Magnet therapies which are claimed to cure conditions ranging from back pain to cancer have no proven benefits, according to a team of US researchers.
A study indicated wrist pain was not aided by magnetic therapy
Sales of the so-called therapeutic devices, which are worn in bracelets, insoles, and wrist and knee bands, top $1 billion worldwide, they said.
But a major review showed no benefits, a British Medical Journal report said.
The team also warned self-treatment with magnets risked leaving underlying medical conditions untreated.
Professor Leonard Finegold of Drexel University in Philadelphia and Professor Bruce Flamm of the Kaiser Permanente Medical Center in California said turning to magnetic therapies could also cause "financial harm".
"Money spent on expensive and unproved magnet therapy might be better spent on evidence based medicine," their report said.
"Sadly some advertisers even claim that magnets are effective for cancer treatment and for increasing longevity; not surprisingly, these claims are unsupported by data."
The team reviewed about 10 earlier studies into the effectiveness of magnetic therapy.
It said many "controlled" experiments used to show the worth of such treatments were suspect because of the difficulty of hiding the presence of a magnet to those involved.
Patients were given dummy magnets and real magnets, but because the real magnets would stick to keyrings in their pockets and other metal items, they could tell if they were part of the real study or the control.
An example of this was a study in chronic pelvic pain which reported improvement.
But the researchers admitted the ability to ensure patients did not know if they had the real or dummy magnets was compromised.
The team does refer to one study on the effects of carpal tunnel syndrome - a painful wrist condition - in which the magnets and the sham treatments were boxed so they could not be identified.
In this, they said, there was no statistical difference between patients with real and sham magnets, with both reporting an improvement in their condition.
The study concludes: "Magnets are touted by successful athletes, allowed to be widely advertised, and sold without restrictions, so it is not surprising that lay people think that claims of therapeutic efficacy are reasonable."
Dr Max Pittler, research fellow in complementary medicine at the Universities of Exeter and Plymouth, said the study suggested static magnets which were most often used to treat pain of different origins were not likely to be effective.
He said: "Although the authors cite only selected studies to back up their statements, also a systematic assessment comes to the conclusion that the evidence is not compelling for the effectiveness of static magnets for reducing pain above non-specific effects."
Debbie Shimadry, director of World of Magnets, said use of magnets was not a cure, but a treatment that was very effective for relief of symptoms of joint-related disorders.
She said: "It is important to have the right type of magnet of the right strength, placed in the right area.
"Magnetic bracelets worn around the wrist are not going to allievate pain all over the body - they need to be placed directly over the area of injury for the right length of time."