Older people

Copper bracelets and arthritis

“Copper bracelets and magnetic wrist straps are useless for relieving pain in people with arthritis,” BBC News has reported. It said that the first tightly controlled study of the interventions found no benefit in treating pain or stiffness from the condition.

This report is based on a robust study design and shows good evidence that magnetic wrist straps and copper bracelets have little or no effect on pain, physical function or stiffness in osteoarthritis.

According to this study, any benefit these devices may have, which the BBC reports can costs between £25 and £65, may be explained by the placebo effect.

Where did the story come from?

The research was carried out by Dr Stewart J. Richmond and colleagues from the University of York, Durham University, the University of Hull and other medical institutions in Hull, UK.

It was funded by a grant from the Wolds Primary Care Research Network (WoReN). The bracelets were supplied free of charge by a manufacturer, MagnaMax Healthcare (Ontario, Canada).

The study was published in the (peer-reviewed) medical journal Complementary Therapies in Medicine.

What kind of scientific study was this?

This study tested the effectiveness of a typical magnetic wrist strap sold for reducing pain and stiffness and for improving physical function in patients with osteoarthritis. It was a randomised, double-blind, placebo-controlled crossover trial.

The study recruited 45 people with osteoarthritis from GP practices in Yorkshire between February and June 2005. Patients were over 40 years old and were receiving active treatment for pain with either prescribed NSAIDs or opioid painkillers (e.g. paracetamol or codeine).

The participants were randomly allocated to one of four sequences of treatment. Each of the four sequences featured the same devices being worn, but in different orders, for a total of 16 weeks. Each device was worn for a minimum of eight hours a day for four weeks. The four treatments were:

  • Standard (the experimental device): a MagnaMax wrist strap. These commercially available bracelets have a leather strap with a magnetic insert that is worn directly against the skin. The researchers demonstrated that these bracelets had a strong magnetic field.
  • Attenuated (i.e. weakened): a MagnaMax wrist strap with a much weaker magnetic field.
  • De-magnetised bracelet (a dummy):  A MagnaMax wrist strap that was completely non-magnetic.
  • Copper bracelet (the placebo): these were worn on the wrist, weighed 13g and had no magnetic properties.

The participants completed questionnaires at the beginning of the study and at the end of each four-week treatment phase. The questionnaire measured experience of pain on a number of different scales, but the main one used in the analysis was the pain subscale of a wider questionnaire called WOMAC Osteoarthritis Index 3.1.

In crossover studies, each participant is their own ‘control’ because they all have the intervention under study and the placebo treatment. After 16 weeks, the researchers compared the changes in pain scores across the treatment groups. They adjusted for treatments (from self-report diaries, prescription records and pill counts) and for time. Complete data was available from 42 patients for analysis.

What were the results of the study?

The researchers found no difference between the four devices in change in pain outcomes as measured on the WOMAC Osteoarthritis Index, (their primary measure of outcome) or on the other measurements of pain experience (which were considered secondary measures of pain).

When pain was measured on the PRI sensory pain subscale (a secondary measure), the standard magnetic wrist strap seemed to reduce pain more than other bracelets, with a reduction of  2.52 points out of a possible 42 points on the PRI sensory scale (95% confidence interval, 4.05 to 0.99 reduction).

What interpretations did the researchers draw from these results?

The researchers conclude that “magnetic therapy, involving the use of a magnetic wrist strap over a period of four weeks, had no statistically significant therapeutic effect among patients with osteoarthritis for the primary pain outcome measure”. They say that there was also no effect on physical function, stiffness or use of medication. Together these results can be taken “as evidence against the use of magnetic therapy”.

The researchers suggest that the reported analgesic benefits associated with the devices may therefore be due to the psychological effects of a placebo.

What does the NHS Knowledge Service make of this study?

This study provides good evidence that magnetic and copper wrist bracelets have no effect on pain, stiffness or physical function in osteoarthritis.  Although the trial was small, the researchers made sure they had enough participants to detect a difference of 25% between the standard bracelet and the placebo devices on the WOMAC A scale. However, the trial may have been too small to detect a significant difference on the other measures used.

The researchers discuss how their results contradict those from a previous study, which looked at similar devices and found that a standard bipolar magnetic wrist strap reduced pain compared to weak or demagnetised devices. The researchers suggest that this previous study had weaknesses that may explain the contradictory findings. For example, there are signs that it had problems with blinding, as it found that a fully magnetised device had a significant effect compared to the demagnetised dummy device, but not to the attenuated (weakened) strap.

Crossover trials of drugs usually need a ‘washout’ period before the participant switches to another treatment group. This is to allow the effects of the previous treatment to pass before measuring the effects of a new treatment. However, the researchers say they did not consider washout periods to be necessary as the study included multiple control devices (attenuated and weakened bracelets and the copper bracelet).

The study did not control for multiple testing. This means that the researchers performed several different statistical tests and did not adjust for the fact that the more tests are performed, the more likely there are to be positive findings by chance alone. They say that this may explain the one significant finding of their study, that pain was reduced by the standard magnetic wrist strap on the PRI sensory pain subscale.

The researchers also highlight that one possible explanation for the lack of effect in their study was the short time (four weeks) of each treatment period. They say that this is likely to affect the copper bracelet treatment most, but argue that four weeks should have been sufficient for the magnetic wrist strap to have had an effect.


NHS Attribution