Can a GLP-1 Help Knee Osteoarthritis? What STEP 9 Found

STEP 9 asked a question that had never been answered in a proper randomized trial: if you treat obesity with a GLP-1, does knee osteoarthritis pain actually improve?

The trial (Bliddal H, et al., *New England Journal of Medicine*, 2024) enrolled 407 adults who had both a body mass index of 30 or higher and clinically diagnosed moderate knee osteoarthritis with meaningful daily pain. Participants were randomly assigned to weekly semaglutide 2.4 mg — the dose sold as Wegovy — or placebo, with both groups receiving lifestyle counseling. The trial ran 68 weeks, long enough for weight loss to plateau and for any pain benefit to stabilize rather than reflect a short-term placebo bump.

The main outcome was the change in the WOMAC pain score, a validated 100-point questionnaire that osteoarthritis researchers have used for decades. It asks about pain while walking, climbing stairs, lying in bed, sitting, and standing. Higher scores mean worse pain, so a larger drop is better.

Why this trial mattered: for years, the standard advice to someone with knee osteoarthritis and obesity was "lose weight, your knees will thank you." That advice was based on biomechanics and observational data, not on a randomized trial of a medication that reliably produces weight loss. Doctors were giving advice that most patients could not act on, and nobody had proven what would happen if they could.

Osteoarthritis is also not a small problem. It is the leading cause of chronic pain and disability in adults worldwide, and the knee is the most commonly affected joint. Women are affected more than men, and the gap widens sharply after menopause — a pattern we explore in [why joints ache during menopause](/blog/menopause-joint-pain-why-everything-aches-and-what-helps).

Pain dropped substantially in both groups — but far more on semaglutide.

Over 68 weeks, the semaglutide group's WOMAC pain score fell by 41.7 points. The placebo group's fell by 27.5 points. The difference between them was 14.1 points on a 100-point scale, and it was statistically significant.

Two things are worth pulling out of those numbers.

First, the placebo group improved a lot too — nearly 28 points. That is not a flaw in the trial; it reflects something real about osteoarthritis. Pain fluctuates, people enrolled in trials get structured lifestyle support and regular clinical attention, and the simple act of tracking symptoms changes them. Any claim about an osteoarthritis treatment that does not have a placebo comparison should be treated with suspicion.

Second, the 14.1-point gap is clinically meaningful, not just statistically significant. Osteoarthritis researchers generally consider a change of around 8 to 10 points on the WOMAC pain scale to be the smallest difference a patient would actually notice. The semaglutide advantage was well past that threshold — roughly the difference between "my knee hurts when I take the stairs" and "I take the stairs without thinking about it."

Physical function improved as well. Participants on semaglutide scored higher on the SF-36 physical function measure, which asks about real-world capability: carrying groceries, bending, kneeling, walking several blocks. Pain relief that does not translate into doing more is of limited value, so this secondary result matters.

Almost certainly both — and the mechanics alone are impressive enough to explain a large share of the benefit.

The mechanical argument. Every pound of body weight translates into roughly three to four pounds of force through the knee joint during walking, and considerably more when climbing stairs or rising from a chair. A person who weighs 220 pounds and loses 13.7% has shed about 30 pounds — which removes somewhere between 90 and 120 pounds of peak load from each knee with every step. Multiply that by several thousand steps a day and the cumulative relief is enormous.

The inflammation argument. Fat tissue is not inert padding; it is metabolically active and secretes inflammatory signaling molecules including interleukin-6 and TNF-alpha. These circulate and contribute to joint inflammation independently of mechanical load, which is part of why obesity is also associated with osteoarthritis in the hands — joints that bear no body weight at all. Losing fat mass reduces this inflammatory background, and there is evidence that GLP-1 receptor activation has direct anti-inflammatory effects beyond what weight loss alone would produce.

What STEP 9 could not separate. The trial was not designed to disentangle these mechanisms, and it did not measure cartilage structure on imaging. So we know pain and function improved; we do not know whether the joint itself is structurally better off, or whether the disease process slowed. Those are different claims, and only the first one is supported.

One caution worth naming: rapid weight loss on a GLP-1 also costs lean muscle mass if protein intake and resistance training are not deliberately maintained — and the muscles around the knee, particularly the quadriceps, are what stabilize the joint. Losing weight while losing quad strength can partially undo the benefit. Our guide to [preserving muscle with strength training on a GLP-1](/blog/strength-training-on-glp1-how-to-preserve-muscle-while-losing-weight) covers the protocol.

Because three separate things converge in midlife, and they compound each other.

Estrogen withdrawal. Estrogen receptors are present in cartilage, bone, and the synovial lining of joints. When estrogen falls during the menopause transition, cartilage turnover shifts and inflammatory signaling increases. This is one reason so many women report new joint aches in perimenopause that they had never experienced before — including in joints with no prior injury.

Body composition change. The menopause transition is associated with a shift toward central fat storage and a decline in lean muscle mass, even when total weight stays flat. Less muscle around the knee means less shock absorption, and more visceral fat means more circulating inflammatory signals. We cover the fat redistribution piece in [why belly fat shifts in menopause](/blog/glp1-menopause-visceral-fat-why-belly-fat-shifts).

Bone changes underneath the joint. The bone directly beneath knee cartilage remodels as bone density declines, altering how load is distributed across the joint surface. This is a slower process, but it means osteoarthritis and osteoporosis are not entirely separate stories.

The practical implication: for a woman in her fifties with knee pain and a BMI over 30, a GLP-1 is addressing several of these threads at once. But it is not addressing the estrogen piece — that is a separate conversation about whether hormone therapy is appropriate for you, and our guide to [HRT patch vs. pill vs. gel](/blog/hrt-patch-vs-pill-vs-gel-which-is-safest) is a starting point for that discussion with your clinician.

It is also worth knowing that STEP 9's participant group was predominantly female, which makes the results particularly relevant to this population — though the trial was not designed to report results separately by menopause status.

STEP 9 does not answer that question, and it is important to be precise about why.

The trial measured pain and function over 68 weeks. It did not measure joint structure on MRI, it did not track cartilage thickness, and it did not follow participants long enough to count knee replacement surgeries. Those would require a different trial design running for years.

What we can say honestly:

• •Symptom relief was real and clinically meaningful. People hurt less and did more.
• •We do not know whether the underlying joint disease slowed. Osteoarthritis is a structural condition, and feeling better does not automatically mean the cartilage is better.
• •Observational data hints at fewer joint replacements in people who lose substantial weight, but observational data cannot separate the effect of weight loss from everything else that differs between people who lose weight and people who do not.

There is a separate and more immediate consideration if surgery is already on your calendar. GLP-1 medications slow stomach emptying, which raises the risk of aspiration under anesthesia. Anesthesiology guidance now includes specific instructions about holding GLP-1 doses before elective procedures — we cover the current rules in [GLP-1s before surgery](/blog/glp1-before-surgery-what-the-2024-guidance-says).

What to do with this information. If you have knee osteoarthritis and obesity, STEP 9 is a legitimate reason to raise GLP-1 therapy with your doctor as part of a pain management plan rather than purely as a weight intervention. Pair it with the things that independently work for knee osteoarthritis: quadriceps strengthening, low-impact aerobic activity, and adequate protein to protect muscle during weight loss. The medication reduces load; the muscle work protects the joint. Doing one without the other leaves benefit on the table — [how much protein you really need on a GLP-1](/blog/protein-on-glp1-how-much-you-really-need) has the specific targets.