The two headline ingredients in almost every joint supplement have been tested in one of the largest trials the NIH has ever funded for a supplement. Most reviews of this category don’t mention what it found.
Glucosamine and chondroitin were tested properly, and they did not beat placebo. The NIH-funded GAIT trial, published in the New England Journal of Medicine in February 2006, randomised patients with painful knee osteoarthritis to glucosamine 1,500 mg/day, chondroitin 1,200 mg/day, both together, celecoxib 200 mg/day, or placebo, for 24 weeks. There were no significant differences between glucosamine or chondroitin and placebo. The prescription anti-inflammatory celecoxib did beat placebo.
And the placebo group did remarkably well. A 60.1% response rate was recorded in the placebo arm — six in ten people improved on nothing at all. That single number explains almost every enthusiastic joint supplement testimonial you will ever read, including the honest ones. 🦵
- GAIT: the trial that settled it
- The 60% number and why it matters
- The comparison arm nobody quotes
- What’s happened since 2006
- What’s actually in this kind of formula
- The “16 ingredients” problem
- “FDA-approved facility” means less than you think
- Reading the review sites that rank for this
- What actually works for joint pain
- Why exercise is the awkward answer
- If you still want to try it
- Frequently asked questions
- Your checklist
GAIT: the trial that settled it
Most supplement categories can be criticised for having no good evidence. This one is different, and more interesting: the evidence exists, it’s excellent, and it’s inconvenient.
The Glucosamine/chondroitin Arthritis Intervention Trial — GAIT — was a multicentre, double-blind, placebo- and celecoxib-controlled trial published in the New England Journal of Medicine on 23 February 2006. Participants with painful knee osteoarthritis were randomised to one of five arms for 24 weeks:
| Arm | Dose | Result vs placebo |
|---|---|---|
| Glucosamine | 1,500 mg/day | No significant difference |
| Chondroitin | 1,200 mg/day | No significant difference |
| Both combined | 1,500 + 1,200 mg/day | No significant difference |
| Celecoxib (prescription NSAID) | 200 mg/day | Statistically significant benefit |
| Placebo | — | 60.1% response rate |
Note the doses. These were not token amounts — 1,500 mg of glucosamine and 1,200 mg of chondroitin daily are the standard, fully adequate doses, sustained for six months. This was not a study designed to fail.
You cannot dismiss GAIT as underdosed, too short, or badly designed. It was an NIH-funded, multicentre, double-blind trial testing full doses against both a placebo and an active drug. It is close to the best evidence a supplement has ever been given.
To be complete: an exploratory analysis suggested a possible benefit from the combination in a subgroup with moderate-to-severe pain. That was a secondary finding in a smaller subgroup, it was not the trial’s primary result, and it has not been established by subsequent work. It’s worth knowing about, and it’s not a basis for a purchase.
The 60% number and why it matters
Of everything in GAIT, the placebo response rate is the finding you should carry with you longest.
60.1% of people taking a sugar pill met the response threshold.
Sit with that. Six in ten people who took nothing improved enough to count as responders over 24 weeks. This is not people lying, and it isn’t weakness of mind. Several real things are happening at once:
- Osteoarthritis pain fluctuates. It has good weeks and bad weeks regardless of treatment.
- Regression to the mean. People start a supplement when pain is at its worst. From a peak, the likely direction is down — with or without a capsule.
- Genuine placebo analgesia. Expectation measurably changes pain perception. The relief is real even when the mechanism isn’t the pill.
- Behaviour changes alongside. People who start a joint supplement often also start walking more, losing a little weight, or paying attention to their knees. Those things work.
If six in ten people improve on a placebo, then a product with zero pharmacological effect will still generate an enormous volume of genuine, heartfelt five-star reviews. Every one of those reviewers is telling the truth about their experience. This is precisely why controlled trials exist — they are the only tool that separates “people got better” from “the product made people better.” Any review of a joint supplement that leans on testimonials rather than trials is answering a question you didn’t ask.
The comparison arm nobody quotes
GAIT included a fifth arm that gets far less attention than it deserves: celecoxib, a prescription NSAID, at 200 mg/day.
Celecoxib produced statistically significant pain relief versus placebo — roughly 70% response against the placebo’s 60%. The supplements did not.
The design point is elegant. A trial that finds nothing can always be accused of being insensitive — too small, too short, the wrong outcome measure. GAIT pre-empted that by including a drug known to work. The trial could detect a real effect, because it detected one. It simply didn’t detect one from the supplements.
