There are people reading this—you may be one of them—thinking, “Of course cycling is good for your knees, you idiot. Why do you think I switched from running?” I hear you. But there’s a big difference between something that’s “not bad” for your knees and something that’s actually good for them. The claim in a new study is that people with a history of cycling are less likely to develop osteoarthritis in their knees, and that’s a claim with some interesting implications both for cycling and for how we think about osteoarthritis.

　　The study was published last month in Medicine Science in Sports Exercise. It’s from a big team led by Grace Lo of Baylor College of Medicine, analyzing data from the Osteoarthritis Initiative, which followed thousands of older adults with and without osteoarthritis starting in 2004. This particular analysis zeroes in on 2,600 subjects who had knee X-rays, reported subjective knee pain, and filled out a questionnaire reporting their lifetime history of various physical activities during four periods of their life: 12 to 18 years old, 19 to 34, 35 to 49, and over 50.

　　About half of the subjects reported some history of cycling, and the headline result is that these people were less likely to have osteoarthritis in their knees. Osteoarthritis, or OA for short, is when the cartilage that helps your joints move smoothly starts to break down; it’s sometimes referred to as the “wear-and-tear” version of arthritis, though we’ll come back to why that’s a misnomer. You can diagnose OA with an X-ray showing the bones on either side of the joint coming closer together, or even coming into contact with each other. But what you see in an X-ray doesn’t always correspond to what you feel. The study used three different outcomes: persistent knee pain, radiographic OA (based on the X-ray), and symptomatic OA (when you have both knee pain and radiographic OA).

　　People with any history of cycling were 17 percent less likely to have knee pain, 9 percent less likely to have radiographic OA, and 21 percent less likely to have both. About half of the cyclists only reported cycling regularly during one of the four age periods, usually the youngest. But there was a pronounced dose-response relationship: cycling during 1, 2, 3, or 4 of the life periods reduced the risk of symptomatic OA by 17, 19, 28, and 43 percent. In other words, lifelong cyclists cut their risk of symptomatic knee OA by almost half.

　　What This Means for Cycling and Osteoarthritis Is this an obvious and expected finding? There’s a smattering of evidence that, for example, cycling squeezes knee cartilage in a way that produces less deformation than running, and might even “condition” cartilage to become more resilient over time. You could extrapolate that to suggest that cycling should be good for your knees. But other scientists take the opposite perspective. In a study last year, Spanish researchers claimed that “repetitive concentric movements, such as while cycling, can produce the microtrauma that leads to osteoarthritis.”

　　There’s another possible mechanism that has nothing to do with the knees themselves. In recent years, OA researchers have pushed back against the idea that the condition is a consequence of wear-and-tear on overused joints. It’s well established that obesity is a significant risk factor for OA, and it was long assumed that extra weight overloaded the joints. But that doesn’t explain why, for example, obesity doubles the risk of OA in non-weight-bearing joints like the hand. Instead, the systemic inflammation associated with a disrupted metabolism and accumulated fat deposits might be a key trigger.

　　In other words, cycling’s apparent benefits may have more to do with keeping weight off than with the details of cartilage compression in the knee. If that’s the case, you’d expect the same to be true for other activities like swimming and—dare I say it—running. And in fact, that’s true. Lo has previously published data from the Osteoarthritis Initiative finding that rates of OA were lower for people with a history of swimming and strength training. And yes, they were also lower for runners, consistent with prior evidence that running doesn’t actually ruin your knees, despite what every non-runner and ex-runner you know will continue to insist.

　　Not All Activity Lowers Your Arthritis Risk At this point, you might wonder whether Lo’s data shows that everything magically reduces OA rates. But there’s one exception to the streak of good news: a history of playing football, even just as a teenager, ramped up knee OA risk by as much as a factor of two. The elevated risk remained even after adjusting for BMI, which suggests that in this case the problem is likely related to acute knee injuries suffered on the field, a category known as post-traumatic OA. That’s bad news for football players, but a good sign that Lo’s data is capable of picking up risk factors.

　　The trickier question is what happens if you do develop knee OA. Running with the condition is tricky though not impossible. As encouraging as the new data is, more than half of the current or former cyclists in the study had radiographic evidence of OA, and a quarter of them had symptomatic OA. Is cycling more compatible with OA than other activities like running, or perhaps even capable of slowing its progression? At this point, there’s no evidence either way.

　　There are still some important caveats about the new study. Since it’s a retrospective, observational study, it’s impossible to pin down cause and effect. We only know that more cycling is associated with less knee pain, not that it causes it. But when you look at the whole body of evidence, including the swimming, strength training, running, and football data, the picture is encouraging. Yes, some of the cyclists you know will end up with bad knees—but it’s not about the bike.