Knee Osteoarthritis Treatment: What Actually Helps, According to the Trials

Close to a third of Maine adults told a federal survey in 2023 that a doctor had diagnosed them with arthritis. That is 32.7 percent, against 25.4 percent nationally, which makes the question of osteoarthritis in the knee joint, and what treatment actually earns its place, a fairly local one. The short answer from the trial literature is unglamorous. Structured land-based exercise, weight loss of at least 5 percent of body weight for people carrying extra, and self-management education have the strongest support. The 2019 American College of Rheumatology and Arthritis Foundation guideline strongly recommends all three, along with topical and oral NSAIDs, cane use, tibiofemoral bracing and tai chi, and it strongly recommends against glucosamine, chondroitin, platelet-rich plasma and stem cell injections for the knee. Nothing on that list regrows cartilage. The realistic goal is less pain and better function, not a repaired joint.

Maine’s numbers are high, and age does not fully explain them

The 32.7 percent figure comes from CDC’s Chronic Disease Indicators, built on BRFSS survey responses. New Hampshire sat at 28.6 percent that year.

The obvious explanation is age, and age does most of the work. Arthritis climbs steeply through the decades: in Maine, 12.0 percent of adults aged 18 to 44 report it, 37.4 percent of those 45 to 64, and 54.3 percent of those 65 and older. Women report it more often than men, 36.1 percent versus 29.0 percent.

But age adjustment does not erase the gap. On an age-adjusted basis Maine still runs 27.0 percent and New Hampshire 23.8 percent, against 22.5 percent for the country. Whatever is happening here, age structure is not the whole of it.

Inside the region the spread is wide enough to map. CDC’s PLACES county estimates put Washington County, Maine, highest at 39.7 percent crude (30.1 percent age-adjusted), with Piscataquis at 37.4 and Lincoln at 37.2. Cumberland County is lowest at 27.2 percent crude. In New Hampshire, Coos County leads at 33.8 percent while Hillsborough sits at 25.8.

Two caveats before anyone builds a theory on that. The survey asks about doctor-diagnosed arthritis of any type, not knee osteoarthritis specifically, so these figures cover rheumatoid arthritis, gout and everything else alongside OA. And Mainers who have arthritis are slightly more active than the national average, not less: 30.4 percent report physical inactivity, versus 32.8 percent of US adults with arthritis. New Hampshire is 29.3 percent. The usual inactivity story does not fit the data.

The disease is a whole-joint problem, not just worn cartilage

Osteoarthritis is the most common form of arthritis, and CDC estimates about 33 million US adults have it. The mental picture most people carry, cartilage wearing down like a brake pad, is too narrow. The National Institute of Arthritis and Musculoskeletal and Skin Diseases describes OA as “a degenerative joint disease, in which the tissues in the joint break down over time,” and the tissues it names include cartilage, tendons and ligaments, the synovium, bone, and, in the knee, the meniscus. The whole joint is involved.

That framing matters for expectations, because nothing in the guidelines rebuilds any of those tissues.

CDC lists five risk factors: repetitive joint stress or injury, obesity, family history, older age, and sex, noting that women are more likely than men to develop OA after 50. The lifetime figures put it in perspective. A 2008 analysis in Arthritis and Rheumatism estimated the lifetime risk of developing symptomatic knee osteoarthritis at 44.7 percent. With a history of knee injury it rose to 56.8 percent, and among people who were obese it approached two in three.

Why stairs are the hard part

Researchers have measured knee loading directly, using instrumented implants that report force from inside the joint. In five people, peak knee contact force reached 346 percent of body weight going down stairs and 316 percent going up, compared with 261 percent for level walking and 107 percent for standing on both feet.

Descending is the most demanding thing an ordinary knee does. Which is worth sitting with in a place where ordinary life includes the last mile down off a ridge (Acadia alone lists more than 150 miles of trails) and four months of icy granite steps and cellar stairs.

None of that causes osteoarthritis, and nothing in this evidence says cold or damp weather damages a joint. Whether pressure changes affect how a joint feels is a separate question, and we looked at it in our piece on cold weather, barometric pressure and joint pain. What the loading data does explain is why a knee that behaves on flat ground announces itself on the way down, and why body weight is not a cosmetic variable in this particular disease. Stair confidence also turns into a falls question once the ice arrives, which is its own subject and one we covered in balance training and fall prevention.

