Complementary therapies: what the guidelines actually say

1. The same therapy, different answers

This is the part no general wellness page will tell you. The recommendation depends on which arthritis you have. Massage is the clearest example: it is recommended against in osteoarthritis and for in rheumatoid arthritis and axial spondyloarthritis. That is not a mistake, and it is explained further down.

"Conditional" means most people should be offered it, but it is a decision to make with your clinician and your own preferences count. "Strong" means almost everyone should follow it. The certainty grade is the guideline's own confidence in the evidence, and where it says very low, that is worth knowing: it means the recommendation rests on weak studies, not that the therapy is proven.

2. Why massage is a no in osteoarthritis and a yes in inflammatory arthritis

The guidelines are not contradicting each other. They asked different questions and demanded different proof.

In osteoarthritis, the ACR looked for benefit on outcomes specific to OA and did not find it. Its stated reasoning is that the massage studies suffered from a high risk of bias, included small numbers of patients, and did not demonstrate benefit for OA-specific outcomes. In rheumatoid arthritis, the ACR accepted very low certainty evidence that massage improves pain and judged that worth a conditional yes, because the harm and burden are low.

There is a detail in the osteoarthritis guideline worth repeating, because it is unusually honest. The patient participants on the panels disagreed. The guideline records that they noted some studies showed positive outcomes with minimal risk, and felt strongly that massage was beneficial for symptom management. The panel recommended against it anyway, on the evidence. So if massage helps your knee, you are not doing anything wrong, and you are in the company of the patients who sat on that guideline panel. It just is not something the evidence can currently support recommending.

3. The two safety warnings

Chiropractic and rheumatoid arthritis. The ACR conditionally recommends against chiropractic spinal adjustment for managing RA. What makes this different from the other conditional recommendations is the reasoning: no studies met the eligibility criteria at all, so the recommendation is made in the absence of evidence, on the grounds of the potential cervical spine complications that can occur. RA can affect the joints at the top of the neck, and that is why a forceful adjustment there is a different proposition than it is in someone without RA.

Spinal manipulation and axial spondyloarthritis. This is the strongest recommendation on this page: the ACR strongly recommends against it. A spine affected by axSpA can be stiff or fused and is more prone to fracture. The same guideline strongly recommends that people with a fused or markedly stooped spine wear a medical alert bracelet, so that first responders know to treat the spine as brittle after a fall or crash. That tells you how the guideline views the risk.

If you are on a medicine that suppresses your immune system, mention it before any treatment that breaks the skin, including acupuncture and dry needling, and mention it to the practitioner as well as to us.

4. Where there is simply no guidance

For psoriatic arthritis, we checked the GRAPPA 2021 treatment recommendations directly: they do not mention yoga, tai chi, acupuncture, massage, or mindfulness anywhere. That is not an oversight on our part; there is no graded recommendation to report. The same is broadly true for lupus and myositis, where exercise is addressed but these specific therapies are not.

That does not mean they do not work. It means nobody has graded them for those conditions, and anyone who tells you otherwise is going beyond the evidence. If you have PsA and yoga helps, the RA and axSpA recommendations are the closest reasonable guide, and that is a conversation to have with your team rather than a rule to follow.

Dry needling is not the same as acupuncture, and no rheumatology guideline we have read names it at all. There is no recommendation for or against it in any of these conditions.

5. What to do with all this

Tell us what you are doing. Not so we can talk you out of it: most of these carry a conditional yes, and a conditional yes means your preference genuinely counts. We ask because two of them have a neck-related warning, because anything that breaks the skin matters if you are immunosuppressed, and because if something is helping we would rather know what is working for you.

If you want one place to start, tai chi is the only therapy on this page with a strong recommendation behind it, for knee and hip osteoarthritis. For inflammatory arthritis, movement in general has better evidence than any of the passive treatments here, and the axSpA guideline says so directly by preferring active exercise over passive treatments such as massage, ultrasound and heat.