I’ve spent 20 years coaching people, and no piece of fitness advice has cost more people more benefit than this one: don’t run, you’ll wreck your knees.
It sounds sensible. It gets repeated by well-meaning friends, family, and sometimes clinicians. And when you look at the largest body of evidence we have, it points the other way.
Quick Answer : In a meta-analysis of 114,829 people, hip and knee osteoarthritis prevalence was 3.5% in recreational runners, 10.2% in sedentary people, and 13.3% in competitive runners. Recreational running was linked to less arthritis than doing nothing at all.
Where the “Running Ruins Your Knees” Belief Came From
The belief isn’t stupid. It comes from a reasonable-sounding mental model: your joints are like the tyres on a car. Finite tread. Use them hard, wear them out faster.
Running does load the knee. Each stride sends force through the joint at roughly two to three times your body weight. Multiply that across 5,000 strides in a 5K, then across years, and the image of grinding cartilage writes itself.
Three other things reinforced it. Early research focused heavily on elite and competitive athletes — people running 60, 80, sometimes 100+ miles a week — and their outcomes were quietly generalised to everyone. Doctors saw injured runners in clinic and rarely saw the healthy ones who never needed an appointment. And “rest it” has always been the lower-risk thing to tell a patient.
What we now know is that cartilage doesn’t behave like tyre rubber. It’s living tissue that responds to load. Under regular, moderate loading it adapts — much like bone gets denser and muscle gets stronger. The absence of loading isn’t neutral for a joint. It’s its own kind of problem.
What the Largest Analysis Actually Found
In 2017, researchers led by Dr. Eduard Alentorn-Geli published a systematic review and meta-analysis in the Journal of Orthopaedic & Sports Physical Therapy. They pooled 17 studies covering 114,829 people and compared hip and knee osteoarthritis prevalence across three groups.
| Group | Hip & knee OA prevalence | What it suggests |
|---|---|---|
| Recreational runners | 3.5% | Lowest prevalence of the three groups |
| Sedentary non-runners | 10.2% | Nearly 3× the recreational runners |
| Competitive runners | 13.3% | Highest — elite and very high-volume athletes |
Read the middle row again, because it’s the one that changes the conversation. The group that wasn’t running had almost three times the arthritis prevalence of the group that was.
If your reason for avoiding running is protecting your knees, this data suggests avoidance may not be the protective choice you think it is.
What I see in the gym: The knee problems that walk into my facility almost never come from running itself. They come from doing too much too soon, from hips and glutes that were never trained, and from ignoring a niggle for six weeks. Those are programming errors — fixable ones — not an indictment of the sport.
Why Competitive Runners Are the Exception
Notice that the 13.3% figure belongs to competitive runners — higher than sedentary people. So the relationship isn’t “more running is always better.” It’s a U-shaped curve.
The U-shape, in plain terms:
- Too little (sedentary): 10.2% — joints lose the loading stimulus that keeps cartilage healthy; often paired with higher body weight and weaker supporting muscle.
- The middle (recreational): 3.5% — enough load to stimulate adaptation, enough recovery to absorb it.
- Too much (competitive): 13.3% — cumulative volume, high intensity, long exposure, and a much higher rate of prior joint injury.
Related research defines the high-volume threshold at roughly 57 miles (92 km) per week. That’s a serious training load — well above what almost anyone reading this is doing. If you run three or four times a week, you’re not near the risky end of the curve. You’re in the protective middle.
Prior injury likely explains a chunk of the competitive group’s numbers too. A previous ACL tear or meniscus injury is one of the strongest predictors of later osteoarthritis, and competitive athletes accumulate more of those.
What the Study Could Not Tell Us
Being honest about evidence means naming its limits, and this analysis has real ones. The authors said so themselves.
- Most of the included evidence was low quality. The finding is directional, not definitive.
- Risk of bias was potentially high, and heterogeneity between studies was substantial.
- Some comparisons rested on very few studies, with wide confidence intervals.
- It’s association, not proof of cause. Reverse causation is plausible: people with early joint pain may stop running, landing them in the “sedentary” group and making that group look worse.
