Chondroitin and glucosamine for knee pain: the trials and reviews, figure by figure
What we found
The GAIT trial randomly assigned 1583 patients with symptomatic knee osteoarthritis (mean age 59 years, 64 percent women) to glucosamine, chondroitin sulfate, both, celecoxib or placebo for 24 weeks, and against a placebo response rate of 60.1 percent the response rate was 5.3 percentage points higher for chondroitin (P=0.17) and 6.5 percentage points higher for the combination (P=0.09), neither significant, while celecoxib was 10.0 percentage points higher (P=0.008) https://pubmed.ncbi.nlm.nih.gov/16495392/
In an exploratory analysis of GAIT patients with moderate-to-severe pain at baseline, the response rate was 79.2 percent with combined therapy versus 54.3 percent with placebo (P=0.002) https://pubmed.ncbi.nlm.nih.gov/16495392/
The LEGS trial followed 605 people aged 45-75 years with chronic knee pain for 2 years, and all four groups reported reduced knee pain over the first year with no significant between-group differences (p=0.93), while the combination reduced 2-year joint space narrowing by a mean difference of 0.10 mm compared with placebo (p=0.046) https://pubmed.ncbi.nlm.nih.gov/24395557/
A 2010 network meta-analysis of 10 trials in 3803 patients with hip or knee osteoarthritis found pain differences against placebo on a 10 cm visual analogue scale of -0.3 cm for chondroitin and -0.5 cm for the combination, against a prespecified minimal clinically important difference of -0.9 cm, and industry-independent trials showed smaller effects than commercially funded trials (P=0.02 for interaction) https://pubmed.ncbi.nlm.nih.gov/20847017/
The 2015 Cochrane review of chondroitin pooled 43 randomized trials with 9,110 participants and found pain 10 points lower than placebo on a 0-100 scale in studies shorter than 6 months (18 vs. 28), rated low-quality evidence, with 53 of 100 people on chondroitin and 47 of 100 on placebo reaching a 20% improvement in knee pain https://www.cochrane.org/CD005614/MUSKEL_chondroitin-osteoarthritis
A 6-month Spanish trial in 164 patients with moderate-to-severe knee pain found the chondroitin-plus-glucosamine combination reduced global pain by 11.8 mm (19% reduction) versus 20.5 mm (33% reduction) with placebo, so the combination was inferior to placebo in the modified intent-to-treat population https://pubmed.ncbi.nlm.nih.gov/27477804/
The CONCEPT trial of a pharmaceutical-grade chondroitin sulfate followed 604 knee osteoarthritis patients for 182 days and found pain fell by 42.6 mm with chondroitin sulfate 800 mg/day versus 33.3 mm with placebo (p=0.001) https://pubmed.ncbi.nlm.nih.gov/28533290/
The US National Center for Complementary and Integrative Health, in a page last updated October 2023, says "It's still uncertain whether glucosamine and chondroitin are helpful for knee osteoarthritis symptoms" and notes the 2019 ACR/AF guideline strongly recommends against both for knee osteoarthritis https://www.nccih.nih.gov/health/glucosamine-and-chondroitin-for-osteoarthritis
By our arithmetic from the CONCEPT figures above, the difference between chondroitin and placebo is 9.3 mm on a 100 mm scale, roughly the same size as the Cochrane short-term difference of 10 points.
Who was studied
Every trial above enrolled adults with diagnosed, painful knee osteoarthritis: GAIT's mean age was 59 years, LEGS took people aged 45-75 with radiographic narrowing of the joint, and MOVES, CONCEPT and the Spanish trial took Kellgren-Lawrence grade 2-3 osteoarthritis or patients diagnosed to American College of Rheumatology criteria. None enrolled healthy runners, triathletes or people with runner's knee (patellofemoral pain) without osteoarthritis.
