Many members of the biomedical establishment would like to contain, curtail, and control the practice of complementary and alternative medicine (CAM). There are a number of ways this can be achieved, ranging from outright suppression to absorbing CAM modalities into the conventional medical system, ideally with the CAM modalities’ own paradigms replaced by a biomedical theoretical framework. The means of achieving this is promotion of the ideal of evidence based medicine, executed within a biomedical framework, with the intention of discrediting CAM modalities if ineffective, or else subsuming them into ‘one medicine’ (meaning biomedicine); in this agenda, inter-paradigm issues are ignored, with the CAM modalities regarded as empirical ‘techniques’ in need of biomedical research to determine the ‘mechanisms’ with which they work, and hence how they should be practised, instead of as valid, autonomous systems. This can be done in the name of putting CAM on a solid scientific, modern, evidence based basis, compatible with current knowledge and the need for rational allocation of resources for medical interventions.
Because it is common for biomedical adherents to condemn CAM systems such as traditional Chinese medicine as unscientific or pseudo-scientific, I will consider these claims briefly. The great physicist Richard Feynman wrote "In general we look for a new law by the following process. First we guess it. Then we compare the consequences of the guess to see what would be implied if this law that we guessed is right. Then we compare the result of the computation to nature, with experiment or experience, compare it directly with observation, to see if it works. If it disagrees, it is wrong. In that simple statement is the key to science" (Gribbin and Gribbin, 1998).
It is apparent to anyone who understands TCM and many other styles of oriental medicine that these do, in fact, work in this way. There is a close correspondence between, say, the TCM concept of the Spleen, various symptoms, tongue body and tongue coat appearance, and pulse. Treatment is rationally applied on the basis of a rationally achieved diagnosis, and each diagnosis and treatment is a mini-experiment to see if the system works. Repeated observations over many hundreds of years have validated this model. It already meets Feynman’s requirements. There is no reason why this process could not be extended to laboratory conditions. Employing Feyman’s definition of science, TCM is a scientific method. Thomas Kuhn (1996) has observed that two different, perhaps competing, paradigms can both be scientific. The mistake that many scientists make is to consider their own paradigm the only valid scientific modality, in effect appropriating the term ‘scientific’. Biomedical adherents who condemn traditional Chinese theory as unscientific do so either in ignorance, or because they have come to equate science exclusively with modern, mainstream science. When they argue that certain issues must be resolved scientifically, this should not imply that the modern scientific set of beliefs and assumptions are the only valid bases for conducting scientific enquiry. Indeed, several eminent scientists such as Bohm (1981) and Capra (1983 and 1997) have criticised the inadequacies of the mainstream scientific paradigm (also see Churchill, 19991*). Rational, rigorous theory and methodology, based on acute observation of the natural world, are not exclusive to the modern scientific paradigm. Traditional oriental medical paradigms do share these characteristics, albeit it from a different standpoint or worldview. This is similar to different languages being capable of serving the functions that languages perform, even though they contain unique concepts and expressive potential. Some might argue that it would be better to have only one common language; the counter argument is that each language has unique characteristics, and that the world is richer for preserving diversity. Kuhn has observed that any particular paradigm is limited in the range of phenomena it can accommodate (1996). There are powerful arguments for a multi-paradigmatic approach in medicine, leading to pluralistic medicine rather than ‘one medicine'.
It would be unethical for a practitioner of any medical modality to recommend his technique while discouraging patients from using another that was known to be superior, assuming such a comparison was possible. Reliable research is necessary for such guidance. With this simple statement come remarkable complexities – how is illness defined and viewed, what is meant by ‘efficacy’, by what criteria do we view an experimental situation? An illustration of this problem was afforded by an alternative practitioner who asked biomedical researchers of a CAM study what steps had been taken to standardise geopathic stress. They hadn’t considered geopathic stress at all, and to them it was irrelevant if not fanciful, but in her paradigmatic framework it was an extremely significant factor, the absence of the consideration of which invalidated the scientific methodology of the research. Another example is afforded by the Dimond et al, 1960 clinical trial in which a surgical technique to improve blood supply to the heart for angina was no more effective than simply making an incision and closing the wound straight away. This example is often used to show the power of placebo effect, even in regard to acupuncture’s effectiveness (White,1998b), but from a traditional Chinese paradigmatic viewpoint, the ‘placebo’ procedure could have had a real effect on the blood and qi of the chest, accounting for the favourable response. The conclusion that the favourable response was a placebo effect rests on prior belief in the exclusive validity of biomedical understanding – ‘there is no (biomedical) explanation why the dummy surgical procedure should have had a good effect, therefore placebo effect was involved.’ It is obvious that many assumptions can influence the design and interpretation of clinical trials. Many biomedical adherents will ridicule geopathic stress and the concept of qi and blood, but that is not ultimately a valid response in many situations in which CAM is being tested. In other words, the underlying assumptions made in research and its interpretation can be all important. In the present climate of evidence based medicine, research into CAM is being and will continue to be done. It will be used to support a number of recommendations that could lead to drastic changes for CAM. CAM professions therefore have a duty to involve themselves in research, not least to avoid incorrect conclusions drawn that could have an unwarranted effect on the practice of those modalities. A notable danger for CAM professions will be a shift to the biomedical paradigm as the basis of their practice, ultimately arising from biomedical assumptions underlying research and its interpretation and use.