- The safe upper limit couldn’t be determined. The research can’t tell you exactly how many miles a week is too many for you.
- Exposure beyond 15 years showed a different pattern than shorter exposure — long-term effects are less settled.
So the accurate summary isn’t “running prevents arthritis.” It’s this: the best available evidence does not support the claim that recreational running damages knees, and it points toward inactivity carrying its own joint risk. That’s a more modest claim than the headline. It’s also more useful, because it’s true.
What This Does Not Mean: Running Is Still Not Risk-Free
None of the above means you can’t get hurt running. Running has a genuinely high injury rate — roughly half of runners deal with an injury in a given year. The distinction that matters is overuse injury versus joint degeneration. They’re not the same thing.
| Common running injury | What it actually is | Usual cause |
|---|---|---|
| Runner’s knee (patellofemoral pain) | Irritation around the kneecap | Weak hips/glutes, poor tracking |
| IT band syndrome | Lateral knee irritation | Volume spikes, hip weakness |
| Shin splints | Tibial bone/muscle stress | Rapid mileage increase |
| Achilles tendinopathy | Tendon overload | Too much speed work, too soon |
| Stress fracture | Bone overload injury | Cumulative load without recovery |
Every one of these is a load-management problem, not evidence that running degrades joints. They’re largely preventable, and they resolve with appropriate treatment.
There’s also a group this article genuinely doesn’t apply to: people with existing significant joint damage, established osteoarthritis, or a history of joint surgery. “Running is fine for healthy knees” and “running is fine for your knees today” are different statements. The second one needs an examination, not an article.
If You Have Been Told Not to Run
A lot of people carry a decade-old instruction from a doctor and treat it as permanent. Sometimes that advice was exactly right. Sometimes it reflected the guidance of its time. Either way, it’s worth revisiting properly rather than assuming.
Please don’t read this as “ignore your doctor.” That’s not the message. The message is that a conversation from 2014 may deserve an update in 2026 — and that conversation should happen with a clinician who can actually examine you, not with an article.
If you want to reopen it, these questions tend to make the appointment more productive:
- “Was that advice based on something specific you found in my knee, or general caution?”
- “Has anything changed in the guidance since we last discussed this?”
- “If running isn’t advisable, is there a loading progression that would be — walking, then walk-run intervals?”
- “What specific signs should make me stop and come back to you?”
A sports medicine doctor or physiotherapist is usually better placed for this than a general appointment, simply because they see athletic populations daily and are closer to the current evidence.
How to Start Running Without Hurting Yourself
If you’re cleared to run and starting from scratch, the single biggest predictor of getting hurt is how fast you ramp up — not running itself.
The walk-run progression
Don’t start with continuous running. Start with intervals and let your tendons, bones and connective tissue catch up to your enthusiasm — they adapt slower than your cardiovascular system does, which is exactly why motivated beginners get hurt.
| Week | Session structure | Frequency |
|---|---|---|
| 1–2 | Run 60 sec / walk 2 min × 8–10 rounds | 3× per week |
| 3–4 | Run 90 sec / walk 2 min × 8 rounds | 3× per week |
| 5–6 | Run 3 min / walk 90 sec × 6 rounds | 3× per week |
| 7–8 | Run 5 min / walk 90 sec × 4 rounds | 3× per week |
| 9+ | Build toward 20–30 min continuous | 3× per week |
Three rules that prevent most injuries
- Never increase your longest run by more than ~10% in a week. Sudden jumps in a single run’s distance are the most consistently identified injury risk factor.
- Train your hips. Glute bridges, clamshells, single-leg work, twice a week. Most “knee” pain in runners is a hip problem showing up one joint lower — strengthening upstream fixes it more often than anything you do to the knee itself.
- Respect the signals. Mild soreness that settles within a day is normal adaptation. Sharp pain during a run, pain that alters how you move, swelling, or discomfort lasting several days means stop and get assessed.
Where Walking Still Wins
None of this makes running the right answer for everyone. Walking holds genuine advantages that no amount of arthritis data erases.
