Is Running Bad for Your Knees? The Evidence Says the Opposite
The fear that running grinds your knees into arthritis is one of the most repeated ideas in fitness. A meta-analysis of 125,810 people found knee and hip arthritis in 3.5 percent of recreational runners versus 10.2 percent of people who did not run. Here is the honest read of what running does to a knee, and the load-management rules that actually protect it.
Key Points
- A 2017 systematic review and meta-analysis of 25 studies and 125,810 people found hip and knee osteoarthritis in 3.5 percent of recreational runners, compared with 10.2 percent of sedentary non-runners (Alentorn-Geli and colleagues, 2017).
- The same analysis found a higher rate, 13.3 percent, in competitive and elite runners. The relationship is a U-shape: too little movement and very high-volume competitive load both sit above the recreational sweet spot.
- A separate cross-sectional study of 2,637 adults in the Osteoarthritis Initiative found no increased risk of symptomatic knee arthritis in runners, and stated plainly that in people without arthritis, running does not appear to be detrimental to the knees (Lo and colleagues, 2017).
- Cartilage responds to a run by thinning a small amount, roughly 3 to 5 percent, then recovering. A 2023 meta-analysis of 446 knees found these changes small and transient, not damage (Coburn and colleagues, 2023).
- The mechanism is adaptation. Cartilage and bone get stronger in response to repeated, tolerable load, the same way muscle does. The real risk is not running, it is bad load management: too much, too soon, or training through a warning-sign injury.
- Structured exercise is a core, first-line treatment for knee osteoarthritis, not something to avoid once arthritis is present (OARSI guideline, Bannuru and colleagues, 2019).
The idea that running wears out your knees sounds obvious, and it is mostly wrong. When researchers pooled data on more than 125,000 people, recreational runners had less knee and hip arthritis than people who sat still. Running was linked to healthier joints, not more broken ones.
There is a catch, and it is honest. The lowest arthritis rate belonged to recreational runners. People who ran at a very high competitive volume for many years had a higher rate. So the protective effect is about a sensible dose, not unlimited mileage. Your knee is a tissue that adapts to load, and like any tissue it can be overloaded.
What actually hurts knees is not the act of running. It is going from zero to too much, or ignoring a real injury and pushing through it. If your knees are stiff, sore, or you are starting from scratch, the fix is to scale the load and build tolerance, and sometimes to get the joint moving well first. That is exactly what hands-on care is for.
Why This Question Matters
"The clinical question is not whether running is safe for knees. It is what dose, and how you build up to it."
Almost every week, someone in my office tells me they quit running because a friend, a relative, or an old coach warned them it would grind their knees to bone. It is one of the most durable myths in fitness. It feels intuitive: pounding a hard surface thousands of times must wear something down. The problem is that the human knee is not a machine part, and the data do not support the fear.
This matters because inactivity is genuinely dangerous. Osteoarthritis is the most common form of arthritis, and its risk factors include age, prior joint injury, obesity, and joint overuse (National Institute of Arthritis and Musculoskeletal and Skin Diseases). Two of those, obesity and deconditioning, get worse when people stop moving out of fear. If the fear of running is unfounded, then talking someone out of that fear is real preventive care.
The Study: 125,810 People
The 2017 review published in the Journal of Orthopaedic and Sports Physical Therapy set out to answer the exact question people worry about. The authors gathered every study they could find comparing joint arthritis in runners versus non-runners, then split the runners into two meaningful groups. Recreational runners were everyday runners with no elite competition history. The competitive group was made of professional, elite, or ex-elite athletes with very high training volumes over many years (Alentorn-Geli and colleagues, 2017).
Splitting those groups is the whole point. Lumping a weekend 5K runner together with someone who logged 100-mile training weeks for two decades hides the real signal. Separated, the pattern is clear and, honestly, more useful than a simple yes or no.
This is a meta-analysis of mostly observational studies, so it reports associations, not proof of cause. I cover what that limits below. But the sample size and the sensible group split make it the anchor for any honest answer to the knee question.
What They Found: A U-Shaped Curve
"Recreational running was linked to less than half the arthritis rate of sitting still."
Here is the finding that should end the myth for most people. Across the pooled data, hip and knee osteoarthritis showed up in 3.5 percent of recreational runners, 10.2 percent of sedentary controls, and 13.3 percent of competitive runners (Alentorn-Geli and colleagues, 2017). Read that again. The everyday runners had the healthiest joints in the study, better than the people who did not run at all.
| Group | OA Prevalence | 95% Confidence Interval |
|---|---|---|
| Recreational runners | 3.5% | 3.4% to 3.6% |
| Sedentary controls | 10.2% | 9.9% to 10.6% |
| Competitive / elite runners | 13.3% | 11.6% to 15.2% |
The competitive number is the honest caveat, and I will not bury it. Runners at the elite end, very high mileage for many years, often with old injuries and surgeries in the mix, had the highest rate of the three. That is why the true shape of the relationship is not a straight line but a U. Too little load and extreme chronic load both associate with more arthritis. The recreational middle is the sweet spot.
