Published: 22/07/2004, Volume II4, No. 5915 Page 22 23

Patient choice may be highlighting public interest in complementary therapies and alternative medicine but evidence of their efficacy is still lacking, says Edzard Ernst

Complementary and alternative medicine (CAM) remains highly controversial.While consumers have voted with their wallets in its favour, it has remained on the periphery of mainstream medicine and has been neither scientifically accepted nor financially integrated into the NHS.

But the climate may be changing. For the first time, the National Institute for Clinical Excellence recently considered CAM and stated that multiple sclerosis patients 'should be informed that there is some evidence to suggest the following items might be of benefit, although there is insufficient evidence to give more firm recommendations: reflexology and massage, fish oils, magnetic field therapy, neural therapy, massage plus bodywork, tai chi, multi-modal therapy'.

1This change has been supported by prominent champions who use the media to demand that alternative medicine should be available to all on the NHS.

The logic behind this argument was recently encapsulated by Prince Charles in The Times: 'If [complementary and alternative treatments] are proven to work, they should be made more widely available.'

But what does 'proven to work' actually mean? For some enthusiasts the answer seems disarmingly simple.

Conservative MP for Bosworth David Tredinnick, treasurer of an all-party parliamentary CAM lobby, once injured his wrist, to which he applied a mixture of homeopathy, acupuncture and aromatherapy.He stated in a letter to The Daily Telegraph: 'I do not know which of these affected me, but I am better. I do not need to be a burden on the health service.'

Though anecdotes can be powerful, however, few healthcare experts would contend that they amount to proof. A wealth of evidence exists to show that many people use CAM. The most recent survey of 1,000 UK citizens was carried out in December 2003 and published recently. Its results suggest that 25 per cent of the population have tried CAM and that 'only one in 10 people who has tried complementary therapy thinks it has made no difference'.

But what does this data really demonstrate? Does it prove that CAM works? In any branch of medicine, including CAM, 'proven to work' can only mean one thing: evidence complying with current standards to demonstrate without reasonable doubt that a given CAM intervention for a given medical condition does more good than harm.

This is where things often become complicated.

Chiropractic spinal manipulation as a treatment of back pain is perhaps the most accepted form of CAM. It may therefore serve as a suitable example to demonstrate some of the issues at hand.

The best evidence about the effectiveness of any treatment is usually provided by Cochrane reviews. A recent state-of-the-art Cochrane review of spinal manipulation for low back pain included a total of 39 randomised clinical trials.Meta-regression models were used to assess the effects on acute or chronic back pain, as well as acute short and long-term pain and function.

For patients with acute back pain, spinal manipulation proved to be superior only to sham therapy or interventions known to be ineffective or harmful. It had no advantage over care by GPs, analgesics, physical therapy, exercise therapy or back school.

For patients suffering chronic back problems, the results were similar. The authors concluded that 'there is no evidence that spinal manipulation therapy is superior to other standard treatments for patients with acute or chronic low back pain'.

The complexity of the issues increases further if one also considers safety aspects. The risks associated with spinal manipulation include stroke and death.

Chiropractors are adamant that such events are rare, but the truth is that, due to under-reporting, nobody can be quite sure.

Chiropractors also argue that such complications only occur in association with upper spinal manipulation, which is unrelated to lower back pain. But this argument is only superficially convincing. Chiropractors view the spine as a functional entity and will manipulate the upper spine when they diagnose a misalignment of its structure, even if the patient complains of problems in the lower back.

We must therefore add safety to the already complex issue of efficacy and ask, does CAM generate more harm than good? We really need to know how the riskbenefit balance of one therapy compares to other therapies. The Cochrane review shows that spinal manipulation is not more effective than other treatments for back pain. Some of these options are, however, not associated with significant risks: exercise therapy or back school, for example. Does this mean that such approaches are preferable?

Promoters of CAM are optimistic about the economic implications, but the evidence sadly does not support this. Specifically regarding chiropractic spinal manipulation for back pain, a recent cost evaluation from the US suggested that it is about twice as expensive as medical care.

2Surveys may suggest that chiropractic is a highly popular treatment for painful backs, but are we convinced it is superior, safer or cheaper than competing treatments?

Regardless of the complexity of the issues, CAM enthusiasts insist that the current agenda is to find a way of integrating CAM with conventional approaches to medicine. Based again on the existing evidence, however, this conclusion needs to be qualified.

The gaps in our knowledge are, in most cases, simply too large to justify advocating integration of CAM into routine healthcare. The overriding implication at this stage must be to conduct more research into those areas of CAM that so far have shown most promise.

At the same time we must apply well-established principles of evidence-based medicine.We need to know which forms of CAM generate more good than harm and we require evidence of acceptable quality.

Double standards are not likely to help anyone - least of all the patient.

Professor Edzard Ernst holds the chair of complementary medicine at Peninsula Medical School, Plymouth, and is co-author of The Desktop Guide to Complementary and Alternative Medicine .For details of PMS research on this subject e-mail nicola. watson@pms. ac. uk.

Evidence-based information is available at www. ex. ac. uk/FACT


1 National Institute for Clinical Excellence.Management of multiple sclerosis in primary and secondary care.NICE,2003.

2 Stano M et al.Chiropractic and medical care costs of low back care.American Journal of Managed Care 2002; 8(9):802-9.

Key points

Complementary medicine can be unsafe and expensive.

Even accepted therapies, such as chiropractic spinal manipulation, rest on doubtful evidence.

There are insufficient grounds to justify integrating these therapies into routine healthcare.