Knee pain can make a runner question everything. Was it the extra long run, the faster session, years of accumulated mileage, or is running itself gradually wearing the joint down?
A large study of 3,804 marathon runners offers a useful answer: pain is common, but a long running history was not associated with a higher likelihood of diagnosed hip or knee arthritis. Previous joint injury, previous surgery, age, family history, and body mass index were more closely associated with arthritis.
The study at a glance
Hartwell and colleagues surveyed runners registered for the 2019 or 2021 Chicago Marathon. Of 37,917 runners contacted, 3,804 completed every question and were included in the analysis, giving a 10% response rate.
- Average age: 43.9 years, with a range from 18 to 83
- Average running history: 14.7 years
- Average weekly distance: 27.9 miles, or about 44.9 kilometres
- Average training pace: 8:52 per mile, or about 5:31 per kilometre
- Average marathons completed: 9.5; the median was five
- First-time marathoners: 17.1%
- Runners who had completed five marathons or fewer: 67.1%
The survey separated two outcomes: hip or knee pain during the previous year that was severe enough to prevent running, and hip or knee arthritis previously diagnosed by a doctor.
How common was knee pain?
Just over one-third of the runners - 36.4%, or 1,383 people - reported hip and/or knee pain during the previous year that had prevented them from running.
- 728 runners reported knee pain without hip pain
- 383 reported hip pain without knee pain
- 272 reported both hip and knee pain
Taken together, 1,000 of the 3,804 participants reported pain involving the knee. That is approximately 26.3% of the full sample, calculated from the study’s reported counts. Roughly one in four respondents had knee pain significant enough to interrupt running.
How common was arthritis?
Doctor-diagnosed hip and/or knee arthritis was reported by 7.3% of the runners, or 277 participants.
- Knee arthritis alone: 194 runners, or 5.1% of the full sample
- Hip arthritis alone: 56 runners, or 1.5%
- Arthritis in both the hip and knee: 27 runners, or 0.7%
Among runners aged 65 or older, 21.7% reported hip and/or knee arthritis. Among those younger than 65, prevalence was 6.7%. This was not a comparison with matched non-runners, so it cannot prove running prevents arthritis.
Did more running mean more arthritis?
No association was identified between cumulative running history and diagnosed hip or knee arthritis after the researchers adjusted for other variables.
- Number of marathons completed was not a significant predictor of arthritis
- Number of years spent running was not a significant predictor
- Average weekly mileage was not a significant predictor
- Average training pace was not a significant predictor
This challenges the idea that joints behave like tyres, losing a little more material with every mile. Human tissue responds to load, recovery, injury, age, and other biological influences. The study does not establish exactly how adaptation works, but it does not support a simple mileage-equals-arthritis relationship.
What was associated with arthritis?
Previous hip or knee surgery was associated with 5.85 times the odds of arthritis. A previous hip or knee injury that forced a break from running was associated with 5.04 times the odds.
Each additional year of age was associated with an 8% increase in the odds of arthritis in the adjusted model. Family history was associated with 3.47 times the odds. Each one-unit increase in body mass index was associated with a 10% increase in the odds. These are statistical associations, not proof of what will cause or prevent arthritis in an individual runner.
Previous injury also mattered for current symptoms. Runners with an injury history that had stopped them running had 3.30 times the odds of reporting hip or knee pain that prevented running during the previous year.
What predicted pain that stopped running?
Pain and arthritis did not follow exactly the same pattern. Previous injury and surgery were associated with both outcomes. Female runners had higher adjusted odds of reporting hip or knee pain, but sex was not a statistically significant predictor of arthritis under the corrected threshold.
The model found very small reductions in pain odds with each additional weekly mile and completed marathon. This does not prove adding mileage treats knee pain. Runners who hurt may reduce their distance, while comfortable runners continue accumulating miles and races.
Pain is a signal, not a diagnosis
Pain was much more common than diagnosed arthritis: 36.4% reported hip or knee pain that stopped running, while 7.3% reported arthritis. The categories can overlap, but they are not interchangeable.
The study did not determine the cause of each episode or measure its exact location, sensation, duration, or run-to-run behaviour. A useful record asks where pain appeared, when it began, what changed it, and whether its intensity was stable, rising, or settling.
What the study cannot tell us
- It was cross-sectional, so it can identify associations but cannot establish causation.
- Arthritis diagnoses were self-reported without new examinations or imaging review.
- Only 10% of contacted marathon registrants completed the survey.
- Chicago Marathon registrants are a selected group; runners forced out of the sport may be underrepresented.
- Several anatomical, genetic, traumatic, and metabolic influences were not fully captured.
- Adjusted analyses combined hip and knee outcomes, so their risk estimates are not for knee pain alone.
What should runners take from this?
The defensible conclusion is not that running is always safe for every knee. It is that, in this large sample, a greater lifetime running dose was not associated with greater odds of diagnosed hip or knee arthritis.
- Do not treat every painful run as evidence that the joint is wearing out.
- Do not ignore an individual pattern because running was reassuring at group level.
- Keep a clear record, especially after an injury, surgery, or return to training.
If knee pain is persistent, worsening, follows a significant injury, or affects normal movement as well as running, a qualified healthcare professional can assess it. A tracking record can support that conversation but cannot replace an examination.
From a painful moment to a useful pattern
Memory is not designed to compare dozens of runs. With TENSION, runners can paint the painful or tense area on a 3D body map, record its intensity, save the session, and return to their history later.
- Map the exact zone around the knee instead of writing only “knee pain”
- Record intensity immediately after a run and again the next morning
- Note whether the sensation followed a long run, speed work, hills, rest, or a return after time off
- Observe whether the hip or lower leg appeared at the same time
- Review whether the pattern settled, stayed stable, or became more frequent
This record does not diagnose the problem. It turns scattered moments into a visible timeline, helping runners notice patterns, describe symptoms more precisely, and share a clearer history with a qualified professional.
Read the study - Hartwell MJ, Tanenbaum JE, Chiampas G, Terry MA, Tjong VK. Does Running Increase the Risk of Hip and Knee Arthritis? A Survey of 3804 Marathon Runners. Sports Health. 2024;16(4):622-629.