It’s the most commonly affected joint of all, with around 365 million people living with it. Six in ten of those people are women.

If you’re looking at herbs for osteoarthritis of the knee, there’s more research to go on than you might expect. The knee has been studied more than any other joint, so there’s something solid to work from. The quality of it varies a great deal, though, and there’s a bigger question underneath the research that none of it touches.

Two things about the knee change what’s likely to help. The first is that it carries you. Every step puts several times your body weight through the joint. That’s why knee pain changes the way you walk, in a way that pain in your hand doesn’t.

The second is that the joint lies close to the surface. There’s thin skin over the joint line and no thick layer of muscle in the way. So anything you rub into the skin has a short distance to travel before it reaches the tissue underneath. That’s the reason so much of the herbal research on knees looks at oils rubbed in rather than capsules. It’s also why what you put on your knee and what you take orally are two separate questions, and I’ve kept them apart below.

Osteoarthritis gets described as wear and tear, as though it were the inevitable result of using a joint for long enough. That isn’t what it is. Something started it, and that matters more than which herb you choose.

Between 10 and 12 per cent of all osteoarthritis is post-traumatic, meaning it follows an injury, and it can appear ten years or more after the event. The knee you hurt in your twenties can be the knee that aches in your fifties, and most people have long stopped connecting the two.

After a meniscal tear, the odds of going on to develop osteoarthritis rise around threefold for a minor tear and nearly eightfold where the damage was severe. After a cruciate ligament rupture the risk depends heavily on whether the menisci were preserved during surgery.

Body weight matters too. Researchers describe obesity as linked to knee osteoarthritis not simply through mechanical load but through chronic low-grade inflammation and altered signalling from fat tissue. Leptin is one of those signals. It rises with body weight and tracks with both pain scores and structural damage in the joint. The same pattern appears in osteoarthritis of joints that carry no weight at all, which points to something systemic rather than purely mechanical.

For women over 40 there’s another layer again. Falling oestrogen changes joints, muscle and connective tissue at the same time, and I’ve written about that in musculoskeletal syndrome of menopause.

This is where herbal medicine works differently from the conventional approach. A painkiller or an anti-inflammatory is aimed at the pain, and it does that job. A herbalist is trained to start further back, with what set the joint off and what is keeping it inflamed, and to work on that alongside the pain.

It also means that buying an anti-inflammatory herb off a shelf isn’t really herbal medicine. It’s using a remedy as a sticking plaster. The reason the joint got to this stage is still sitting underneath, and if that doesn’t change, the problem tends to come back.

Conventional care for knee osteoarthritis has changed, and plenty of people are still working from the older advice. The current NICE guideline came out in October 2022. It says not to routinely offer paracetamol, unless nothing else is suitable and it’s only used occasionally. That surprises people, because paracetamol was the standard answer for years.

For the knee specifically, the guideline says to offer a topical anti-inflammatory first. An oral one is only considered if the topical doesn’t work. So conventional care has moved towards the skin as well, which matters when you’re weighing up a herbal oil.

Ahead of any medicine, though, comes exercise. It’s offered to everyone with osteoarthritis, and where someone is carrying extra weight, losing some reduces pain and improves how the joint works. None of that rules out herbs. It tells you where they fit alongside everything else.

Nigella sativa is black seed, also called black cumin or kalonji. It isn’t the love-in-a-mist you might have growing in the garden, which is a different species, Nigella damascena. That one contains an alkaloid called damascenine, and its seeds are not the ones used medicinally. The seeds of Nigella sativa have been used medicinally for centuries, and the oil pressed from them has anti-inflammatory and pain-relieving actions.

It’s the herb that comes up most often for knees, and nearly all the research on it uses the oil externally. You massage it into the affected area rather than taking it orally, which sets it apart from the others in this post.

The trial that compared it with paracetamol

The trial people usually mean, when they mention black seed and knees, was published in 2016. Forty people with knee osteoarthritis used 1ml of Nigella sativa oil, massaged in every eight hours for three weeks. Then, after a month off, the same people took 325mg of paracetamol every eight hours for three weeks. Comparing people with themselves like that removes a lot of the variation between one person and the next.