Note also the size of celecoxib’s effect: about ten percentage points over placebo. That’s a real, prescription-strength anti-inflammatory, and its advantage over a sugar pill is modest. It’s a useful calibration for how much any oral treatment can be expected to do for knee osteoarthritis.
What’s happened since 2006
Twenty years is a long time, and the fair question is whether GAIT has been overturned.
It hasn’t. Subsequent reviews revisiting what has been learned since GAIT have not produced a clear, replicated reversal. The picture remains that glucosamine and chondroitin do not reliably outperform placebo for knee osteoarthritis pain, with continued debate about particular formulations, particular patient subgroups and particular endpoints — the kind of debate that persists precisely because no large, unambiguous positive result has arrived.
What has changed is the marketing. The ingredients that failed the test are now bundled with a dozen others, rebranded with trademarked names, and sold with claims about bioavailability rather than efficacy.
What’s actually in this kind of formula
Products in this space typically build outward from the same two ingredients. Taking the common components in turn:
| Ingredient | Role claimed | Where the evidence sits |
|---|---|---|
| Glucosamine | Cartilage building block | Tested at full dose in GAIT — no significant benefit vs placebo |
| Chondroitin | Cartilage support | Same — no significant benefit vs placebo |
| MSM | Sulphur, inflammation | Small studies, limited quality |
| Turmeric / curcumin | Anti-inflammatory | Some signal; absorption is the limiting problem |
| Boswellia | Anti-inflammatory | Some small positive trials |
| BioPerine / black pepper | Boosts absorption | Real effect on curcumin absorption — but it improves uptake, not efficacy |
| Branded chondroitin | “Superior bioavailability” | Better absorption of an ingredient that didn’t beat placebo |
The last two rows contain the key move in modern joint supplement marketing, and it’s worth naming explicitly. Bioavailability claims and efficacy claims are different claims. Demonstrating that more of an ingredient reaches your bloodstream tells you nothing about whether that ingredient helps once it arrives. If the parent compound didn’t beat placebo at a full dose, absorbing it better is solving the wrong problem.
The “16 ingredients” problem
Long ingredient lists read as generosity. Structurally, they’re closer to the opposite.
A capsule holds a finite amount of material. Sixteen ingredients in a capsule means each one is present in a fraction of what a single-ingredient product would contain — and studied doses for these compounds are not small. Glucosamine’s studied dose is 1,500 mg on its own; that’s already a substantial capsule before you add fifteen other things.
Take the supplement facts panel and compare each ingredient’s amount against the dose used in the research for that ingredient. Glucosamine 1,500 mg/day, chondroitin 1,200 mg/day — those are the GAIT doses. If a “16-ingredient” formula contains 500 mg of glucosamine, it is delivering a third of a dose that at full strength didn’t beat placebo. If the amounts are hidden inside a proprietary blend, you cannot do this check at all, which is itself the answer.
“FDA-approved facility” means less than you think
This phrase appears constantly in supplement marketing and it is worth understanding precisely, because it sounds like a regulatory endorsement and isn’t one.
The FDA does not “approve” facilities. Facilities that make dietary supplements are required to register with the FDA and to follow current Good Manufacturing Practice regulations, and they may be inspected. Registration is a requirement, not an award. GMP compliance concerns how consistently a product is made — that what’s on the label is in the bottle, made cleanly.
None of it says anything about whether the product works. A facility can be impeccably GMP-compliant while producing capsules of something with no clinical effect. Manufacturing quality and efficacy are entirely separate questions, and this phrasing invites you to hear the second when only the first is being claimed.
The genuinely useful quality signal is independent third-party testing — NSF, USP, Informed Choice — which verifies contents against the label. That’s meaningful, and it’s still not evidence of benefit.
Reading the review sites that rank for this
If you search this product category, the first page fills with sites whose names suggest independent consumer testing, awarding “Editor’s Choice” badges and top rankings.
A few things to check before trusting one, all of which take seconds:
- Does it cite GAIT? A genuinely independent review of a glucosamine product that never mentions the largest trial of glucosamine is telling you something.
- Does every product link to a purchase page? Look for affiliate parameters in the URL.
- Is one product always number one across every comparison on the site?
- Does the disclosure exist, and is it above the fold or at the bottom in grey?
