Multiple studies consistently point in the same direction: recreational running does not increase the risk of hip and knee osteoarthritis, and may even be associated with a lower chance of it than an inactive lifestyle. This evidence is purely associative and has not yet come from controlled experiments, but the findings are consistent enough to put the popular myth into perspective. For people without prior joint problems, moderate running is not harmful to the joints; the biggest risk factors are a previous injury or surgery, older age and a higher BMI, not running itself.
The persistent idea that running causes osteoarthritis is not supported by the available studies. A large literature review concludes that the vast majority of clinical research finds no link between running and the breakdown of joint cartilage. Although running can cause overuse injuries to muscles, tendons and bone, it has not been shown to damage the cartilage in the hip or knee1.
Recreational runners even appear to have less osteoarthritis than people who do not exercise at all. In a comparative study, 3.5% of recreational runners had hip or knee osteoarthritis, compared with 10.2% of sedentary controls. The statistical difference was just short of significant (odds ratio 0.86; 95% CI: 0.69-1.07), but the trend is clear: moderate and regular running is not associated with a higher risk of osteoarthritis. People who have been running for fewer than 15 years even have a significantly lower chance of osteoarthritis compared with non-runners (odds ratio 0.60;2.
Competitive running at elite level tells a different story. Among elite runners, the prevalence of hip and knee osteoarthritis was 13.3%, nearly four times as high as among recreational runners. This difference relative to non-runners also just missed statistical significance (odds ratio 1.34; 95% CI: 0.97-1.86), but the higher load involved in elite sport does appear to send a risk-increasing signal. This applies specifically to the extremes of the sport, not to the average runner2.
Notably, the number of kilometres, the number of marathons completed, years of running experience and pace barely matter in themselves. In a survey of 3804 marathon runners, none of these training characteristics proved to be a significant predictor of osteoarthritis. Clear risk factors, however, were: older age, a higher BMI, a hereditary predisposition to osteoarthritis, and in particular a previous knee or hip injury or surgery3. That last factor is crucial: if a joint is already damaged, running can increase the risk of osteoarthritis.
The practical conclusion is therefore nuanced but encouraging for the everyday runner. Moderate running is probably not harmful to the joints, but researchers cannot yet pinpoint exactly where the boundary of 'too much' lies4,5. Anyone who has no prior joint injuries, maintains a healthy weight and runs at a recreational level need not fear that running will cause osteoarthritis. Anyone who does have a history of knee or hip problems would be wise to discuss this with a doctor or physiotherapist before significantly increasing their training load.
All claims are based on associative research (cross-sectional studies, surveys, literature reviews); no randomised controlled trials are available. Causality has therefore not been established. Selection bias is possible: people with joint complaints are more likely to stop running, which may skew prevalence figures among active runners in a favourable direction.