Acupuncture – A Scientific Appraisal is a potentially influential book that promotes the idea of research being used to determine the practice of acupuncture. In the introduction Ernst and White boldly state "Since the development of the concept of evidence-based healthcare, therapies must establish their efficacy, safety and cost-effectiveness by means of rigorous studies". In the book’s conclusion, they write "The current evidence of the efficacy of acupuncture could be compatible with several eventual conclusions, each with important consequences for the integration of this therapy within the health services, for the training of practitioners, and for further research. Different outcomes are, of course, likely in different conditions". The different outcomes are:
Outcome 1: If acupuncture does more harm than good, it should be abandoned.
Outcome 2: If sham needling is shown to be just as effective as acupuncture, the safest form of acupuncture should be administered (not necessarily utilising needles), with practitioners trained in effective placebo therapy.
Outcome 3: "Traditional Chinese acupuncture could prove to be superior to western medical (and to sham) for a particular condition. In this case, TCA [traditional Chinese acupuncture] should be integrated into healthcare, and should be the method taught in acupuncture colleges. Research should address the question of optimising methods of diagnosis and treatment, as well as exploring the underlying mechanisms".
Outcome 4: "Western acupuncture could prove superior to TCA (and to sham) for a particular condition. In this case, acupuncture should be practised within a western medical context. Teaching should concentrate on conventional diagnostic methods and treatment approaches. Research should optimise the clinical technique and further explore underlying mechanisms. TCA would become a subject for sociologists and medical historians to study".
Ernst and White don’t specify how action based on these outcomes is to be taken - with or without the consent of traditional acupuncturists, as self-regulation measures or measures imposed by regulatory authorities, by pressure mounted in the media, etc. It would be helpful if they addressed this issue. The evidence-based project that Ernst and White espouse is extraordinarily ambitious – recognising that acupuncture is practised according to different styles, they go as far as proposing that individual schools should be encouraged to provide rigorous [my underlining] evidence of success of their approach.
While good research is something desirable, one must guard against bad research being used as a basis of action. One must question whether this project is realistic in the first place, for many reasons – including the formidable practical difficulty of performing enough sufficiently large scale, rigorous and reliable studies that reflect the subtleties and variety of ways that traditional acupuncture is practised. One’s opinion regarding the importance of research as the basis of medical practice will be relevant. If one believes that medical modalities should only be practised after a condoned method of research has determined that they ‘work’, one will reach very different conclusions than if one believes that people have a right to choose whatever form of treatment they desire, that different treatments can have value beyond narrow considerations of whether they ‘cure’ a particular condition, and that research should only be used as one factor in making informed choices Churchill, 1999). One’s definition of acceptable research, ‘work’ and ‘cure’ will be important considerations. It is a weakness of Acupuncture – A Scientific Appraisal that crucial issues such as these are not explored. Ernst and White’s bias is towards a use of acupuncture with underpinnings in the biomedical worldview that is highly restrictive of the way acupuncture should be practised, leading to a reduced and reductionist practice of acupuncture.
This book can be seen in the context of a biomedical agenda to subject CAM to biomedical scrutiny. As was said in the editorial of the New England Journal of Medicine (Angell and Kassirer 1998) "It is time for the scientific community to stop giving alternative medicine a free ride. There cannot be two kinds of medicine - conventional and alternative". Although many CAM practitioners have welcomed research into their fields, they have assumed that this would be in the context of their own paradigms being respected. By contrast, the 'one medicine' approach is associated with the intention of subordinating CAM modalities to biomedical practice and theory. This is a project for which the consensus of the CAM community has not been sought, begging the issue of what group of people have a right to determine how medical issues are to be considered, and by what criteria. Biomedical exponents have the answer – it is ultimately they, with the methods and outlook they condone. While the affective tone in Acupuncture – A Scientific Appraisal is neutral, similar statements by the biomedical community are often quite angry, exasperated, even sarcastic and insulting.
