Does Tai Chi Work for Pain and Mobility?
Tai chi has an image problem. To anyone who has only seen it from a distance — a group moving very slowly in a park at seven in the morning — it looks too gentle to do anything. That impression is wrong, and it is wrong in an interesting way: tai chi has been tested in proper randomised trials against real comparators, and for several kinds of pain it holds up as well as the exercise programmes people take more seriously.
It is not a cure, and it does not help everything it gets credited with. But the honest summary is more favourable than most people expect.
What it actually is
Tai chi is a Chinese martial art practised, in its health form, as a sequence of slow continuous movements. You stay standing, weight shifting from one leg to the other, knees soft, arms moving in wide unhurried arcs, breathing steadily. Nothing is fast, nothing is impact, and nothing is held long enough to be a strain.
Physically it is more demanding than it looks. Moving slowly through a partial squat while transferring your weight is sustained single-leg loading — the quadriceps and hips work continuously rather than in bursts, and the constant weight transfer keeps the balance system busy the entire time. That combination is unusual. Most gentle exercise gives you one or the other.
The attention component is real but often oversold. You do have to concentrate on what your body is doing, which crowds out other things for an hour. Calling it “meditation in motion” is fair; calling it a spiritual practice you must buy into is not. Plenty of people do it purely because their knees hurt less afterwards.
Where the evidence is genuinely good
Knee osteoarthritis
This is the strongest case. A randomised trial published in the Annals of Internal Medicine in 2016 compared twelve weeks of tai chi against standard physical therapy in people with knee osteoarthritis and found the two produced similar improvements in pain and function — with tai chi doing somewhat better on depression and quality-of-life measures.
That result is reflected in guidelines. The American College of Rheumatology, in its guideline for hand, hip and knee osteoarthritis developed with the Arthritis Foundation, gives tai chi a strong recommendation for knee and hip OA. Strong recommendations in that document are not handed out generously; most of what is on the shelf at a pharmacy does not get one.
Fibromyalgia
A trial in The BMJ in 2018 compared tai chi with supervised aerobic exercise — the intervention usually recommended first for fibromyalgia — over 24 weeks. Tai chi did as well or better, and the groups doing it twice a week or for longer did best. For a condition where exercise is essential and also the thing people find hardest to tolerate, an option that is easier to stay with matters more than a marginal difference in effect size.
Balance and falls
The falls-prevention evidence is the most established of all, and it is the reason public health programmes fund tai chi classes. Trials consistently show reductions in fall rates among older adults, and a purpose-built programme called Tai Ji Quan: Moving for Better Balance outperformed both stretching and conventional multimodal exercise in a randomised trial of older adults at high risk of falling.
This deserves more weight than it usually gets from people whose main complaint is pain. If pain has made you cautious on stairs or unsteady on uneven ground, the fall you don’t have is worth more than any amount of pain relief.
Back pain
The evidence here is real but thinner — modest improvements in pain and disability for chronic low back pain, from a smaller and more variable set of trials. Reasonable to try, not something to promise.
Where it doesn’t deliver
Tai chi will not change the structure of an arthritic joint, a narrowed spinal canal or a damaged disc. Claims that it improves bone density are not well supported. It is not a treatment for inflammatory arthritis in place of medication, and nobody serious suggests it is. And its effect on pain, while real, is in the same modest range as most exercise interventions — meaningful across a population, variable in any one person.
The other honest caveat is blinding. You cannot conceal from someone that they are doing tai chi, so expectation contributes to the results, as it does in every trial of exercise. The response to that is not to dismiss the findings but to notice that the comparators — physical therapy, aerobic exercise — carry the same problem and tai chi still matched them.
The styles, and which ones suit a painful body
“Tai chi” is not one thing, and walking into the wrong class is the most common reason people conclude it isn’t for them. The traditional family styles differ considerably in how hard they are on knees and hips.
| Style | Character | Suits you if… |
|---|---|---|
| Yang | The most widely taught. Large, slow, even movements; moderate stance depth. | You want the default. Most classes you find will be this. |
| Sun | Higher, more upright stances, shorter steps, a following step that keeps the feet close together. Least demanding on the knees. | You have arthritic knees or hips, limited range, or balance concerns. |
| Wu | Smaller frame, compact movements, a characteristic forward lean. | Space is tight and large arm movements are uncomfortable. |
| Chen | The oldest family style. Low stances, spiralling movements, occasional bursts of speed and stamping. | You are fit and want a martial workout. Not a starting point for painful joints. |
Alongside the family styles there are shortened sequences built for specific purposes, and these are usually the better entry point:
- The simplified 24-form — a Yang-based sequence standardised in the 1950s and by far the most commonly taught form in the world. Takes about five minutes to perform and a few months to learn. If a class advertises “short form”, this is usually it.