Two other large studies point the same direction. A meta-analysis in the American Journal of Sports Medicine found no meaningful difference in knee symptoms between runners and non-runners and even a protective association against needing knee surgery (Timmins and colleagues, 2017). And a cross-sectional analysis of 2,637 people in the Osteoarthritis Initiative found runners had no increased risk of symptomatic knee arthritis, concluding that in people without arthritis, running does not appear to be detrimental to the knees (Lo and colleagues, 2017).
Does Running Wear Out Cartilage?
The intuition behind the myth is that cartilage is like a brake pad that only wears down. That is not how living tissue works. When researchers imaged knees before and after running, they did see cartilage compress and lose a little thickness and volume right after a run, in the range of about 3.3 to 4.9 percent depending on the region. But those changes were small, and they recovered. The authors of that 2023 review concluded a single bout of running is not detrimental to knee cartilage (Coburn and colleagues, 2023).
"Cartilage is not a brake pad. It is a living tissue that adapts to the load you give it."
Think of it the way you think about muscle. Load it in a tolerable dose, let it recover, and it comes back stronger. Cartilage and the bone beneath it behave similarly. Repeated, moderate impact signals the joint to maintain and strengthen its tissue. That is the load-then-recover adaptation model, and it is the same principle that governs the companion research on why movement, not rest, is first-line care for ongoing pain, which I break down in this Forward Review on activity and chronic pain. Rest does not condition a joint. Sensible load does.
The failure mode is not impact itself, it is overload without recovery. When you exceed what the tissue is currently ready for, whether through a sudden mileage jump, poor recovery, or training through an injury, you tip from adaptation into irritation. That is where knee problems actually come from, and it is entirely a load-management problem, not an argument against running.
OARSI 2019 guideline (Bannuru and colleagues): for knee osteoarthritis, structured land-based exercise and arthritis education are core treatments, strongly recommended for nearly all patients regardless of disease severity or other conditions.
What this means in plain terms: even after arthritis is diagnosed, the guideline says the answer is more appropriate movement, not less. Motion is medicine for the arthritic knee, prescribed and progressed sensibly.
For the full guideline: Bannuru RR, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage. 2019.
What This Means Clinically
The runners and lifters I see in Tulsa almost never have a running problem. They have a load-management problem wearing a running costume. Someone doubles their weekly mileage before a race and their knee flares. Someone comes back from a layoff and trains like they never left. Someone has a cranky hip or a stiff ankle that quietly forces the knee to absorb load it was never meant to. The knee gets blamed, but the knee is usually the victim, not the culprit.
This is where hands-on care earns its place. When the hip or ankle loses motion, the knee compensates, and the tissue around it guards and tightens. Restoring mechanical motion up and down the chain, plus targeted soft-tissue work, can take the abnormal load off the knee so the joint can do its real job. I am not claiming an adjustment cures arthritis or rebuilds cartilage. I am saying it can restore the movement that lets you load the joint the way the research says keeps it healthy. Knee, hip, and gait complaints are part of the full list of what I work with, laid out on the conditions we treat in Tulsa page.
If You Already Have Knee Arthritis
A diagnosis of knee osteoarthritis is not a stop sign for movement. The guideline evidence is clear that structured, progressed activity is core treatment, not a risk (Bannuru and colleagues, 2019). What changes is the dose and the supervision. For an arthritic knee, that often means starting with lower-impact loading, building the muscle that supports the joint, and progressing carefully rather than avoiding load out of fear. When a knee keeps flaring in a recurring mechanical pattern, ongoing care makes sense: keep the joint and the chain moving, reassess as symptoms change, and adjust the plan. At Vanguard, that is a $40 walk-in visit, no contract, and you stop whenever you want. The goal the whole time is to keep you moving on your own.
Patient Translation: Manage the Load, Keep the Knee
The biggest mistake I see is people going from zero to too much, hurting themselves, and concluding that running is the enemy. It is not. Bad progression is. If you are starting from a stiff, deconditioned baseline, a short easy run or brisk walk you can repeat the next day beats a heroic effort that leaves you limping for a week. Consistency builds tolerance. Tolerance builds resilient tissue.
"The knee-saver is not avoiding load. It is respecting the dose your tissue is ready for today."
This scaling rule is not just for beginners, and I hold myself to it. I train at CrossFit Jenks, and the same principle governs a loaded barbell and a gymnastics movement as it does a recovery run. A kipping pull-up, a butterfly pull-up, or a chest-to-bar pull-up is a great movement when your shoulders, grip, and midline are ready for it, and a fast way to get hurt when they are not. The fix is never to quit moving. It is to scale the movement to the body in front of it, build the prerequisite strength, and progress on purpose. Knees are no different. Build the tissue, then ask more of it.