The following are common characteristics of the biomedicalisation project for CAM:
1. The assumption that CAM requires regulation, and that this cannot be left to CAM modalities alone.
2. The assumption that evidence based medicine, employing criteria that biomedical exponents condone, should be a primary criterion for assessing CAM. .
3. The perception that CAM modalities are not scientific.
4. The assumption that if CAM modalities are ‘shown to work’, the mechanisms with which they operate, understood in biomedical terms, should be sought and become the basis of their modes of practice.
5. Within this process of assessing the effectiveness of CAM, and seeking its biomedical mechanisms, CAM modalities should cease as independent activities, and become ‘integrated’ into conventional medicine, so that there is only ‘one medicine’.
There are certain secondary characteristics:
6.There is an assumption that many CAM modalities may not work at all, but to the extent that they do, this is largely or entirely due to placebo effect.
7. The attraction to CAM may be due to the credulity of the public, and even to mental problems suffered by CAM patients, for which psychiatric help might be desirable (see Holland, 1999).
8. To the extent that CAM modalities do work this is empirical, but the non-scientific aspects of their theories do not stand up in the ‘light of current knowledge’, and may be mystical, pseudo-scientific, vague, and definitely invalid and in need of proper scientific research and explanation.
An overriding assumption is that modern science is privileged beyond other paradigms, and affords the proper basis with which to deal with these issues. This is an absolutist, not relativistic, paradigm position, and is philosophically invalid (Kuhn, 1996, Capra 1983, Churchill 1999). It is a position that can only logically lead to CAM modalities not being practised according to their own theoretical frameworks, but according to biomedical theory, if they are allowed to be practised at all. Not surprisingly, paradigm and inter-paradigm issues are usually glossed over by exponents of the biomedical project.
A paradigm involves many assumptions, and constitutes a world-view which is limited in scope. It cannot be absolute, and does not reveal the truth. Different paradigms are incommensurable, meaning that their frames of reference are different so that the world depicted in one paradigm cannot be translated into another, just as not everything that can be said in one language can be said in another. There is no philosophically correct manner with which to determine the validity of one paradigm by another. (See Kuhn 1996, Churchill 1999)
At this juncture, some who support the biomedical project might object that its aim is simply to determine whether CAM modalities work for specific conditions, that there is no further agenda to usurp the CAM’s own theoretical frameworks. (On the other hand, many advocates of subjecting CAM to biomedical scrutiny, including contributors to Acupuncture – A Scientific Appraisal, are explicit about their desire to develop biomedical models for practice, and to jettison the traditional theories.) Doesn’t Ernst and White’s outcome 3, which suggests the possibility of making TCA the basis of acupuncture’s practice if warranted by clinical evidence, demonstrate this clearly?
If one considers other writings of Ernst and White, it is clear that they give privileged status to the biomedical paradigm. White has maintained "What these ancient Chinese doctors observed cannot be disputed, it is only their explanations, based on Taoist philosophy and a vitalistic view of the world, that are unacceptable in the face of current knowledge". (White 1998a) Ernst has asserted that yin and yang have not been substantiated by scientific research – a comment that betrays ignorance of inter-paradigm issues, and demonstrates his belief that it is appropriate to make the biomedical paradigm the basis for judging Chinese theory (Quoted in Laurance, 1998). Ernst and White have both supported the ideal of integrated or ‘one medicine’. White (in conjunction with Filshie) has written about his wish for acupuncture’s "complete integration into conventional medicine over the next decade" (Filshie and White, 1998), and Ernst has said "I think that the way to go about [integrating complementary medicine into orthodox medicine] is to take one form of complementary therapy after another and test it, and if it can be shown to be work in a definite condition then it ceases to be complementary and it is orthodox medicine, and my aim would be to integrate these therapies into orthodox medicine and one day there will only be one medicine" (Quoted in Golby and Hopper, 1998).
