Tomatoes and other nightshade vegetables are safe to eat with arthritis, and no controlled evidence links them to joint inflammation. The dietary change with the largest effect is weight reduction where weight is raised, which substantially reduces knee osteoarthritis pain. A Mediterranean-style pattern comes second, oily fish third, and purine and alcohol restriction fourth for gout specifically. Cutting out nightshades ranks fifth because it is the change people make most often and the one with the least behind it.
Getting enough protein matters more than most single food swaps once you are training an arthritic joint, and the numbers are in PhysioHub’s guide to how much protein you need.
Ranked by the strength of the evidence, with the condition each one actually applies to.
| Change | Applies to | Strength of evidence | Realistic effect |
|---|---|---|---|
| 1. Reducing weight | Knee and hip osteoarthritis, where weight is raised | Strong for the knee | Large |
| 2. Mediterranean-style pattern | Rheumatoid arthritis, and general health in all types | Moderate, trials are small | Modest but real |
| 3. Oily fish and omega-3 | Rheumatoid arthritis mainly | Moderate for RA, weak for OA | Modest |
| 4. Alcohol and purine restriction | Gout only | Established for attack frequency | Useful, secondary to medication |
| 5. Cutting out nightshades | Nothing, on current evidence | None | None expected |
Notice what the second column does. Most diet advice for arthritis is written as though arthritis were one condition. Purine restriction is genuinely useful for gout and irrelevant to osteoarthritis, and weight reduction is transformative for a knee and beside the point for a wrist.
This is a documentary review of published trials and guidance rather than a personalised eating plan. Around a dozen commonly recommended dietary changes were considered and five were kept, ranked on the quality of the evidence and on how large an effect the trials actually found.
The sources read were the US National Institutes of Health summary of the IDEA trial on diet and exercise in knee osteoarthritis, the MEDRA randomised trial protocol for Mediterranean dietary intervention in rheumatoid arthritis, a 2012 meta-analysis of omega-3 in rheumatoid arthritis, Harvard Health on nightshade vegetables and rheumatoid arthritis, and Versus Arthritis on diet. Evidence checked in August 2026.
Elimination diets, dairy exclusion, gluten exclusion and alkaline diets were considered and left out, because the evidence supporting them in arthritis is either absent or limited to uncontrolled reports.
Two honest limits. Nutrition trials are hard to blind and hard to run for long, so the certainty here is lower across the board than for drug trials. And no food prices are quoted in this article, because retail costs were not verified for this review.
Ask which type of arthritis the claim is about. A claim that does not specify is usually recycling advice from one condition and applying it to all of them.
Ask whether it was tested against a control. Joint pain fluctuates on its own over weeks, which is exactly the timescale on which people judge a new diet, and that is why uncontrolled reports are so consistently positive.
Ask what the effect size was, not just whether it was significant. A statistically significant change in a blood marker is not the same as a change you would notice.
Be suspicious of anything that removes a whole food group. The cost is nutritional and social, it is paid every day, and the benefit needs to be correspondingly clear.
Check whether it competes with the thing that works. Effort spent on an elimination diet is effort not spent on the exercise programme, which has far stronger evidence.
What it involves
A sustained reduction of around 10% of body weight, achieved alongside exercise rather than by diet alone, so that muscle is preserved while load is reduced.
Who it suits
People with knee or hip osteoarthritis whose weight is above the healthy range. It does not apply to anyone already at a healthy weight, and it does very little for hand or spinal osteoarthritis.
What the evidence supports
The IDEA trial, summarised by the US National Institutes of Health, found that combining diet with exercise produced better pain and function outcomes in knee osteoarthritis than either intervention alone. Around 10% loss is the threshold most consistently associated with meaningful symptom improvement.
Limits and cautions
Rapid weight loss without resistance training costs muscle, and muscle around an arthritic knee is precisely what you are trying to keep. Losing weight while joints hurt is its own practical problem, covered in exercising when your joints hurt.
Why it ranks first
It is the only dietary change on this list with an effect size comparable to medication, and the mechanism is both mechanical and inflammatory, so it works on two fronts at once.