- Tai Chi for Arthritis — twelve movements drawn from Sun style, designed specifically for people with joint pain and taught in a structured programme with a seated variation. This is the one recommended by arthritis organisations, and the Tai Chi for Health Institute lists certified instructors.
- Tai Ji Quan: Moving for Better Balance — eight forms adapted from the 24-form and rebuilt around balance training, which is the programme behind much of the falls research. Its programme site explains the structure and where it is delivered.
- Seated tai chi and qigong — if standing for thirty minutes is not currently possible, the upper-body movement and breathing can be done from a chair. It is not equivalent, since you lose the balance and leg-loading component that does much of the work, but it is a genuine starting point rather than a consolation prize.
If you are choosing blind, a Sun-style or Tai Chi for Arthritis class is the safer bet. If the only class near you is Yang style, that is still fine — ask the instructor to let you work at a higher stance.
Talk to your doctor or physiotherapist before starting if any of these apply. Recent joint replacement or spinal surgery; dizziness, fainting or a balance disorder; severe osteoporosis, where the twisting in some forms needs modifying; unstable angina or uncontrolled blood pressure; or a knee or hip that gives way. None of these rules tai chi out — most of them change how high you stand and what you hold onto — but they are worth naming before your first class rather than after it.
How much, and how long before you judge it
The trials that found benefit generally used sessions of about an hour, once or twice a week, for twelve weeks or more, with practice at home in between. That is the realistic dose. People who go twice and conclude it does nothing have not tested it.
Expect the first month to be frustrating in a specific way: you will spend it learning where your feet go, not feeling any better. The physical benefit arrives after the choreography stops taking all your attention. Knowing that in advance is most of what gets people through it.
Ten or fifteen minutes at home on the days between classes is worth more than a longer session once a week. Slow movement is easy to fit into a small space and needs no equipment, which is the practical argument for it over almost anything else on this list.
Finding a class in Canada
Tai chi is unusually well established here. The Taoist Tai Chi Society was founded in Toronto in 1970 and now runs classes in every province, often in church halls and community centres, on a membership model that is inexpensive by the standards of any supervised exercise. Its location finder is the fastest way to see what exists near you.
Beyond that, look at municipal recreation centres, which frequently run older-adult tai chi at a subsidised rate; seniors’ centres; and hospital or community health centre rehabilitation programmes. Arthritis Society Canada is a useful starting point for arthritis-specific programmes and local support.
On cost: tai chi classes are not an insured health service, so you pay for them, but they are typically a fraction of what private physiotherapy costs per session in Canada — and unlike physiotherapy, the price does not rise with the number of visits. For a long-term condition, that difference compounds.
Questions worth asking before you sign up:
- Which style, and is there a shorter or seated form available?
- Is there a beginners’ class, or does everyone learn in one group?
- Has the instructor taught people with arthritis, joint replacements or balance problems before?
- Can you stand near a wall or chair while you are learning?
- Can you try one class before committing to a term?
An instructor who answers the third question with a shrug is not the right one for you. An instructor who immediately offers a modification is.
The best exercise for a long-term condition is not the one with the largest effect in a trial. It is the one you will still be doing in a year.
So, does it work?
For knee and hip osteoarthritis, yes — well enough that a major rheumatology guideline recommends it strongly. For fibromyalgia, yes, at least as well as the aerobic exercise usually prescribed first. For balance and falls, yes, and that may be the most valuable part. For back pain, probably, modestly. For everything else it gets credited with, the evidence thins out quickly.
What makes it worth considering is not that it beats other exercise — mostly it matches it. It is that it is low-impact, needs no equipment, can be scaled from a chair to a deep stance, is cheap and widely available in Canada, and is done in a room with other people, which is not a trivial feature of something you are meant to keep doing for years. If your pain has made ordinary exercise feel like a negotiation, that combination is worth an hour of your week for three months to find out.