Ernst and White’s Outcome 3 mentions the need to continue to explore ‘the underlying mechanisms’ of acupuncture. (Comments to this effect are made by biomedical researchers almost every time a CAM modality is shown to have positive clinical results.) These mechanisms are within the framework of biomedical theory. Forms of medicine are more than techniques – they are practised in the light of a theoretical understanding. Chinese medical systems have their own way of explaining how and why acupuncture works, but advocates of ‘one medicine’ seek biomedical explanations. It is only a short, and inevitable step, to maintain that the modality should then be practised in accordance with that biomedical understanding, whence the modality has become fully biomedicalised. In any call for ‘one medicine’, the situation could not be otherwise. A medicine which had a multipli"city" of incommensurable paradigmatic bases could not be called ‘one medicine’, or integrated. The entire direction of modern science has been towards theoretical compatibility within diverse scientific disciplines to achieve an overall identity – one continuous scientific worldview from subatomic particle to cosmos. Scientists are hardly likely to make an exception to traditional Chinese medical theory, with its hard to fathom concepts such as yin and yang. If we do achieve ‘one medicine’, traditional Chinese acupuncture will truly be ‘a subject for sociologists and medical historians to study’ as Ernst and White suggest.
A very considerable, probably insuperable problem for a rigorous evidence based medicine project to regulate acupuncture is the magnitude of research required to do justice to the multipli"city" of styles and complexity with which TCA as it is practised. In Acupuncture- A Scientific Appraisal, Ernst considers the evidence of acupuncture’s efficacy to only be conclusively positive for treating three conditions: dental pain, low back pain, and nausea/vomiting. Studies of nausea/vomiting were extremely simplistic: only one point (Neiguan P-6) was tested. Traditional acupuncturists would rarely needle this one point alone. For just one of the many conditions associated with nausea/vomiting, morning sickness, Obstetrics and Gynaecology in Chinese Medicine by Giovanni Maciocia lists seven different patterns, each with distinct treatments involving several acupuncture points. Those treatments are only a guide – most practitioners would never apply a formulaic, identical treatment each time they treated the patient, needle stimulation would not be identical for each patient, each time, nor for each point, nor would the needles be retained a uniform amount of time. More than one disease pattern could exist in that patient, requiring those patterns to be treated concurrently or sequentially, and the disease pattern could alter during the course of treatment. Other TCM texts would advocate different treatment strategies from Maciocia’s, and TCM is only one style of acupuncture among many. A very complex set of variables arise in the real-life acupuncture treatment of morning sickness, many orders of complexity greater than those in a conventional drug trial, and even large-scale conventional drug trials very often yield uncertain results, and are extremely expensive. With such difficulties, it is unlikely that clinical evidence can ever provide more than very general guidelines for the practice of acupuncture. If total reliance was put on biomedically acceptable research to determine the minutiae of TCA’s practice, it would almost certainly become be a very restricted, impoverished form of treatment.
I summarise a number of points that are relevant in the debate about subjecting acupuncture to scientific scrutiny to determine how it should be practised.
1. An evidence-based medicine approach to stringently regulate the practice of acupuncture is questionable because rigorous, large-scale studies that reflect the scope and aims of TCA as it is actually practised are probably unachievable. Pressure to practise based on studies that were inadequate to act as guidelines would be unacceptable to traditional acupuncturists, and would be unfair to patients.
2. Clinical studies are open to bias in construction and interpretation. Different traditionally based acupuncture styles have incommensurable ways of seeing illness, health and goals of treatment – there are evident differences between Worsley Five-Element style and TCM style acupuncture, for example, that cannot be reductionistically represented by research into each one’s rate of success in treating a particular symptom. Standardisation of research criteria can be inappropriate, and research results must be interpreted with specific reference to all assumptions that were made in the study. Moreover, as Resch, Ernst, and Garrow concluded in their study 'Does Peer Review Favor the Conceptual Framework of Orthodox Medicine?', European Journal of Clinical Nutrition "Despite a remarkably large within-group variation in both groups, there seems to be a relevant reviewer bias against papers dealing with unconventional medical concepts". Meta-analyses of similar data can reach opposite conclusions, due to subjectivity. As Morton Lindbaek writes (1999) "Meta-analyses based on small studies are not easy to interpret, and conclusions should be drawn with caution … [In] meta-analyses too there is an element of subjectivity in the research question posed, the choice of outcome measures, and the evaluation of whether significant differences also are clinically important". Since biomedical adherents have a prominent role in promulgating and interpreting research, there is a strong danger that their assumptions and biases, many of which might be hidden, will be expressed. CAM professions should take great care in protecting against potential bias, as well as developing their own interpretations and clinical studies, and being active in the arenas in which research is considered.