Both reduced pain, and the reduction was greater with the oil. The participants averaged nearly 76 years old, so this was an older group with long-standing knee osteoarthritis rather than the mildest cases. The oil still came out ahead of the paracetamol.

What happens when you look at all the trials together

One trial on its own doesn’t settle anything. A 2024 systematic review gathered six randomised trials covering 370 people with knee osteoarthritis. Three of the trials using the oil on the skin found better pain relief than placebo. The trials using it orally were more mixed, with only one of the two showing a clear effect. That suggests it does more when it goes onto the joint than when it’s taken orally.

On safety the reviewers were clear. Both the topical and the oral forms were well tolerated, and no serious adverse events were reported in any of the trials.

On how well it works, the reviewers were more cautious. The trials vary in size and design, so they stopped short of a formal clinical recommendation and asked for better ones. That is a comment on the research rather than on the plant.

How I use it on the knee

This is a simple thing to be getting on with in the meantime. It’s safe for most people and you can start today. It works on the pain and the inflammation, though, rather than on whatever started the osteoarthritis in the first place.

Start with a good quality cold pressed black seed oil. It can go on neat, and I’d suggest 10 to 15 drops for each application.

If your skin is sensitive, or you notice any irritation, dilute it. Half and half works, and you can go lower still on the black seed oil if you need to. I’d suggest diluting it anyway if you’re going to use it over the longer term. Extra virgin olive oil, sweet almond oil and jojoba all work as the base, and a diluted oil spreads more easily over the knee.

Warm a small amount in your hands first, then massage it in for around five minutes. Work the front of the knee, the sides and the back. The massage isn’t just a way of getting the oil on. It helps the oil to penetrate the tissues, and it brings blood flow to the area.

Twice a day is a reasonable starting point, and three times a day is fine if you’d rather. Keep it going consistently for at least three weeks before you decide whether it’s doing anything. That’s the same length of time the trials ran for. It suits adults with mild to moderate knee osteoarthritis, and particularly anyone who wants to try an alternative to pain medication.

Not everything has to go on the skin. Several herbs taken orally have been tested on knee osteoarthritis, and Boswellia has the best evidence behind it, so I’ll start there.

Boswellia

Boswellia is the resin from the frankincense tree, known in Ayurvedic medicine as salai guggul or shallaki, where it has been used for centuries for inflammatory conditions. More trials have looked at it for osteoarthritis than at any other herb taken orally. A meta-analysis of seven trials covering 545 people found reductions in pain, in stiffness and in how well the joint worked. The doses used ran from 100mg to 250mg. The reviewers also concluded that at least four weeks is needed before any effect can be judged.

A larger comparison published in 2025 looked at seven different supplements across 39 trials and 4,599 people. Boswellia came first for pain and first for function. It was also the only one of the seven where the pain reduction reached statistical significance. The others may well be doing something, but the trials weren’t clear enough to show it.

The authors noted that the trials are small and that the pattern of results suggests some publication bias, which means studies finding no effect may never have been published. That is a fair caution, and it applies across supplement research rather than to Boswellia in particular.

Turmeric

Turmeric has been used as an anti-inflammatory for a very long time, and the European Medicines Agency’s assessment records its long history in Ayurvedic medicine specifically for arthritis. It also has the other real body of modern evidence, and here the form matters more than the dose. A 2025 analysis of 17 trials found that all the turmeric preparations reduced pain compared with placebo. The ones that did best were the bioavailability-enhanced preparations. Those use piperine from black pepper, or nanoparticles, or emulsions, and they reached the threshold that counts as a difference a patient would actually notice.

The reason comes down to absorption. Plain curcumin powder is poorly absorbed from the gut, so much of what you swallow never reaches your bloodstream. That’s why the preparation is the thing to look at on a label, rather than the milligrams on the front of the tub.

The reviewers rated the certainty of the evidence as low and asked for better trials. I’ve written more on how turmeric and Boswellia work together for joint pain, and there’s a broader piece on natural anti-inflammatory supplements.

