Lying on a bed, allowing someone to stick needles into your flesh up to a depth of four centimetres involves some courage. Yet for Emma, 24, an expatriate living in Oman, any small discomfort in her acupuncture sessions is nothing compared with the pain she was suffering before.
Emma's problems began during an operation, when gas pumped into her stomach led to a severe backache that simply wouldn't go away.
"I had nothing to lose", she says of her decision to try acupuncture. Physiotherapy had achieved nothing, the painkillers she had been prescribed gave her stomach cramps and the pain was so bad that she could no longer bend over to tie her shoelaces.
As it turned out, the treatment appeared to work miraculously.
"When I woke up the next day I reached for the painkillers and realised that nothing hurt. There was a kind of numb ache where it had once been sharp and constant. That ache decreased over a few days and even the need for a few ibuprofen after a day on my feet was gone within a week. I had been so worried that I was going to be hunched and needing help forever, but I felt fine."
Emma is one of many with anecdotal evidence of acupuncture's success in treating pain. But what about hard, scientifically tested evidence?
The past 10 years have seen hundreds of clinical trials on acupuncture as a treatment for pain. Many are inconclusive and some are conducted with less than perfect rigour, but they nevertheless make fascinating reading.
One of the largest studies to date - a German trial into acupuncture as a treatment for chronic low back pain, published in 2007 - involved 1,162 patients who had been suffering low back pain for an average of eight years. A third were treated with traditional acupuncture. A third underwent "sham" acupuncture, in which the needles were inserted in non-acupuncture points to a depth of 1-3 mm (as opposed to the usual 5 to 40mm). A further third were given conventional therapy involving treatments such as physiotherapy, massage and heat therapy.
The results were startling. Almost half of the patients given "true" acupuncture reported improvement compared with only 27 per cent of those given conventional treatment. Intriguingly, there was no significant difference between the success rates of those given true acupuncture and those given sham acupuncture.
A more positive outcome for "true" acupuncture came in the form of a German study on 424 patients suffering from chronic shoulder pain, published in 2010. It randomly assigned patients to true and sham acupuncture and to conventional orthopaedic treatment. In this study, the true acupuncture outperformed both the sham treatment and the conventional treatment.
Debate has raged over the conclusions to be drawn from these and other studies, with critics arguing over their methodology and interpretation.
Some, such as Edzard Ernst, professor of complementary medicine at the Peninsula Medical School at the Universities of Exeter and Plymouth, are broadly dismissive of the results.
"I do sympathise with decisions to make the best use of acupuncture while hoping that research will bring more clarity," he says. "My feeling is, however, that the rigorous trials will eventually demonstrate that it is an elaborate placebo."
Advocates of acupuncture are slower to jump to conclusions.
"There are paradoxes here, various possible explanations," says Mark Bovey, a research officer at the British Acupuncture Council. "Nobody knows which is correct, or more correct than others. There is no basis to conclude that acupuncture must be acting as a placebo - it presumably does to some extent, but we don't know how much."
Another difficulty with performing clinical trials on acupuncture is that it is a holistic therapy. It is based on the notion of qi, or vital energy, which is said to flow around the body and which can be blocked by a number of factors including stress, poor nutrition, infection or injury.
Acupuncture is said to focus on all factors that contribute to disease, not just the presenting symptoms, and some of its advocates are cautious about submitting it to traditional trials. The British Acupuncture Council, for example, asserts that it "doesn't neatly fit the prevalent research protocols that have been developed in response to symptom-based treatments".
Add to this the fact that pain - and particularly chronic back pain - is a complex condition that can be affected by stress and other psychosocial elements, and the fact that by the time patients submit to trials their perceptions of conventional therapies may have been affected by years of failed treatments (whereas their expectations of acupuncture may be more positive), and it's hardly surprising that the trials are controversial.
Either way, acupuncture remains one of the most credible and mainstream alternative therapies; the UK's National Institute for Clinical Excellence, for example, suggests that patients be offered a course of acupuncture as a treatment option for low back pain.
As for Emma, her visits to the acupuncturist continue to make both practical and economic sense.
"I don't know how it works, and to be honest it doesn't matter to me", she says. "Why question something if it does the job well? One session of acupuncture in Oman costs me 5 Omani rials [Dh50], while if I ignore it and have to buy painkillers they can cost RO2-RO3 every few weeks, so it works out more cost effective to take the acupuncture."
Whatever you think of the treatment, it's hard to argue with the maths.