A few practical rules protect knees without giving up running. Increase weekly volume gradually rather than in big jumps. Keep at least one or two easy or rest days so tissue can recover and adapt. Strengthen the muscles around the hip and knee, because strong muscle is a shock absorber the joint relies on. And keep the neighboring joints, the hips and ankles, moving well so the knee is not forced to cover for them. If everything is stiff and you do not know where to start, that is exactly the moment a hands-on reset helps most, by getting motion back so the strengthening and the miles become possible.
Most knee aches from running are mechanical and respond to load management and hands-on care. A few signs are not mechanical and need prompt medical evaluation, not a chiropractor. Seek urgent or emergency care if you have any of these:
- A hot, red, swollen knee with fever or feeling unwell, which can signal joint infection
- A knee that locks, buckles, or gives way, or a pop with immediate swelling after a twist or fall, which can signal a ligament or meniscus tear
- Inability to bear weight or put the leg straight after an injury, which can signal a fracture
- Calf pain, swelling, warmth, or redness in one leg, which can signal a blood clot and needs same-day evaluation
- Numbness, spreading weakness, or a foot that drags, which points to a nerve problem, not a simple joint ache
A careful evaluation screens for these before any treatment or exercise advice. Honest care means knowing what a joint problem is not.
The anchor meta-analysis pools mostly observational studies, so it shows association, not cause. Runners may self-select for healthier joints, lower body weight, and fewer prior injuries, and those factors, not running itself, could drive part of the lower arthritis rate. The design cannot fully separate them.
The competitive-runner finding of 13.3 percent comes from elite athletes who often carry old injuries and surgeries, so it should not be read as proof that mileage alone causes arthritis. Confounding by injury history is likely.
The cartilage-imaging evidence for the load-then-recover model is rated low certainty by its own authors, and the follow-up windows were short. It supports that a single run is not harmful, not that any amount of running is safe for every knee.
These findings are about people without significant existing joint disease starting or continuing recreational running. They do not override individual medical advice for a knee that is already injured, unstable, or severely arthritic. If you have red-flag signs or a known joint problem, get evaluated before changing your training.
Running does not wear out normal knees. The largest analysis we have found arthritis in 3.5 percent of recreational runners versus 10.2 percent of people who sat still. Cartilage adapts to sensible load and recovers from it. The real risk is bad load management, not impact. Scale the dose, respect recovery, keep the whole chain moving, and your knees are more likely to thank you for running than to punish you for it.
Walk in to Vanguard Chiropractic in Brookside
If a stiff, cranky knee is keeping you from the movement you know would help, a hands-on evaluation can find where the load is going wrong and get the joint moving again. Walk in, get assessed, and train with a plan. No referral and no appointment needed.
Vanguard Chiropractic
1334 E 41st St, Brookside, midtown Tulsa, OK 74105
$40 per visit. No contracts. Stop any time.
Walk-in hours: Mon to Fri 10am to 7pm and Sat 10am to 2pm
Veteran-owned. Cash-pay, no insurance hassle.
Frequently Asked Questions
Is running bad for your knees?
Does running cause knee arthritis?
Does running wear down cartilage?
Can you run if you already have knee arthritis?
How do I protect my knees when I run?
Can a chiropractor help with knee pain from running?
Editorial Disclosure
The Forward Review is editorial commentary, not peer reviewed. Written and medically reviewed by Jason Forward, DC, a licensed chiropractor at Vanguard Chiropractic in Tulsa, Oklahoma. Commentary reflects clinical interpretation and does not represent the views of the original study authors, their institutions, or the guideline bodies cited. Individual results vary.
Conflict of interest: Dr. Forward is a practicing chiropractor whose clinical work includes hands-on care intended to restore movement. No financial relationships with the cited authors or publishers.
Corrections policy: Errors are corrected with a dated correction notice appended to the article. Original text remains accessible via revision history.
Level of evidence: Three systematic reviews or meta-analyses and one large cross-sectional cohort (primary evidence), one international clinical guideline, and one government health resource. Certainty ranges from low (cartilage-imaging data) to moderate (pooled prevalence).
- What Actually Helps a Child With ADHD, Ranked by Evidence Instead of PopularityForward Rev. 2026;1(2)
- Hands-On Therapy for Plantar Fasciitis: What a 2026 Study FoundForward Rev. 2026;1(2)
- Why Treating the Mid-Back Helps the Shoulder: 2026 Thoracic Manual Therapy EvidenceForward Rev. 2026;1(2)
- Shoulder Pain Without Surgery: What the 2025 Evidence Says About ExerciseForward Rev. 2026;1(2)
View Vol. I, No. ii and the full archive