Rosehip

Rosehip is one you can grow yourself, or pick from a hedge in autumn. A meta-analysis of three trials covering 306 people found rosehip powder reduced pain compared with placebo. The dose was 5g a day, taken over an average of just over three months. The number needed to treat came out at six. That means six people would need to take it for one of them to get a response they wouldn’t otherwise have had.

All three trials scored highly for quality, which is more than can be said for a lot of this field. There’s more on rosehip and joint pain in a separate post.

Ginger

Ginger has the longest record of the four for this exact use. The European Medicines Agency’s assessment documents it in traditional Chinese medicine for rheumatism, and in Western herbal practice for rheumatic and muscular disorders. It even records a traditional dose for rheumatic complaints, a quarter to one gram of the dried rhizome three times a day.

The modern trials are more mixed. In the 2025 comparison of seven supplements, ginger came third for pain and fifth for function, and neither result reached statistical significance. A trial tests one fixed dose in a mixed group of people, so a result like that tells you what those particular trials could demonstrate. It doesn’t undo several centuries of use for the same problem.

How the herbs compare across the wider research

Beyond those four, a lot of other plants have been tested, and the evidence thins out as you go. A network meta-analysis published in 2025 compared 26 botanical extracts across 36 trials and 3,285 people. Cucumber extract came top for pain and for function, and ashwagandha came top for stiffness.

The authors were careful about it. Only two small studies lay behind the cucumber result, and they described it as an exploratory signal needing a large trial to confirm.

That’s true of a lot of this research. Plenty of plants show an effect, and very few have been funded to the standard that would settle the question either way.

Why a trial understates a herb

Herbal medicine wasn’t built from trials. It was built from watching what happened to people, over centuries, and the trials arrived afterwards to check some of it. So where a trial hasn’t been run, or comes back unclear, that isn’t evidence against the plant. It usually means nobody funded the study, which is a question about research money rather than about the herb.

How these trials are designed changes what the results can tell you as well. A trial gives one fixed herb, at one fixed dose, to a group of people picked out by their diagnosis. That’s more or less the opposite of how herbal medicine is practised. Matching the herb, or more often the formula, to the person in front of you is the method, and a trial design removes it on purpose.

So when a herbal result looks modest, part of what’s being measured is the absence of that matching. It doesn’t tell you the plant did nothing. An average across a mixed group falls somewhere between the people who improved and the people who didn’t.

This is also the honest case for seeing a herbalist rather than buying something off a shelf. A woman who reads about Boswellia and orders a bottle is doing exactly what the trial did. She’s taking a fixed dose, with nobody asking what’s actually going on in her knee, or in the rest of her.

Safety and what to check first

There are a few things to check before you start. Boswellia, turmeric and ginger can all affect blood clotting, so tell your prescriber if you’re taking an anticoagulant. Turmeric at supplement doses isn’t suitable if you have gallstones.

Nigella sativa oil was well tolerated on the skin in all the trials. If your skin is sensitive, patch test a small area of your forearm and leave it 24 hours before using it properly. Don’t apply it to broken or irritated skin. Avoid it altogether if you’re allergic to the buttercup family, Ranunculaceae, because black seed belongs to that family.

If your knee is hot and swollen rather than stiff and achy, see your GP before you try anything. That’s a different picture and it needs looking at.

What herbs can and can’t do here

Everything above is what these plants have been shown to do for the inflammation and the pain, and for some of them the evidence is reasonable. What none of them do is answer why this knee, and why now.

That’s the question I’d be asking. What happened to the joint, what else is going on in the body, what’s driving the inflammation, and what else you’re taking that a herb might interact with. Those answers change what’s worth using, in what form, and alongside what else. They also change whether a herb is the right place to start at all.

None of which means doing nothing while you wait. The black seed oil is worth getting on with, and it’s the most sensible of these to try on your own, because it goes on the skin and the safety picture is straightforward. Just hold it as something to be getting on with, rather than as the answer. Without knowing what started the osteoarthritis, anything you use is working on the pain rather than the cause, and the cause is still there.

A personalised consultation can help you understand what is happening in your knee, and find an approach that fits. One thing worth ruling out first is gout, which rises in women after 40 and is often mistaken for osteoarthritis.