- Are the “cons” real? “The only downside is it sells out quickly” is not a criticism.
We disclose our own position plainly: this site uses affiliate links, including in the box below. What you should hold us to is whether we recommend the thing we earn from — and in this article, we don’t.
What actually works for joint pain
Having spent this long on what doesn’t, here’s what the evidence supports. It’s less convenient and considerably more effective.
| Intervention | Evidence strength | Notes |
|---|---|---|
| Exercise and strengthening | Strong — first-line | Quadriceps strengthening for knee OA; works, and it’s free |
| Weight management | Strong | Every kilogram lost reduces knee load substantially |
| Physiotherapy | Strong | An individual programme beats a generic one |
| Topical NSAIDs | Good for knee/hand | Fewer systemic effects than oral — ask a pharmacist |
| Oral NSAIDs | Effective (celecoxib beat placebo in GAIT) | Real GI, cardiovascular and kidney risks — medical advice needed |
| Walking aids, footwear | Helpful, underused | Reduces load; no downside |
| Glucosamine / chondroitin | Did not beat placebo in GAIT | Safe, but that’s a different claim |
Why exercise is the awkward answer
Everyone in this field knows exercise is first-line, and almost nobody wants to hear it — for reasons worth taking seriously rather than dismissing.
It’s effortful, it’s slow, and it is counter-intuitive: being told to move a joint that hurts sounds like advice from someone who hasn’t tried. And there’s a genuine early cost — starting to strengthen a knee that has been guarded for years is uncomfortable before it’s better.
But the mechanism makes sense. Stronger muscles around a joint absorb load that would otherwise pass through the joint surfaces. Better conditioning improves function and reduces the stiffness that comes from avoiding movement. Nothing swallowed can do that, because the problem is mechanical.
See a doctor or physiotherapist before starting an exercise programme for a painful joint, particularly if the pain is new, severe, one-sided, follows an injury, or comes with swelling, redness, heat or fever. Those features need assessment rather than exercise. When you do start: begin far below what you think you can manage, progress weekly rather than daily, and expect some discomfort that settles within 24 hours. Pain that is worse the next morning means you did too much. A physiotherapist earns their fee here by getting the starting point right.
Resistance Bands for Joint Exercise
Strengthening is first-line for knee osteoarthritis — and unlike the supplement aisle, it has the evidence to match
We’re not recommending a joint supplement, because the two ingredients that define the category were given a full, fair, NIH-funded test and did not beat placebo. What did beat placebo in that trial was a prescription drug — and what beats both, in the wider evidence base, is strengthening the muscles around the joint. Resistance bands are the cheapest way to do that at home. They allow very light starting loads, which matters enormously when a joint is sore and you need to begin below the threshold that provokes it, and they let you progress in small increments rather than jumps. Get a set with graded resistances so you have somewhere to go. This is a tool to support a programme designed for you — not a substitute for getting one.
- Choose a set with graded resistances so you can progress gradually
- See a physiotherapist first if pain is new, severe, swollen or one-sided
- Start lighter than feels worthwhile — this is the most common mistake
- Progress weekly, not daily; soreness should settle within 24 hours
- Focus on the quadriceps for knee osteoarthritis
- Consistency over months beats intensity over days
If you still want to try it
Reasonable people take glucosamine and chondroitin knowing all of the above. They’re inexpensive, they have a good safety record, and a 60% placebo response is still 60% of people feeling better. If that’s your decision, make it a well-run experiment:
- Check the doses against GAIT — 1,500 mg glucosamine, 1,200 mg chondroitin daily. Anything materially less is below what was tested.
- Avoid proprietary blends. If you can’t see the amounts, you can’t run the check.
- Set a decision date before you start — say 12 weeks — and write down your current pain and function.
- Change one thing at a time. Starting a supplement and a walking habit in the same week guarantees you learn nothing.
- Stop if it hasn’t helped by your date. The commonest failure isn’t buying it; it’s still buying it two years later out of inertia.
- Tell your doctor. Particularly relevant if you take warfarin, are diabetic, or have a shellfish allergy — most glucosamine is derived from shellfish, though non-shellfish versions exist.
- Don’t let it delay assessment. Persistent joint pain deserves a diagnosis; osteoarthritis is not the only cause.
Frequently asked questions
Does glucosamine and chondroitin actually work?