The main trials and reviews (figures as printed by each source)
| Study (type) | Year | People studied | Duration | Result as printed | Source |
|---|---|---|---|---|---|
| GAIT (randomized trial, NEJM) | 2006 | 1583 patients with symptomatic knee osteoarthritis, mean age 59 years, 64 percent women | 24 weeks | Response vs placebo (60.1 percent): chondroitin +5.3 percentage points (P=0.17), combination +6.5 percentage points (P=0.09), celecoxib +10.0 percentage points (P=0.008) | https://pubmed.ncbi.nlm.nih.gov/16495392/ |
| Wandel et al. (network meta-analysis, BMJ) | 2010 | 10 trials in 3803 patients with knee or hip osteoarthritis | Trials pooled over multiple time points | Pain vs placebo on 10 cm scale: glucosamine -0.4 cm, chondroitin -0.3 cm, combination -0.5 cm; minimal clinically important difference -0.9 cm | https://pubmed.ncbi.nlm.nih.gov/20847017/ |
| Cochrane review of chondroitin (Singh et al.) | 2015 | 43 trials, 9,110 participants with osteoarthritis | 1 month to 3 years | Pain 10 points lower than placebo on 0-100 scale under 6 months (18 vs. 28), low-quality evidence; 53 vs 47 of 100 reached 20% pain improvement | https://www.cochrane.org/CD005614/MUSKEL_chondroitin-osteoarthritis |
| LEGS (randomized trial, Ann Rheum Dis) | 2015 | 605 people aged 45-75 years with chronic knee pain and medial joint narrowing | 2 years | No significant between-group pain differences (p=0.93); combination joint space narrowing mean difference 0.10 mm vs placebo (p=0.046) | https://pubmed.ncbi.nlm.nih.gov/24395557/ |
| MOVES (non-inferiority trial vs celecoxib, Ann Rheum Dis) | 2016 | 606 patients with grade 2-3 knee osteoarthritis and moderate-to-severe pain | 6 months | WOMAC pain change -185.7 with combination (50.1% decrease) vs -186.8 with celecoxib (50.2% decrease); no placebo arm | https://pubmed.ncbi.nlm.nih.gov/25589511/ |
| Roman-Blas et al. (randomized trial, Arthritis Rheumatol) | 2017 | 164 patients with grade 2-3 knee osteoarthritis and moderate-to-severe pain | 6 months | Pain fell 11.8 mm (19% reduction) with combination vs 20.5 mm (33% reduction) with placebo; combination inferior to placebo in mITT | https://pubmed.ncbi.nlm.nih.gov/27477804/ |
Where this fails
All of this evidence comes from older adults with diagnosed knee osteoarthritis, so it says nothing direct about a healthy runner with sore knees after a long block, about patellofemoral pain, or about preventing knee damage from mileage. The positive results (the GAIT moderate-to-severe subgroup, CONCEPT) are an exploratory subgroup and a trial of one pharmaceutical-grade product, and the BMJ analysis found commercially funded trials reported larger effects than independent ones. MOVES compared the combination with celecoxib and had no placebo arm, so its 50.1% decrease cannot show how much of the improvement was the supplement; in GAIT the placebo response alone was 60.1 percent. Supplement products differ in content and grade, and the trials used specified preparations and doses, so a shop product may not match any of them; the NCCIH page also notes a bleeding risk with warfarin and that glucosamine may raise blood glucose in some people. This is a summary of research, not medical advice.
Sources
- https://pubmed.ncbi.nlm.nih.gov/16495392/ - Clegg et al., GAIT, New England Journal of Medicine 2006 (doi 10.1056/NEJMoa052771), abstract; read 2026-11-15
- https://pubmed.ncbi.nlm.nih.gov/24395557/ - Fransen et al., LEGS, Annals of the Rheumatic Diseases 2015 (doi 10.1136/annrheumdis-2013-203954), abstract; read 2026-11-15
- https://pubmed.ncbi.nlm.nih.gov/20847017/ - Wandel et al., network meta-analysis, BMJ 2010 (doi 10.1136/bmj.c4675), abstract; read 2026-11-15
- https://www.cochrane.org/CD005614/MUSKEL_chondroitin-osteoarthritis - Singh et al., Cochrane review of chondroitin for osteoarthritis, published 28 January 2015, plain-language summary; read 2026-11-15
- https://pubmed.ncbi.nlm.nih.gov/25589511/ - Hochberg et al., MOVES, Annals of the Rheumatic Diseases 2016 (doi 10.1136/annrheumdis-2014-206792), abstract; read 2026-11-15
- https://pubmed.ncbi.nlm.nih.gov/28533290/ - Reginster et al., CONCEPT, Annals of the Rheumatic Diseases 2017 (doi 10.1136/annrheumdis-2016-210860), abstract; read 2026-11-15
- https://pubmed.ncbi.nlm.nih.gov/27477804/ - Roman-Blas et al., Arthritis & Rheumatology 2017 (doi 10.1002/art.39819), abstract; read 2026-11-15
- https://www.nccih.nih.gov/health/glucosamine-and-chondroitin-for-osteoarthritis - NIH NCCIH, Glucosamine and Chondroitin for Osteoarthritis, last updated October 2023; read 2026-11-15