Osteoarthritis knee joint treatment, ranked by the evidence

Both major guidelines land in roughly the same place, and it is not where the advertising is.

OARSI’s 2019 guideline names the core treatments for knee OA as arthritis education and structured land-based exercise programs, with or without dietary weight management. The ACR and Arthritis Foundation guideline strongly recommends exercise, weight loss for patients who are overweight or obese, self-efficacy and self-management programs, tai chi, cane use, tibiofemoral bracing, topical NSAIDs, oral NSAIDs, and intra-articular glucocorticoid injections.

Notice what is missing: a recipe. ACR states outright that current evidence is insufficient to recommend a specific exercise prescription. Odd, for the best-supported treatment in the field. Honest, too.

The exercise evidence has also grown less certain over time, which deserves saying plainly. The 2015 Cochrane review of land-based exercise called its evidence high quality: across 44 trials and 3,537 participants, exercise reduced pain by the equivalent of 12 points on a 0 to 100 scale, held for two to six months afterward, and was comparable to published estimates for NSAIDs. The 2024 update pooled far more, 139 trials and 12,468 participants, and found a similar raw effect, 13.14 points for pain and 12.53 for function against no treatment or usual care. But it compared those against minimal important difference thresholds of 12 points for pain, 13 for function and 15 for quality of life, and concluded the benefits were “of uncertain clinical importance.” It also found no difference in effect between types of exercise. Real, modest, and less impressive than the 2015 version implied.

Weight has the cleanest mechanism behind it. The IDEA trial randomized 454 overweight and obese adults aged 55 and older with knee OA for 18 months. Diet plus exercise produced mean weight loss of 10.6 kg (11.4 percent), diet alone 8.9 kg, exercise alone 1.8 kg. The combined group reported less pain (3.6 versus 4.7 on a 0 to 20 scale) and better function than either single approach, and the dieting arms lowered knee compressive force by roughly 200 N compared with exercise alone. The ACR guideline puts the threshold at a loss of at least 5 percent of body weight for clinical and mechanistic change, with benefits continuing to increase at 5 to 10 percent, 10 to 20 percent, and beyond 20 percent. Doing that after 60 without shedding muscle along with the fat is a genuine problem, and one we have written about separately in protein needs and muscle loss in older adults.

Two head-to-head trials are worth knowing about. In 156 patients treated in the US Military Health System, physical therapy beat a glucocorticoid injection at one year, with a mean WOMAC score of 37.0 versus 55.8 on a 0 to 240 scale where higher is worse. And a 52-week trial of 204 people found tai chi statistically indistinguishable from standard physical therapy at 12 weeks (WOMAC improvement of 167 points versus 143, a difference whose confidence interval crossed zero), with benefits maintained out to a year. The tai chi group did significantly better on depression scores and on the physical component of quality of life.

Then there is walking, which costs nothing. In 1,212 people aged 50 and older followed in the Osteoarthritis Initiative cohort, those who walked for exercise had lower odds of developing new frequent knee pain (odds ratio 0.6) and less progression of medial joint space narrowing (odds ratio 0.8). That study is observational, and its authors present it as proof of concept warranting further study rather than a settled result.

The running question comes up constantly. A meta-analysis of 25 studies covering 125,810 people found hip and knee OA in 3.5 percent of recreational runners, 10.2 percent of sedentary non-running controls, and 13.3 percent of competitive runners. Before anyone reads that as protection, the authors state they could not determine whether the associations were causal or confounded by factors such as previous injury.

What the trials rule out

ACR strongly recommends against glucosamine, chondroitin sulfate, combination glucosamine-chondroitin products, platelet-rich plasma injections, stem cell injections, and transcutaneous electrical nerve stimulation for the knee. It conditionally recommends against vitamin D, fish oil, intra-articular hyaluronic acid for the knee, prolotherapy, intra-articular botulinum toxin, massage therapy, modified shoes, and wedged insoles.

The glucosamine verdict rests substantially on GAIT, which randomized 1,583 patients with symptomatic knee OA over 24 weeks. The placebo response rate was 60.1 percent, a number that says something on its own about how knee pain behaves in trials. Glucosamine came in 3.9 percentage points above placebo (P=0.30), chondroitin 5.3 points (P=0.17), the combination 6.5 points (P=0.09). Celecoxib managed 10.0 points (P=0.008). An exploratory subgroup with moderate-to-severe pain did favor the combination, and the authors called that exploratory for a reason.

