It’s absolutely wild to me how much damage some people have done to their joints by following outdated advice.
When I see knee x-rays where all the cartilage has worn away and, instead, bone is grinding against bone, my heart sinks. At that stage, there’s really only one option left — a knee replacement.
I wish we could have had a conversation sooner. Because there’s been a profound change in how we understand the prevention and treatment of knee arthritis.
And a brand new study has reinforced this change in direction, offering fresh insights into the most promising strategies we can take.
Plus, there’s also a medication that’s recently been proven to help.
Table of Contents
The Old Paradigm
The advice many people follow today is rooted in a paradigm that was established more than a hundred years ago.
In the early 1900s, a prominent British physician, Sir Arbuthnot Lane, was puzzling over the bones of horses. At the time, cars were only just beginning to take over the roads, but horses were still a crucial part of the transportation network. As these horses pulled heavy loads across the hard, stony streets of London and other cities, they developed distinctive bony growths in their leg joints.
Lane’s attention was drawn to this problem because of an ongoing medical debate. These bony growths were a defining feature of osteoarthritis, a condition that affected many people, but whose cause was still poorly understood.
To his mind, the evidence from London’s cab horses was compelling. It pointed to a simple explanation: wear and tear. Use a joint enough, and it eventually breaks down. We now know this conclusion is wrong, but at the time, it made intuitive sense.
The key target of this breakdown was thought to be cartilage. Cartilage has several unusual properties. It contains no nerves and no blood vessels, which means it heals extremely poorly once damaged [1]. Even more importantly, because it is non‑vascularized and non‑innervated, cartilage cannot mount a typical inflammatory repair response.
So if we imagine a joint being used repeatedly over time, it’s easy to see how this tissue might gradually wear down. And if the tissue has virtually no ability to repair itself, then it can only tolerate so much stress before failure. The more we use it, the faster that failure point seems to arrive.
From this logic came a set of recommendations that sounded sensible — but were deeply flawed. If we want our knees to last into old age without pain, we should avoid using them too much. And when arthritis symptoms first appear, that’s a sign we should dramatically reduce activity.
The Paradigm Shift
More recently, this simple picture has changed — and become far more nuanced.
The shift began with a basic question: not everyone develops knee arthritis. So what separates those who do from those who don’t?
The wear‑and‑tear paradigm offers a straightforward answer. Those who are more physically active should be at higher risk. More movement means more wear. If the theory is correct, arthritis risk should rise in direct proportion to activity levels.
Some data initially appeared to support this idea.
A large meta‑analysis found that people in physically demanding occupations — including farmers, builders, and floor layers — had a 52% higher rate of knee arthritis compared to those in sedentary jobs [2]. These occupations involve frequent kneeling, squatting, lifting, and climbing, all of which place higher mechanical stress on the knees.
Another meta‑analysis compared people with high physical activity levels to those with moderate activity. Those in the high‑activity group had a 26% higher risk of knee arthritis [3].
There’s also a strong association between aging and increased arthritis risk, which further seems to fit the wear‑and‑tear model [3].
At first glance, all of this appears to reinforce the old story — joints can only take so much use, much like those horses plodding along London’s cobbled streets.
But it turns out this is only part of the story.
Other evidence points in a very different direction.
For one thing, the relationship between high physical activity and arthritis risk has been inconsistent. A cohort study that followed more than 1,100 older adults at high risk for knee arthritis for up to ten years found that long‑term participation in strenuous activities such as jogging, swimming, tennis, skiing, and cycling did not increase the risk of developing knee arthritis [4].
Then there’s evidence from the opposite end of the activity spectrum.
Researchers analyzed data from the National Health and Nutrition Examination Survey (NHANES), tracking physical activity levels and osteoarthritis diagnoses in the U.S. population from 2007 to 2020. What they found was clear: people who were more sedentary were significantly more likely to be diagnosed with arthritis [5].
Based on these findings, the authors explicitly recommended that individuals at risk of osteoarthritis should aim for regular physical activity, directly contradicting the traditional wear‑and‑tear hypothesis [5].
So why would being sedentary increase arthritis risk?
With long‑term inactivity, the muscles surrounding the knee joint shrink and weaken. This muscle atrophy reduces joint stability and protection, making cartilage more vulnerable to damage [3].
