Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Magnetic insoles for foot pain

Randomised trial

Here's a neat conversation piece. How much is spent in the USA every year on magnetic devices to treat pain? Answer is $500 million, with a total worldwide market to date above $5 billion. To put that into some sort of perspective, that $500 million is just half the annual sales that the pharmaceutical industry defines as a 'blockbuster'.

And what do you think is the evidence for magnets affecting pain? You guessed it. None. There is a trial in a Cochrane review of interventions for plantar heel pain [1], and that was negative, and poor. A new, well-conducted, randomised trial provides a powerful negative, and a great example of trial design.

Randomised trial

The trial was conducted in patients responding to an advertisement for people with foot pain for at least one month, present on more days than not, and aggravated by standing or walking. For inclusion people had to be over 18 years, with pain of at least moderate intensity, and characteristics of plantar heel pain on examination. Exclusions were sensible, especially possible participants with other diagnoses.

Magnetic and non-magnetic insoles were identical in appearance and manufacture, and were cushioned soles with embedded foil which was either magnetised or not magnetised. The manufacturer provided insoles using a random tracking code, and they were then mixed in a large box.

Participants were asked to use the insoles for at least four hours a day for at least four days a week for eight weeks. Information was collected at baseline, and at four and eight weeks. Outcomes were pain, response to treatment, and other outcomes using diaries to record hours that insoles were worn, and daily pain intensity.


Randomised were 101 patients, 57 with magnetised and 44 with non-magnetised insoles. Six were lost to follow up, three in each group. Use of insoles was above 80%. At baseline patients in the two groups were very similar. Current treatments included use of NSAIDs or paracetamol in 60-70%, but also ice, elevation, stretching and massage.

There was a fantastic reduction in morning pain intensity with the magnetised insoles (Figure 1). Unfortunately there was exactly the same reduction with placebo. The proportions of patients who felt all or mostly better at four and eight weeks were the same with magnetised and non-magnetised insoles (Figure 2).

Figure 1: Pain scores over eight weeks

Figure 2: Patients all or mostly better at 4 and 8 weeks

There were no other differences between magnetised and non-magnetised insoles. About half in each group guessed correctly which insole they had, so blinding was maintained. Insoles were checked to ensure that they were correctly magnetised. Belief or non-belief in the utility of magnetised insoles made no difference to the result.


A beautiful study demonstrating pretty conclusively that magnetised insoles do not help plantar heel pain. It is another example showing that with complementary therapies, the better the trial, the more negative the result.

The manufacturer of the insoles funded the study, which was brave of them. They might want to take away that perhaps it was the use of the insoles, rather than the magnetic field from them, that helped. Figure 1 certainly seems to show that. The trouble is that patients may have entered the trial because, in a chronic fluctuating disorder, the pain was particularly bad at the time. If it then got better because it was fluctuating, reduced pain is just what would be expected.

There is also a complex philosophical question. We protect people from ineffective and unsafe chemicals, but we do much less for crackpot devices. OK, we don't know that magnetised insoles are unsafe, but no woman who could possibly be pregnant could be enrolled in the trial. And isn't parting with $500 million for no benefit an adverse event of a sort? Bandolier could find useful and interesting things to do if anyone has the odd $500 million going spare, and we wouldn't turn our nose up at smaller amounts, either.


  1. Crawford F, Thomson C Interventions for treating plantar heel pain. In: The Cochrane Library, Issue 3, 2003. Oxford: Update Software
  2. MH Winemiller et al. Effect of magnetic vs sham-magnetic insoles on plantar heel pain. A randomised controlled trial. JAMA 2003 290: 1474-1478.

previous or next story