What it involves
Vegetables, fruit, pulses, wholegrains, nuts, olive oil and oily fish as the base of the diet, with less red and processed meat and fewer refined carbohydrates. It is a pattern rather than a list of permitted foods, which is why it is sustainable.
Who it suits
People with rheumatoid arthritis have the most direct evidence. Everyone else gets the general cardiovascular and metabolic benefits, which matter because inflammatory arthritis raises cardiovascular risk independently.
What the evidence supports
Randomised trials including MEDRA have tested Mediterranean dietary intervention in rheumatoid arthritis and reported improvements in disease activity scores and symptom severity. The trials are small and short, which is why this sits second rather than first.
Limits and cautions
This supports medical treatment rather than replacing it. Nobody should reduce disease-modifying medication on the strength of a dietary change, and the trials did not test that.
Why it ranks second
It has the best evidence of any specific eating pattern in arthritis, it removes nothing essential, and its benefits extend well beyond the joints.
What it involves
Two portions of oily fish a week, or a supplement providing a meaningful dose of EPA and DHA. The trials that found benefit used doses well above what a typical supermarket capsule contains.
Who it suits
People with rheumatoid arthritis and other inflammatory arthritis. The osteoarthritis evidence is considerably weaker.
What the evidence supports
A 2012 meta-analysis found omega-3 supplementation reduced joint pain intensity, morning stiffness, number of tender joints and use of anti-inflammatory medication in rheumatoid arthritis. Effective doses in the trials were around 2.7g of combined EPA and DHA per day, sustained for at least three months.
Limits and cautions
The dose is the part people get wrong, and the timeframe is the second part: eight to twelve weeks before judging it. Fish oil at high doses can interact with anticoagulants, so check with a pharmacist if you take one.
Why it ranks third
The rheumatoid evidence is genuinely supportive and the effect is modest, so it belongs alongside treatment rather than near the top of a priority list.
What it involves
Reducing beer and spirits, and moderating high-purine foods such as red meat, offal, shellfish and oily fish, alongside reducing sugar-sweetened drinks. Adequate fluid intake sits with it.
Who it suits
People with gout, and nobody else on this list. This section has no application to osteoarthritis or rheumatoid arthritis.
What the evidence supports
The link between purine intake, alcohol and serum urate is well established, and dietary change reliably shifts urate levels. What it does not do is shift them as far as urate-lowering medication does, which is why guidance treats diet as a supporting measure in gout rather than a treatment.
Limits and cautions
Oily fish is high in purines and also the main food source of omega-3, so the advice at number three and the advice here genuinely conflict for someone with both gout and inflammatory arthritis. That is a conversation with a clinician rather than something to resolve from a web page.
Why it ranks fourth
Within gout it is useful and worth doing. Across arthritis as a whole it applies to a minority, which is what places it here.
What it involves
Removing tomatoes, potatoes, aubergines and peppers on the theory that solanine and related alkaloids in the nightshade family promote joint inflammation.
Who it suits
On current evidence, nobody in particular. It is included because it is among the most common dietary changes people with arthritis make, and it deserves a direct answer rather than being ignored.
What the evidence supports
Harvard Health states that there is no convincing evidence that potatoes or tomatoes, or any specific individual food, are bad for rheumatoid arthritis, and no scientific evidence that avoiding them improves it. The solanine argument does not survive contact with the detail either: the alkaloid is concentrated in the leaves, stems and green parts rather than in the fruit and tubers people eat.
Limits and cautions
Tomatoes and peppers are useful sources of vitamin C and carotenoids, and removing them makes a Mediterranean-style pattern harder to follow. If you have genuinely tested this on yourself and are confident a specific food affects you, that individual observation is worth respecting, and it is not a general rule.
Why it ranks fifth
It ranks last because it asks for daily effort and the evidence supporting it is absent, which is the worst combination on any list of this kind.
| Change | Osteoarthritis | Rheumatoid arthritis | Gout | Removes a food group |
|---|---|---|---|---|
| Reducing weight | Yes, knee and hip | Helpful generally | Yes | No |
| Mediterranean pattern | Helpful generally | Yes | Helpful generally | No |
| Oily fish and omega-3 | Weak evidence | Yes | Care, high purine | No |
| Alcohol and purine restriction | No | No | Yes | Partly |
| Cutting out nightshades | No evidence | No evidence | No evidence | Yes |
The last column is the one worth reading twice. The two changes with the best evidence remove nothing, and the change with no evidence behind it is the only one that takes food off the table permanently.
