Pillar
Movement
Fitness, strength, and function get better in joint disease, and the disease activity scores get better a good deal less often than you will read. How much is known depends on which disease you have, by more than tenfold. This page says which is which, rather than averaging across them.
What the research found.
Resistance exercise lowered the DAS-28, a score built from a 28-joint count, a blood marker, and your own health rating, in rheumatoid arthritis, at a standardized mean difference of minus 0.69. That came from seventeen randomized trials and 1,010 patients. It lowered the sedimentation rate too, while pain on a visual scale didn't reliably improve.
In dermatomyositis and polymyositis, exercise improved muscle strength by 0.61 standard deviations, across nineteen studies and 298 patients. Creatine phosphokinase showed no difference between the exercise and control groups. That enzyme marks muscle breakdown, and the fear was that loading swollen muscle would push it up.
The whole randomized evidence base for exercise in Sjogren disease is six trials and 277 people. The standardized mean difference for heart and lung fitness was 0.59. That is the entire literature for one disease, rather than a selection taken from a larger body of it.
A Cochrane review of exercise in juvenile idiopathic arthritis found nothing significant at all. Function, quality of life, and fitness all crossed zero, at weighted mean differences of minus 0.07, minus 3.96, and 0.04. It pooled three trials and 212 participants.
Takken and colleagues, Cochrane Database of Systematic Reviews, 2008
The 2018 EULAR recommendations judged the public health activity targets fit for people with inflammatory arthritis and osteoarthritis. They didn't set a lower target for them. That was a decision rather than an oversight, and it's worth knowing when somebody tells you to take it easy.
Rausch Osthoff and colleagues, Annals of the Rheumatic Diseases, 2018
What exercise reliably does
Exercise in joint disease has been studied across dozens of conditions and hundreds of trials, and the results sort into three groups that get run together all the time. Fitness, strength, and function get better every time, fatigue and quality of life get better often, and the disease scores get better sometimes, in some diseases. The clearest finding across all of it is safety. Supervised strength and aerobic training have been tested in people with active disease and didn't speed up joint damage, which reverses decades of advice.
The disease score question, one disease at a time
Rheumatoid arthritis has much the strongest evidence here, where a review pooled seventeen randomized trials and resistance exercise lowered the DAS-28, a score built from a 28-joint count, a blood marker, and your own health rating, by a standardized mean difference of minus 0.69. Lupus goes the other way, with two reviews finding no real effect on the scores at all, while quality of life improved in both. Myositis answers the fear people raise constantly, because strength improved across nineteen studies while creatine phosphokinase showed no difference between the groups. That enzyme marks muscle breakdown, and the worry was that loading swollen muscle would push it up.
How much the evidence varies
Summaries treat joint disease as one thing and it isn't one thing. How much is known differs by more than tenfold between diseases, so advice that's solid in rheumatoid arthritis can rest on almost nothing in a rarer one, and that difference is the most useful thing on this page. Fibromyalgia has four Cochrane reviews and a network review of 65 trials, while Sjogren disease has six trials and 277 people in all, and relapsing polychondritis has case reports. Advice brought across from one disease to another may well be right, and it's still borrowed.
What exercise doesn't do
No trial here has shown exercise preventing the joint erosion that disease-modifying drugs prevent and that shows on an X-ray. Anyone offering exercise instead of methotrexate, a biologic, or immune suppression is going well past the evidence, and acting on that risks damage nothing later undoes. Bad news deserves as much room as good news here, and a Cochrane review of exercise in juvenile idiopathic arthritis came out flat on every outcome it pooled. The other overclaim worth knowing is that a working muscle makes its own form of a prescribed drug, which is on a great many sites and rests on lab work instead of on people.
What to do this week.
Aim at the ordinary target
European rheumatology guidance judged the general activity targets fit for people with inflammatory arthritis and set no lower one. So aim for aerobic work most days and strength work twice a week.
Start below what you think you can do
Most people quit because they began too hard. Two strength sessions a week is the dose most trials used, and bodyweight moves count, so a gym isn't needed to get the effects on this page.
Cut the load, don't stop
A flare complicates this instead of ending it. Cut the load, move to water, or work the joints that are behaving, because any of those keeps the habit going through a bad spell.
Ask which disease the advice came from
Advice gets brought across from rheumatoid arthritis to rarer conditions all the time. Ask your team which disease a piece of advice was tested in, because the answer is usually available and rarely offered.
Common misconceptions.
Myth. Cardio is more important than strength.
Reality. For autoimmune disease, the muscle-building and insulin-sensitizing effects of resistance training arguably matter more than aerobic capacity alone. The best programs combine both, but if you only have time for one, lift.
In plain words. Both kinds of exercise help. If you only have time for one, lift. Muscle does jobs that walking alone doesn't do.
Questions patients ask.
Will exercise damage my joints or set off a flare?
The trials say no, and this is the best-supported claim on the page. Supervised strength and aerobic training have been tested in people with active disease without speeding up joint damage, and the Cochrane review in lupus covered thirteen trials and recorded no flares and no serious harms. Sore muscles after starting are normal and aren't a flare, though a joint that turns hot, swollen, or much worse is worth a call to your team.