- Far lower injury risk. Ground forces are around 1.2× body weight versus 2–3× when running. For higher body weight, existing joint damage, or anyone returning from injury, that difference is decisive.
- It accumulates without a plan. Nobody accidentally runs three miles. But you can pick up thousands of steps across an ordinary day. If you want the detail on that, I’ve covered it in my guide to how many steps a day you actually need.
- Almost universal accessibility. Age, fitness level, joint status, postpartum recovery — walking accommodates nearly all of it.
- Comparable health outcomes at matched energy expenditure. It simply takes longer. Public health guidance treats 150 minutes of moderate activity as equivalent to 75 minutes of vigorous.
Running’s clearest edges are time efficiency and bone density — the higher impact is a stronger osteogenic stimulus. For a fuller side-by-side, see running vs walking: which is actually better for your health. And if you’d rather build a base first, my guide on high-volume walking covers how to do that sensibly.
Frequently Asked Questions
A. No — not at recreational volumes. In a meta-analysis of 114,829 people, recreational runners had 3.5% hip and knee osteoarthritis prevalence versus 10.2% in sedentary people. Cartilage adapts to regular moderate loading. Risk rises only in very high-volume competitive runners and those with existing joint damage.
A. Not at moderate volumes. The relationship is U-shaped: 10.2% prevalence in sedentary people, 3.5% in recreational runners, 13.3% in competitive runners. Both extremes carry more risk than the middle. Note this shows association, not proven causation.
A. No definitive limit has been established. The higher-risk competitive group trains around 57 miles (92 km) per week or more — far above most recreational runners. Rate of increase matters more than total: sudden jumps in a single run’s distance predict injury better than weekly mileage.
A. It’s patellofemoral pain syndrome — pain around the kneecap, worse on stairs or after sitting. It’s usually not structural damage and responds well to treatment. The cause is often weak hips and glutes causing poor knee tracking, so strengthening upstream typically resolves it.
A. Mild soreness that settles within a day means reduce load, not stop. Sharp pain, swelling, a knee that gives way or locks, altered gait, or pain lasting several days means stop and get assessed by a physiotherapist or sports medicine doctor.
A. Sometimes, with medical guidance. Some research in mild to moderate osteoarthritis found running didn’t accelerate disease progression and sometimes reduced pain. But it depends on severity, affected surfaces, and body weight — decide this with a clinician who can review your imaging, not from an article.
A. Walking generates about 1.2× body weight per stride versus 2–3× for running, making it lower-risk with existing joint problems or higher body weight. For healthy knees, there’s no strong evidence walking produces better long-term joint outcomes
A. Cartilage isn’t like tyre tread — it’s living tissue that adapts to loading. Imaging studies in long-term runners haven’t shown the progressive degeneration the wear-and-tear model predicts. Real risks are sudden load increases, running through pain, and running on already-damaged joints.
A. Age alone isn’t a reason to avoid it, and running’s bone-density benefit matters more as you age. What changes is recovery — progress more gradually, protect rest days, and add strength training twice a week. Get assessed first if you have joint symptoms.
Key Takeaways
- 3.5% vs 10.2%. Recreational runners had roughly a third the hip and knee osteoarthritis prevalence of sedentary people in a 114,829-person meta-analysis.
- The curve is U-shaped. Risk sits at both extremes — inactivity and very high-volume competitive running (around 57+ miles/week). The recreational middle is the protective zone.
- The evidence is directional, not definitive. Study quality was mixed and causation isn’t established. It still doesn’t support the claim that recreational running wrecks knees.
- Overuse injuries are real and common — about half of runners annually — but they’re load-management problems, not joint degeneration.
- Progress slowly, train your hips, respect pain signals. That’s most of injury prevention.
- If you have existing joint damage or were told not to run, get assessed — this article isn’t a substitute for that conversation.
A few related pieces that expand on what I’ve touched on here:
Pingback: Is Functional Strength Training Actually Better Or Just Trendy? - RoutineRX Health & Wellness Routine Academy