The largest, best-designed test says no. The NIH-funded GAIT trial, published in the New England Journal of Medicine in February 2006, randomised patients with painful knee osteoarthritis to glucosamine 1,500 mg/day, chondroitin 1,200 mg/day, both, celecoxib 200 mg/day, or placebo for 24 weeks. There were no significant differences between the supplements and placebo. Celecoxib did beat placebo, which shows the trial was capable of detecting a real effect. An exploratory subgroup analysis hinted at benefit in people with moderate-to-severe pain, but that was a secondary finding and has not been established since.
Why do so many people say a joint supplement worked for them?
Because in GAIT, 60.1% of the placebo group responded. Six in ten people improved on a sugar pill over 24 weeks. Osteoarthritis pain naturally fluctuates; people start supplements when pain peaks, so the likely direction is improvement regardless; placebo analgesia is a real, measurable phenomenon; and people who start a supplement often start moving more at the same time. Every one of those positive reviews can be sincere and accurate about the person’s experience while telling you nothing about the capsule.
Is a 16-ingredient formula better than a simple one?
Usually the reverse, mechanically. A capsule holds a finite amount, so more ingredients means less of each — and the studied doses here are large. Glucosamine’s studied dose is 1,500 mg a day on its own. A formula splitting a capsule sixteen ways is almost certainly delivering fractions of doses that, at full strength, didn’t beat placebo. Check each amount against the research dose; if the amounts are hidden in a proprietary blend, that’s your answer.
What does “manufactured in an FDA-approved facility” mean?
Less than it sounds. The FDA doesn’t approve facilities — supplement manufacturers are required to register and to follow current Good Manufacturing Practice rules, and may be inspected. That’s a baseline requirement about manufacturing consistency, not an endorsement, and it says nothing at all about whether a product works. A perfectly GMP-compliant facility can reliably produce capsules with no clinical effect. The more useful signal is independent third-party testing such as NSF or USP, which verifies that what’s on the label is in the bottle.
What should I do instead?
Strengthening exercise and weight management are first-line for knee osteoarthritis, with physiotherapy to get a programme suited to you. Topical NSAIDs are effective for knee and hand OA with fewer systemic effects than oral versions. Oral NSAIDs work — celecoxib beat placebo in GAIT — but carry genuine gastrointestinal, cardiovascular and kidney risks and need medical advice. Walking aids and appropriate footwear reduce load and are consistently underused. None of that is as easy as a capsule, and all of it has better evidence.
Is it safe to take anyway?
Glucosamine and chondroitin have a good general safety record, which is why taking them despite the evidence is a defensible personal choice rather than a reckless one. Speak to your doctor first if you take warfarin, have diabetes, or have a shellfish allergy — most glucosamine is shellfish-derived, though alternatives exist. If you do try it, run it as an experiment: check the doses against 1,500 mg and 1,200 mg, set a stop date around 12 weeks, don’t change anything else at the same time, and actually stop if it hasn’t helped.
Your checklist
- Know that GAIT tested full doses for 24 weeks and found no benefit vs placebo
- Remember the 60.1% placebo response when reading any testimonial
- Note that celecoxib did beat placebo — the trial could detect real effects
- Treat bioavailability claims as separate from efficacy claims
- Check each ingredient against its studied dose, not the bottle’s total
- Avoid proprietary blends that make that check impossible
- Read “FDA-approved facility” as registration, not endorsement
- Look for third-party testing (NSF, USP) as the real quality signal
- Check whether a review site cites GAIT before trusting its ranking
- Put exercise and weight management first — they’re first-line
- Ask about topical NSAIDs, which are underused
- If you try it: full dose, 12-week limit, change one thing
- Tell your doctor if you take warfarin or have a shellfish allergy
Medical disclaimer. This article is general information and not medical advice. Persistent joint pain should be assessed by a doctor — osteoarthritis is not the only cause, and inflammatory arthritis, infection and injury need different treatment. Seek prompt medical attention for a joint that is hot, red, swollen, or accompanied by fever. Do not start oral NSAIDs without medical advice; they carry gastrointestinal, cardiovascular and kidney risks. Speak to your doctor before taking glucosamine if you take warfarin, have diabetes or have a shellfish allergy. Get professional advice before beginning exercise for a painful joint.
Affiliate disclosure. Some links on this page are affiliate links. As an Amazon Associate we earn from qualifying purchases. We have not recommended the product this article reviews, or any product in its category.
