Complementary medicine encompasses a confusing array of therapies (Table 1). It is often defined by what it is not (e.g. not based on science, not taught in medical schools, etc.) rather than by what it is. A 'positive' definition states that complementary medicine is:

diagnosis, treatment and/or prevention which complements mainstream medicine by contributing to a common whole, by satisfying a demand not met by orthodoxy or by diversifying the conceptual frameworks of medicine.

Complementary medicine has become an important topic, not least because of its popularity in the general population and its prevalent use for pain control. Between 20 per cent and 75 per cent of all adults in industrialized countries have used at least one type of complementary medicine within the past year. Almost all patients suffering from chronic pain try some form of complementary medicine.

Reasons for popularity
The reasons for the popularity of complementary medicine are varied and complex; they change with time, space, and from one therapy to another. They also differ from one individual to the next - for example, a pain patient will have different motives from someone who is 'worried well'. Thus no single determinant exists but a broad range of interacting positive and negative motivations can be identified (Table 2). Some of these amount to a biting criticism of our modern healthcare system.

Obstacles to research
Many providers of complementary medicine are convinced that their therapy defies scientific research. They argue that it is individualized, holistic, intuitive, and so on, and therefore not amenable to study by standard approaches of medical appraisal. There have been many calls call for a 'paradigm shift' in research.

Usually these arguments are based on a series of misunderstandings, and often the problems can be resolved by clearly defining the research question and subsequently identifying the optimal research tool. If this question relates to testing the effectiveness of a treatment, randomized controlled trials usually provide the least biased method for finding a reliable answer.

Why then is there so little research into complementary medicine? A particular problem is that the area lacks both a research tradition and a research infrastructure. Research career opportunities are limited and the field does not attract experienced researchers. Most importantly perhaps, the orthodox attitude remains sceptical and, as a consequence, the funding of complementary medicine research is pitiful.

Is there any evidence at all?
Opinions dominate complementary medicine, and they can be dangerously misleading. The best way to establish the true value of any therapy is through rigorous research. Moreover, each of the numerous techniques has to be evaluated on its own merits; simple answers or broad generalizations are not possible. Some forms of complementary medicine are safe while others are burdened with risks; some are effective while others may be pure placebos.

Unfortunately, with so little investment in research, very few studies have examined the role of complementary medicine in pain control. For only a few treatments is our knowledge sufficiently advanced to allow preliminary conclusions to be drawn (Table 1). In some cases (for example, herbal medicine) the evidence is positive; in other instances (for example, chelation therapy) it is negative.

Possible mechanisms
The confusing array of treatments makes it obvious that there cannot be one single mode of action. Only with some therapies are we anywhere near to understanding how they work.

For instance, herbal medicine acts through the pharmacological actions of the plant constituents - much like conventional drugs. Acupuncture might work through several complicated neurophysiological pathways. Treatments such as aromatherapy or massage might be effective via a relaxation response.

For the majority of treatments, however, no plausible mode of action has yet been identified.

Cost
Complementary medicine is not cheap. The UK's annual total expenditure on complementary medicine is around £1.6 billion, and practitioners' fees can be considerable.

This level of spending has significant implications not only for the nation's healthcare system but also for our economy in general. The ultimate question is whether the use of complementary medicine increases or decreases overall expenditure in our healthcare system. Self-treatment, it has been argued, may ease the burden on the public NHS. To answer it, one would require reliable data. The most rigorous studies to date show no evidence that the use of complementary medicine reduced overall expenditure.

The power of placebo
Some forms of complementary medicine could be 'ineffective' (in the sense of not being better than a placebo) and still alleviate pain through a placebo response. Some argue that, in such instances, complementary medicine should be used regardless, particularly when its use is not associated with serious risks.

Rigorous research could even be seen as counterproductive. We might, for instance, find little 'hard' evidence in favour of homoeopathy and its availability would thus decrease. Patients who previously benefited from it might in turn no longer do so.

On the other hand, if research really showed that homeopathy helps people through powerful non-specific (placebo) effects, the medical community should start seriously considering the power of placebos. The research question then shifts to how placebo effects might be optimized so that more patients can reliably profit from them.

Conclusion

Complementary medicine is frequently used by patients to alleviate pain. Several forms are beginning to be backed up by good evidence. In many instances, however, it is unclear whether complementary approaches are truly effective, harmful or just a waste of time and money. To fill the considerable gaps in our current knowledge, we urgently require more research - and sufficient funds to adequately perform this work.

Professor Edzard Ernst is Director of Complementary Medicine, Peninsula Medical School, Universities of Exeter and Plymouth, Exeter EX2 4NT, UK.
E-mail: edzard.ernst@pms.ac.uk

Further reading
Ernst E, Pittler M H, Stevinson C, White A R (2001) The desktop guide to complementary and alternative medicine. Edinburgh: Mosby.