Platelet-rich plasma got a proper placebo-controlled test in RESTORE, which randomized 288 people with mild to moderate symptomatic knee OA to leukocyte-poor PRP or saline. At 12 months, pain change was -2.1 versus -1.8 on an 11-point scale, a difference of 0.4 points with a confidence interval spanning zero. Medial tibial cartilage volume changed by -1.4 versus -1.2 percent. Of 31 prespecified secondary outcomes, 29 showed no significant difference between groups. ACR’s objection to PRP and stem cells is more fundamental than any single trial: the preparations are so varied and unstandardized that it is, in the guideline’s words, “difficult to identify exactly what is being injected.”

Vitamin D was tested too. A two-year trial of 413 people with symptomatic knee OA and low vitamin D levels found no significant difference from placebo in tibial cartilage volume or WOMAC knee pain, even though blood 25-hydroxyvitamin D rose by 40.6 nmol/L in the treatment group against 6.7 in placebo.

And then the awkward one, which the guideline itself does not resolve. ACR strongly recommends intra-articular glucocorticoid injections. But a two-year randomized trial of 140 patients found that repeated triamcinolone produced significantly greater cartilage loss than saline (index-compartment thickness change of -0.21 mm versus -0.10 mm) with no significant difference in knee pain (-1.2 versus -1.9). Those authors wrote that their findings “do not support this treatment” for symptomatic knee OA. Guideline and trial disagree, openly, and that tension belongs in a conversation with a clinician rather than being settled by an article.

On pain medication, OARSI rates topical NSAIDs Level 1A for the knee, does not recommend any oral NSAID for people with cardiovascular comorbidities or frailty, conditionally does not recommend acetaminophen, and strongly does not recommend oral or transdermal opioids.

Surgery, and the trial that keeps getting misquoted

Total knee replacement has been compared against non-surgical care in a randomized trial exactly once. One hundred eligible patients were randomized. Surgery plus 12 weeks of non-surgical treatment improved the KOOS4 score more than non-surgical treatment alone (32.5 versus 16.0, adjusted mean difference 15.8), and produced more serious adverse events (24 versus 6, P=0.005). The detail that tends to fall out of the retelling: only 13 of the 50 patients assigned to non-surgical treatment, 26 percent, went on to have the replacement within 12 months.

The other trial people reach for is FIDELITY, which randomized 146 patients aged 35 to 65 with a degenerative medial meniscus tear to arthroscopic partial meniscectomy or sham surgery, and found no significant difference in any primary outcome at 12 months. It is a striking result and it is widely cited as evidence that knee surgery does nothing for arthritis. It is not that. The trial excluded knee osteoarthritis by design, so it says nothing about surgery for OA. The misreading is common enough that it is worth flagging every time.

What it adds up to

The treatments with the best evidence behind them are the ones nobody markets: movement of whatever land-based kind a person will actually keep up with, since the 2024 review found no advantage for any particular type; weight management for people carrying extra, where the mechanical and clinical data agree; and structured education about the disease, which both guidelines rate as core rather than supplementary. Everything with a price tag and a testimonial sits lower down the list, and several of the most heavily promoted options sit in the strongly-not-recommended column.

The effects are moderate. The 2024 Cochrane authors declined to oversell them, and there is no reason for anyone else to. But moderate and real still beats a 6.5 percentage point placebo-adjusted nothing, and in a region where a third of adults report arthritis and more than half of everyone past 65 does, the gap between managing a knee and giving up on it works out to a great many Octobers.