Does exercise lower disease activity?
Sometimes, and a good deal less often than you'll read. In rheumatoid arthritis, resistance training lowered the DAS-28, a score built from a 28-joint count, a blood marker, and your own health rating, by 0.69 standard deviations across seventeen trials, which is a moderate effect whose range reaches close to zero. In lupus, two reviews found no effect on the scores at all, though quality of life did improve. So the disease you have changes the answer, and averaging across diseases hides it.
How much should I be doing?
The 2018 EULAR advice gives you the same target as everybody else, and that was a choice rather than an oversight. In practice it means aerobic work most days plus strength work twice a week. Most trials that moved the scores used supervised strength sessions two or three times a week, for eight to twelve weeks, fitted to what the joints would take.
Can exercise replace my medicine?
No trial has shown that, and not one has tested it. Exercise trials measure the disease scores, fitness, strength, and function, and they don't measure the joint erosion your drugs prevent. Stopping your medicine to rely on exercise risks damage that no later treatment undoes, which is why doctors who know this literature don't suggest it. Exercise works alongside treatment, and that's worth having.
Why do I read that exercise makes my muscles produce anti-inflammatory chemicals?
Muscle does release signals when it contracts, and the trouble starts at the next step, because those findings come from lab work rather than from people. Calling them your body's own form of a prescribed drug isn't something any human trial supports. What gets measured in patients is fitness, strength, function, and the disease scores, so judge exercise on those.
Is there evidence for my own condition?
That depends a good deal on which one you have. Rheumatoid arthritis, axial spondyloarthritis, and fibromyalgia have reviews of dozens of trials, while Sjogren disease has six trials and 277 people and relapsing polychondritis has case reports. For adult-onset Still disease and IgG4-related disease, no exercise trial exists, and the nearest evidence is a Cochrane review in juvenile idiopathic arthritis that found no benefit on any outcome.
References.
- Wen Z; Chai Y. Effectiveness of resistance exercises in the treatment of rheumatoid arthritis: A meta-analysis. Medicine. 2021;100:e25019. 10.1097/MD.0000000000025019SR + MA
- Zhang Y; He Z; Yin Z et al. Effect of Exercise Interventions for Rheumatoid Arthritis: A Systematic Review and Network Meta-Analysis of Randomised Controlled Trials. Journal of pain research. 2025;18:5109-5126. 10.2147/JPR.S537227NMA / SR
- Bilberg A; Mannerkorpi K; Borjesson M et al. High-intensity interval training improves cardiovascular and physical health in patients with rheumatoid arthritis: a multicentre randomised controlled trial. British journal of sports medicine. 2024;58:1409-1418. 10.1136/bjsports-2024-108369RCT
- Rausch Osthoff A; Niedermann K; Braun J et al. 2018 EULAR recommendations for physical activity in people with inflammatory arthritis and osteoarthritis. Annals of the Rheumatic Diseases. 2018;77:1251-1260. 10.1136/annrheumdis-2018-213585EULAR recommendations for physical activity in people with inflammatory arthritis and osteoarthritis
- da Silva BISL; Dos Santos BRJ; Carneiro JA et al. Physical exercise for dermatomyositis and polymyositis: a systematic review and meta-analysis. Clinical rheumatology. 2022;41:2635-2646. 10.1007/s10067-022-06281-1SR + MA
- Frade S; O'Neill S; Greene D et al. Exercise as adjunctive therapy for systemic lupus erythematosus. The Cochrane database of systematic reviews. 2023;4:CD014816. 10.1002/14651858.CD014816.pub2Cochrane SR
- Gao Q; Fan C; Zhang Y et al. Effects of physical activity on health-related outcomes in Sjögren’s syndrome: a systematic review and meta-analysis of randomized controlled trials. Frontiers in Immunology. 2026;17. 10.3389/fimmu.2026.1753791Systematic review and meta-analysis of RCTs
- Kong L; Yu C; Wang C et al. Comparative efficacy of different exercise interventions in patients with ankylosing spondylitis: a systematic review and network meta-analysis. PeerJ. 2025;13:e20336. 10.7717/peerj.20336Systematic review and network meta-analysis of 48 studies published between 2002 and 2024
- Macfarlane G; Kronisch C; Dean L et al. EULAR revised recommendations for the management of fibromyalgia. Annals of the Rheumatic Diseases. 2017;76:318-328. 10.1136/annrheumdis-2016-209724EULAR evidence-based society recommendations
- Takken T; Van Brussel M; Engelbert R et al. Exercise therapy in juvenile idiopathic arthritis. Cochrane Database of Systematic Reviews. 2008;2010. 10.1002/14651858.CD005954.pub2Cochrane systematic review and meta-analysis of randomised controlled trials
This page gathers the published research on this subject into one place. The studies behind it were published between 1989 and 2026, and every figure links to the paper it came from. Those studies were peer reviewed. This summary of them was not. Dr. Sarah Luebker is reviewing these pages one at a time and has not reached this one yet, so it carries no medical review date and nothing here is her opinion or her advice to you. Each page gets updated as she reaches it. It is here in the meantime because the science is worth having in one organized place that is easy to find and easy to read. Talk to your own clinician before acting on any of it.