Powder made from the 'velvet' on antlers was tested
Most complementary therapies used by people with rheumatoid arthritis are not effective, a study has suggested.
The Arthritis Research Campaign looked at the scientific evidence available for over 40 treatments.
Two thirds of treatments for rheumatoid arthritis and a fifth of treatments for osteoarthritis were found to be ineffective by the researchers.
The Arthritis Research Campaign said it wanted people who used the therapies to know what evidence was available.
Rheumatoid arthritis is caused by inflammation of the lining (synovium) of the joints.
Osteoarthritis is caused by the breakdown of protective tissue called cartilage in the joints. Inflammation results when the unprotected bones of the joint begin to rub together.
It most commonly affects the joints of the fingers, knees, hips, and spine.
In total, 60% of people with arthritis are thought to use some form of complementary medicine.
The researchers looked at compounds taken by the mouth or applied to the skin.
Effectiveness is measured by improvements in pain, movement or general well-being.
When the researchers examined treatments for rheumatoid arthritis, they found 13 out of 21 complementary medicines were shown to have no or little effect based on the available evidence.
The 13 were: antler velvet powder, blackcurrant seed oil, collagen, eazmov (a herbal mixture), feverfew (herb), flaxseed oil, green-lipped mussels, homeopathy, reumalex herbal mixture, selenium, the Chinese herb tong luo kai bi, vitamins A, C and E, and willow bark.
However, fish body oil was given five out of five in the report, for being effective in reducing joint pain and stiffness.
In addition, six out of 27 treatments for osteoarthritis were shown to have little or no effect based on the available evidence
Capsaicin gel, made from chilli peppers, proved most effective in relieving pain and joint tenderness.
But the effectiveness of glucosamine, a popular supplement used by people with OA which costs around £10 a month, which researchers have previously said was ineffective, again called into question.
For fibromyalgia, which causes widespread pain in muscles and joints, only four products were assessed, none were found to be highly effective with three medicines scoring two out of five, and the fourth just one.
The researchers also examined how safe compounds were.
One - thunder god vine, a traditional Chinese medicine - was given a "red" classification, meaning there were serious safety concerns.
A quarter of the compounds were given an "amber" safety classification, because there were some reported side-effects.
The team said they were unable to evaluate the effectiveness of 36 therapies, including basil, green tea, sarsaparilla and St John's Wort because there was insufficient data.
Professor Gary Macfarlane, from the University of Aberdeen, said while different things worked for different people, "it is useful to also have the scientific evidence available and just as important to know how safe we think they are to use."
Professor Alan Silman, the Arthritis Research Campaign's medical director, added: "We didn't start this saying this was our opportunity to knock complementary medicines.
"The message is not 'don't take them'. The message is 'if you are going to take them, be aware of what the level of evidence is'."
Dr Peter Fisher, clinical director of the Royal London Homeopathic Hospital, said the report focused on tablets and preparations applied to the skin, missing out therapies such as acupuncture and osteopathy.
"I think what really comes across in this report is how sorely under-researched this area is," he said.
Jane Gray, president, of the National Institute of Medical Herbalists added: "This report is a commendable attempt to provide information on self help products for osteo and rheumatoid arthritis."