Physical activity does the opposite. It strengthens the quadriceps and other stabilizing muscles, improves skeletal muscle quality, and optimizes how mechanical stress is distributed across the joint [3]. It also enhances synovial fluid metabolism, helping lubricate the joint and modulate inflammatory processes [3].
Cartilage can be thought of like a sponge. When it’s compressed during movement, fluid is squeezed out. When the pressure is released, synovial fluid flows back in. Because cartilage lacks blood vessels, this compression‑and‑expansion cycle is a critical way nutrients are delivered to cartilage cells.
Taken together, the modern picture looks very different from the old one. Arthritis risk does not rise in a straight line with activity. Instead, it follows a U‑shaped curve. Too little activity weakens the joint and increases risk. Too much activity — or the wrong type — can also increase risk.
A recent meta‑analysis published in 2025 confirmed this complexity. While it concluded that individuals at risk of osteoarthritis should participate in regular physical activity, it also noted that current data are insufficient to define an optimal dose, frequency, or intensity of exercise [5].
The New Study
But what if you already have knee arthritis? Wouldn’t exercise make it worse?
The answer from the evidence is clear: no. Exercise helps. That’s why it’s considered a mainstream treatment for knee osteoarthritis [6].
In the past, people experiencing early arthritis symptoms were often advised to rest their joints and avoid activity. We now know this approach is completely wrong.
The more difficult question is what kind of exercise works best.
A brand new systematic review and network meta‑analysis set out to answer this question. Researchers analyzed 217 randomized controlled trials involving 15,684 participants with knee osteoarthritis [6]. Unlike observational studies, these clinical trials provide much stronger evidence about cause and effect.
The trials examined a wide range of exercise modalities, including aerobic exercise, flexibility training, strengthening exercises, mind‑body practices such as Yoga and Pilates, neuromotor and balance training, mixed exercise programs, and control groups [6].
Outcomes included pain, physical function, walking ability, and overall quality of life, assessed at 4, 12, and 24 weeks [6].
The clear winner was aerobic exercise.
It’s important to clarify what that means. In this study, aerobic exercise included walking, cycling, and swimming — and the authors’ recommendations are specifically tied to these activities [6]. The study does not provide data on running, so it can’t tell us whether running is advisable for people with knee arthritis.
For most of my patients, I recommend something low‑impact, like purchasing a second‑hand exercycle and using it at home.
There were also important limitations. Many of the included trials failed to clearly report exercise intensity, frequency, or duration [6]. As a result, we still don’t have precise guidelines like “X minutes per week at Y intensity.” More research is needed to clarify these details.
But there is another area where the evidence is crystal clear: excess body weight dramatically increases the risk of knee arthritis.
This isn’t just because heavier bodies place greater loads on the knee joints. Obesity also drives systemic inflammation, which plays a major role in the development and progression of osteoarthritis [7].
Diet and exercise are foundational for weight loss, but when they’re combined with GLP‑1 medications, the results can be dramatic.
A recent randomized controlled trial examined the effects of the GLP‑1 medication semaglutide in people with obesity and knee osteoarthritis. The study lasted 68 weeks and enrolled over 400 participants [8].
As expected, those receiving semaglutide lost a significant amount of weight — an average of about 14% [8].
But the most striking result was the effect on pain. Pain was measured using the WOMAC pain scale, which ranges from 0 to 100, with higher scores indicating worse pain. At baseline, participants averaged a pain score of about 71. By the end of the study, those taking semaglutide had reduced their pain by over 40 points, cutting pain levels by more than 50% [8].
I now routinely prescribe GLP‑1 medications to my overweight patients with knee arthritis. For many people, the potential for dramatic pain improvement is what finally motivates them to try these medications — and the results I’ve seen have been very encouraging.
There’s also another medication with similar effects that’s dramatically cheaper. In the next article, I’ll walk you through the studies behind that option and explain why it may represent a true breakthrough for treating knee arthritis.
Reference List
1. https://www.oarsijournal.com/article/S1063-4584(12)01025-4/fulltext
2. https://pmc.ncbi.nlm.nih.gov/articles/PMC7116019/
3. https://pmc.ncbi.nlm.nih.gov/articles/PMC12163862/
4. https://jamanetwork.com/journals/jamanetworkopen/fullarticle/2765373
5. https://pmc.ncbi.nlm.nih.gov/articles/PMC11771141/
6. https://www.bmj.com/content/391/bmj-2025-085242.long