If you have knee osteoarthritis and your weight is raised, this is the highest-value change available to you, and it works best paired with a strengthening programme rather than done alone.
If you have rheumatoid arthritis, a Mediterranean-style pattern with two portions of oily fish a week covers items two and three together, alongside your usual medication.
If you have gout, dietary change is worth making and it is secondary to urate-lowering treatment. Diet alone rarely gets urate low enough to stop attacks.
If you already avoid tomatoes and feel better for it, there is no need to reintroduce them on the strength of this page. Test it properly if you want to know: reintroduce for four weeks and see whether anything actually changes.
If your weight is already healthy and you eat reasonably well, diet is not where your remaining gains are, and the exercise programme is.
If you are considering removing several food groups at once, speak to a dietitian first, particularly if you are older, since unintentional undernutrition does more harm here than any of these foods.
The belief that arrives most often in clinic is that something being eaten is causing the joint pain, and that identifying it would resolve the problem. It is an appealing idea because it offers control. What testing usually shows is a different story: the knee that hurts on stairs has a quadriceps that gave out two years ago, and the hand that aches in the morning belongs to someone who does three hours of gardening on a Saturday and nothing from Sunday to Friday.
The part of the diet question that does hold up is the part nobody enjoys hearing, which is weight where it is raised, and the part nobody thinks of, which is eating enough protein to build muscle while training. An older adult starting a strengthening programme on an inadequate protein intake gets a fraction of the result, and that is a dietary problem with a direct effect on an arthritic joint.
Where osteoarthritis is the diagnosis and you want the structured version of the exercise side, the osteoarthritis page covers what that looks like.
Are tomatoes bad for arthritis?
No. Harvard Health states plainly that there is no convincing evidence that potatoes or tomatoes, or any specific individual food, are bad for rheumatoid arthritis, and no scientific evidence that avoiding them improves it. The solanine argument does not hold up either, since the alkaloid is concentrated in the leaves, stems and green parts rather than in the fruit and tubers people eat.
Which foods should I avoid with arthritis?
For most types of arthritis, no specific food needs avoiding. Gout is the exception, where reducing alcohol, high-purine foods and sugar-sweetened drinks genuinely reduces attack frequency. Beyond that, the evidence supports changing the overall pattern of what you eat rather than removing individual foods.
Does an anti-inflammatory diet work for arthritis?
A Mediterranean-style pattern has the best evidence of any eating pattern tested, with randomised trials in rheumatoid arthritis reporting improvements in disease activity and symptom severity. The trials are small and short, and the effect is modest, so it works alongside medical treatment rather than replacing it.
How much omega-3 do I need for joint pain?
Trials that found benefit in rheumatoid arthritis used around 2.7g of combined EPA and DHA per day for at least three months. That is substantially more than most standard supplements provide, and the osteoarthritis evidence is much weaker than the rheumatoid evidence.
Can diet alone treat arthritis?
No. Weight reduction has a large effect on knee osteoarthritis symptoms, and it works best combined with exercise. For inflammatory arthritis, diet supports disease-modifying medication rather than substituting for it, and nobody should reduce that medication on the strength of a dietary change.
Does dairy make arthritis worse?
There is no good evidence that it does for most people. Dairy exclusion has been proposed repeatedly and tested poorly, and removing it costs calcium and protein at exactly the age when both matter most for bone and muscle. Test it properly if you suspect it, rather than removing it indefinitely on principle.
How long should I try a dietary change before deciding it works?
At least eight to twelve weeks, and ideally with something written down. Joint pain fluctuates over weeks on its own, so a change that coincides with a good spell feels far more convincing than it is. Reintroducing the food afterwards is the only way to know.
PhysioHub – Empowerment through Evidence-Based Education.
Every body is different. Book a thorough assessment and get advice that is specific to you and your goals.
Book an Appointment