Sources

• CDC. U.S. Chronic Disease Indicators (BRFSS-derived), arthritis and physical inactivity among adults with arthritis. https://data.cdc.gov/d/hksd-2xuw
• CDC. PLACES: County Data (GIS Friendly Format), 2025 release. https://data.cdc.gov/d/i46a-9kgh
• CDC. Osteoarthritis (OA). https://www.cdc.gov/arthritis/osteoarthritis/index.html
• National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS). Osteoarthritis. https://www.niams.nih.gov/health-topics/osteoarthritis
• Murphy L, et al. Lifetime risk of symptomatic knee osteoarthritis. Arthritis Rheum. 2008;59(9):1207-13. https://pubmed.ncbi.nlm.nih.gov/18759314/
• Kutzner I, et al. Loading of the knee joint during activities of daily living measured in vivo in five subjects. J Biomech. 2010;43(11):2164-73. https://pubmed.ncbi.nlm.nih.gov/20537336/
• Lawford BJ, et al. Exercise for osteoarthritis of the knee. Cochrane Database Syst Rev. 2024;12:CD004376. https://pubmed.ncbi.nlm.nih.gov/39625083/
• Fransen M, et al. Exercise for osteoarthritis of the knee. Cochrane Database Syst Rev. 2015;1:CD004376. https://pubmed.ncbi.nlm.nih.gov/25569281/
• Kolasinski SL, et al. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the Management of Osteoarthritis of the Hand, Hip, and Knee. Arthritis Care Res. 2020;72(2):149-162. https://pubmed.ncbi.nlm.nih.gov/31908149/
• Bannuru RR, et al. OARSI guidelines for the non-surgical management of knee, hip, and polyarticular osteoarthritis. Osteoarthritis Cartilage. 2019;27(11):1578-1589. https://pubmed.ncbi.nlm.nih.gov/31278997/
• Clegg DO, et al. Glucosamine, chondroitin sulfate, and the two in combination for painful knee osteoarthritis. N Engl J Med. 2006;354(8):795-808. https://pubmed.ncbi.nlm.nih.gov/16495392/
• Bennell KL, et al. Effect of Intra-articular Platelet-Rich Plasma vs Placebo Injection on Pain and Medial Tibial Cartilage Volume in Patients With Knee Osteoarthritis: The RESTORE Randomized Clinical Trial. JAMA. 2021;326(20):2021-2030. https://pubmed.ncbi.nlm.nih.gov/34812863/
• McAlindon TE, et al. Effect of Intra-articular Triamcinolone vs Saline on Knee Cartilage Volume and Pain in Patients With Knee Osteoarthritis. JAMA. 2017;317(19):1967-1975. https://pubmed.ncbi.nlm.nih.gov/28510679/
• Deyle GD, et al. Physical Therapy versus Glucocorticoid Injection for Osteoarthritis of the Knee. N Engl J Med. 2020;382(15):1420-1429. https://pubmed.ncbi.nlm.nih.gov/32268027/
• Messier SP, et al. Effects of intensive diet and exercise on knee joint loads, inflammation, and clinical outcomes among overweight and obese adults with knee osteoarthritis: the IDEA randomized clinical trial. JAMA. 2013;310(12):1263-73. https://pubmed.ncbi.nlm.nih.gov/24065013/
• Lo GH, et al. Association Between Walking for Exercise and Symptomatic and Structural Progression in Individuals With Knee Osteoarthritis. Arthritis Rheumatol. 2022;74(10):1660-1667. https://pubmed.ncbi.nlm.nih.gov/35673832/
• Alentorn-Geli E, et al. The Association of Recreational and Competitive Running With Hip and Knee Osteoarthritis: A Systematic Review and Meta-analysis. J Orthop Sports Phys Ther. 2017;47(6):373-390. https://pubmed.ncbi.nlm.nih.gov/28504066/
• Jin X, et al. Effect of Vitamin D Supplementation on Tibial Cartilage Volume and Knee Pain Among Patients With Symptomatic Knee Osteoarthritis: A Randomized Clinical Trial. JAMA. 2016;315(10):1005-13. https://pubmed.ncbi.nlm.nih.gov/26954409/
• Wang C, et al. Comparative Effectiveness of Tai Chi Versus Physical Therapy for Knee Osteoarthritis: A Randomized Trial. Ann Intern Med. 2016;165(2):77-86. https://pubmed.ncbi.nlm.nih.gov/27183035/
• Skou ST, et al. A Randomized, Controlled Trial of Total Knee Replacement. N Engl J Med. 2015;373(17):1597-606. https://pubmed.ncbi.nlm.nih.gov/26488691/
• Sihvonen R, et al. Arthroscopic partial meniscectomy versus sham surgery for a degenerative meniscal tear. N Engl J Med. 2013;369(26):2515-24. https://pubmed.ncbi.nlm.nih.gov/24369076/
• National Park Service. Acadia National Park, Hiking. https://www.nps.gov/acad/planyourvisit/hiking.htm

This article is for informational purposes only and does not constitute medical advice. Consult a qualified healthcare provider before making any health decisions.