News · Fitness & Exercise
High-impact exercise did not wreck arthritic hips or knees
A British Journal of Sports Medicine analysis of 17,661 Danes with hip or knee osteoarthritis found those doing high-impact activity were less likely to need a hip replacement, not more.
Based on a peer-reviewed registry study in the British Journal of Sports Medicine
- Researchers used Denmark's Good Life with osteoArthritis (GLA:D) registry, covering 5,911 people with hip osteoarthritis and 11,750 with knee osteoarthritis, in the British Journal of Sports Medicine.
- Clinical guidelines consistently recommend low-impact activity for osteoarthritis. This tested whether high-impact activity actually causes harm.
- High-impact activity was associated with 36% to 48% lower odds of total hip replacement within 12 months.
- For knees, it was not associated with any change in total knee replacement risk, in either direction.
- Pain intensity, functional capacity and fear of joint damage did not explain the association, though greater functional capacity slightly reduced knee replacement odds (OR, 0.92; 95% CI, 0.87-0.98).
- Only about 5% of participants did high-impact activity, and they are likely to have had milder disease to begin with. That selection is the study's central limitation.
- Activity was self-reported on a single scale, and the follow-up was 12 months.
Tell a physiotherapist you have knee arthritis and want to run, and you will usually be steered toward a swimming pool. A study in the British Journal of Sports Medicine suggests that instinct deserves re-examination.
Following 17,661 Danes with hip or knee osteoarthritis, researchers found that high-impact activity was associated with a substantially lower risk of total hip replacement, and did not increase knee replacement risk.
The assumption being tested
The advice here has been stable for years. As the authors put it, clinical guidelines consistently recommend low-impact physical activity for the management of hip or knee osteoarthritis.
That guidance rests on a mechanical intuition: a worn joint is a damaged structure, and pounding it must accelerate the damage. Plausible, widely believed, and until now barely tested against what happens to people who ignore it.
Where the data came from
Participants came from the Good Life with osteoArthritis in Denmark registry, a national program that enrolls people with hip and knee osteoarthritis into education and exercise therapy.
The analysis covered 5911 participants with hip osteoarthritis, averaging 66.8 years and 69.5% women, and 11 750 participants with knee osteoarthritis, averaging 66.5 years and 72.7% women. Baseline activity was self-reported using the University of California, Los Angeles Activity Scale and sorted into high-impact, high-intensity, moderate-intensity or low-intensity.
Outcomes were self-reported total hip replacement or total knee replacement at 12 months.
What happened
Among people with hip osteoarthritis, high-impact activity was associated with 36-48% lower odds of hip replacement. Between a third and a half fewer reached surgery.
Among people with knee osteoarthritis, nothing moved. High-impact activity was not associated with altered risk of knee replacement in either direction.
The researchers then asked why, testing three plausible explanations measured at three months: joint pain intensity, functional capacity and fear of joint damage from activity. There was little evidence that any of them mediated the associations. One small thread did appear, where greater functional capacity was associated with a reduction in odds of knee replacement.
So the protective association for hips is real in the data, and unexplained by the obvious mechanisms.
The problem with who exercises
Here the caution has to be firm, because the finding is the kind people quote selectively.
Only 5.9% of the hip group and 5.2% of the knee group engaged in high-impact activity. That is a small, self-selected minority, and the most likely reason they could run or jump is that their osteoarthritis was milder. Milder disease also means less need for a replacement. The association could be substantially, or even entirely, a reflection of who was well enough to participate.
The design cannot rule that out. Activity was captured once, by self-report, on a single scale, and followed for only twelve months, which is a short window for a slow disease.
What it changes
Not the advice for an individual patient, at least not yet. What it does undermine is the confident assumption that impact is inherently damaging to an arthritic joint, for which this is the largest test so far and the answer was no.
The authors pitch it carefully: clinicians may consider recommending higher-impact activities for individuals with hip osteoarthritis, and supporting continued participation in knee osteoarthritis when symptoms, capacity and patient goals allow.
That last clause is doing real work. The finding is permission to keep going, for people already going. It is not yet an invitation to start.
People also ask
Does this mean I should start running with knee arthritis?
No. This is general information, not advice for your joints, and the study cannot support that leap. Only around 5% of participants did high-impact activity, and they were almost certainly people whose osteoarthritis was mild enough to allow it. The honest reading is that high-impact activity did not appear harmful in people already doing it, which is different from showing it is safe to take up. Anyone with osteoarthritis should discuss activity changes with a physiotherapist or doctor.
Why do guidelines recommend low-impact exercise then?
Because the assumption has been that loading a damaged joint accelerates its decline, and because low-impact activity is well tolerated. This study challenges the first half of that assumption rather than the second. The authors note clinical guidelines consistently recommend low-impact physical activity for the management of hip or knee osteoarthritis, and their results suggest that blanket caution may be costing some patients benefit.
Why did hips and knees behave differently?
The study does not settle that. High-impact activity was associated with 36% to 48% lower odds of hip replacement but no change in knee replacement odds. Hip and knee osteoarthritis differ in mechanics, in how they progress and in the threshold at which surgery is offered. The most cautious interpretation is that the protective signal is specific to hips and the knee result is genuinely neutral.
Couldn't the healthier people simply be the ones exercising?
Yes, and this is the objection that matters most. It is called confounding by indication: people with less severe disease can do high-impact activity, and people with less severe disease also need fewer replacements. The researchers tested whether pain, function or fear of damage explained the link and found little evidence they did, but that does not rule out unmeasured severity differences at baseline.
What counts as high-impact activity here?
Activity was self-reported using the UCLA Activity Scale, which grades habitual activity from inactive up to regular participation in impact sports. High-impact sits at the top of that scale and typically covers running, jumping and court sports. It is a single self-reported measure rather than a detailed activity log, which limits how precisely the dose can be described.
References
- Liew BXW, Gronne DT, Roos EM, Skou ST. High-impact physical activity participation and 12-month risk of joint replacement: a longitudinal mediation analysis of 17 661 patients with knee or hip osteoarthritis. British Journal of Sports Medicine (2026).
- National Institute of Arthritis and Musculoskeletal and Skin Diseases. Osteoarthritis.