3. The privileged status scientists claim for their paradigm is invalid. Subtle examples include statements like Filshie’s and Cummings’ in Acupuncture – A Scientific Appraisal"The correlations made between organs and their attributed functions are wildly inaccurate with what we understand today in medical science" (p34). Their error is to imply that what is referred to by ‘organ’ or ‘heart’ in traditional Chinese medicine and biomedicine should correspond. In fact, they are extremely different concepts that cannot be considered apart from the entire theoretical framework of the paradigms in which they occur, and are untranslatable from one framework to the other. This is why many TCA translators carefully select terminology avoiding terms that have associations with the biomedical paradigm. Porkert (1980) translates ‘zang’ as ‘orb’, not ‘organ’, and uses ‘orbis cordialis’ instead of ‘heart organ’. Larre translates qi as ‘breaths’, not ‘energy’. They do this to prevent inter-paradigm confusion and to attempt to convey the sense of those concepts as they appear in their original paradigmatic context.
The problem of incommensurability of paradigms is of utmost significance to CAM modalities. Not taking it into account will distort research, making it appear to validate biomedical concepts to the extent that these exist as implicit starting assumptions. Furthermore, paradigm-ignorant and misleading statements to the effect that CAM theoretical frameworks are invalid because they ‘don’t stand up in the light of modern scientific knowledge’ are often used to attack CAM.
4. The lack of ‘proof’ for a high percentage of biomedical practice – the U.S. Office of Technical Assessment (quoted in Brown, 1998) estimated that only 20% of biomedical procedures are proven – diminishes the biomedical claim of scientific authority.
Many of these important issues to CAM are not discussed in Acupuncture – A Scientific Appraisal, and to the extent they are, it is generally from a pro-biomedical perspective. With the proviso that these issues are important to traditional acupuncturists, the book contains much of value. There can be a middle ground between being only interested in tradition, and accepting only rigorous research within the mainstream scientific paradigm. It is possible to have a pluralistic outlook that recognises the value of diverse paradigms.
The material in this book covers a number of areas, handled in a scholarly manner: a summary of studies on the attitudes towards and use of acupuncture by the public and biomedical profession in the West (White and Ernst), history, nature and current practice of acupuncture: an East Asian perspective (Birch and Kaptchuk), Western medical acupuncture (Filshie and Cummings), neurophysiology of acupuncture analgesia (White), effects of sensory stimulation (acupuncture) on circulatory and immune systems (Lundeberg), clinical effectiveness of acupuncture; an overview of systematic reviews (Ernst), adverse effects of acupuncture (Rampes and Peuker), and a conclusion (Ernst and White). The articles are all interesting, intelligently and professionally presented.
Even if one objects to its biomedical prejudices, there is much that acupuncturists of all persuasions can gain from this book, including summaries of various studies. The article ‘Adverse effects of acupuncture’ might be considered required reading for acupuncture students (and practitioners!). On this topic, in her essay Filshie quotes a personal communication to her from Ernst "The side effects of acupuncture are "dimensionless*" less than those of drugs". This is an interesting qualification to Ernst and White’s sterner comment "All we can state at present is that acupuncture is not free of risks".
*i.e. magnitudes less
Gertrude Stein, on her deathbed, was asked ‘What is the answer?’ to which she replied ‘What is the question?’ If one accepts a free-for-all situation for medical systems, there may be no questions to ask, but as soon as one attempts to be more discerning, considering the responsibility of practitioners for their patient’s welfare, questions arise. The questions and the nature of one’s answers will reflect one’s point of view, inextricably linked with initial assumptions that may be unacknowledged, with possibly other motives coming into play such as vested interest. Before long, a very complex situation arises. Given the dominance of biomedicine in our society, there is a strong possibility of policies that reflect the biomedical viewpoint and interests being imposed whether or not they are the best, necessary, or intellectually valid, solutions. CAM professionals will have to be energetically involved in all areas regarding research, committed to ensuring that fundamental issues and assumptions are explored in an intellectually rigorous manner with regard to paradigm and inter-paradigm issues, if they do not want scientific appraisal to become scientific appropriation.
1. There are many viewpoints and underlying attitudes from which to approach scientific enquiry. It is beyond the scope of this essay to consider this important matter here, but enlightening discussion can be found in Kuhn (1996), Bohm (1981), Capra (1983 and 1997), Naydler (1996), and with more specific regard to Chinese medicine, in Scheid and Bensky (1998).
2. Another term, ‘integrated medicine’, is often used. It is an unfortunate term, because it can imply either a multi-paradigmatic or uni-paradigmatic basis for medicine. ‘One medicine’ implies a uni-paradigmatic basis, and ‘pluralistic medicine’, a multi-paradigmatic basis.