Paper 1, Week 6, MSc Medical Anthropology
Tutorial on Treatment Evaluation, 18.11.2002
Barbara Gerke
Q: Discuss methods used for evaluating the efficacy of medical treatment, present their advantages, and explain why they are not entirely satisfactory.
There is no a priori reason to claim that one kind of knowledge
is more authentic or autonomous than the other forms.
Allan Young (1981:380)
Introduction
Blind assessment and the use of placebos, which look back to a history of over two hundred years, developed mainly after World War II into a method used in biomedicine for the evaluation of medical treatment. This essay will deconstruct briefly the history of placebos in the context of the development of treatment evaluation, and will highlight the advantages of randomised clinical trials (RCT). I will also discuss the limits of RCT in terms of doctor-patient relationships and why RCT are not satisfactory when applied to traditional medical systems, taking the example of Tibetan medicine. Finally, two methods that have been proposed by scholars (Montellano 1975 and Browner et al 1988) to assess the efficacy of traditional medicine will be critically evaluated.
1. Deconstructing the History of Placebo
Placebo (lit. “I will please you”) has a been viewed differently at different times in history. A historical deconstruction of the use of placebo in blind assessments is necessary to understand the impact it had on the process of marginalising the therapeutic success of placebos as well as traditional medical treatment methods in favour of a biomedical supremacy.
In order for biomedicine to establish itself as an authoritative medical system in Europe and North America, it needed a method to “isolate ‘hard’ knowledge and material causality from the contamination of mental delusion, enthusiastic bias, or even calculated deceit.” (Kaptchuk 1998:391). Only with an unbiased scientific method, biomedicine was able to gain the financial and political control over medical education and practice. Once the cultural authority was gained through a “rigorous but open, a thorough but unbiased approach to health and disease” (Sullivan 1993:217), biomedicine could claim to be the most successful and only valid medicine. It is an irony in itself that blind assessment and placebos, which have proven to “produce actual but unexplained therapeutic success” and rely upon a “direct link between knowing and healing” (Sullivan 1993:213), have been illegitimatised in favour of providing the clearest proof that a biomedical drug is producing a therapeutic effect observed.
Not undermining the usefulness of RCT for biomedical drug evaluation, its effect on other medical systems should not be underestimated. The medical authorities of the 20th century proved the therapeutic efficacy of ‘simulation’ – a word carefully avoided in the blind assessment process – and used it for their own ends by imprisoning its valuable therapeutic means. However, recent publications (Moermann 2002) have unravelled the practicability of placebo effects for medical therapy, 107 years after the Polish physiologist, Waclaw Sobieranski, revealed that “autosuggestion played a large role in healing” and that “much of the healing properties of chemical medicine is in reality attributed to autosuggestion” (Sobieranski 1895, quoted from Kaptchuk 1998:418).
2. Advantages and Limitations of RCTs
RCTs are designed to identify and isolate therapies. They can precisely test biomedical pharmacological knowledge that is extrinsic to the clinical encounter, excluding effects of belief, expectations, emotions, etc. (Sullivan 1993:225).
In the process of testing the efficacy of a drug, the element of ‘chance’ is the biggest problem. How to account for the chances that someone may get better with or without taking the drug? Clinical trials have tried to get as close to the ‘truth’ by including large numbers of people of the same age, sex and with single diseases. Randomisation avoids any bias during the trial and double-blind methodology excludes influences of suggestion, as neither the doctor nor the patient knows whether the placebo or the drug is given (Ryan T. R., lecture 12.11.2002).
Because of the complexities involved in medicines, which are hard to verify, biomedicine prefers looking at a single substance. Thus, biomedical pharmacology identifies the single active ingredient in a drug and usually selects a single drug for trials (one reason why its methods is difficult to apply to multi-compounded traditional remedies). The three main factors involved in designing RCTs are low-cost, sustainability and local availability (Ryan T. R., lecture 12.11.2002).
Several authors have criticised RCTs for their limited biomedical approach. Mark Sullivan’s main point against RCT is that this method has diminished the “therapeutic potential of the doctor-patient relationship and of the potential human recourses available in healing.” (Sullivan 1993:213)
Till the 19th century, when medical authority was still unestablished among physicians, patients had a major influence in directing the treatment process and making medical decisions (Duden 1991). Along with biomedical authority taking charge, RCTs have eliminated the patient involvement, because the experimental factors have to be as ‘impersonal’ and ‘controlled’ as possible.
Elisabeth Hsu argues that “efficacy” belongs to the “language” of biomedicine and should be understood as a “means to exert control over medical knowledge.” (Hsu 1996:37). Giving more importance to the actor’s position to evaluate the treatment process, she introduces the aspect of “therapeutic quality” for the patient’s subjective perspective, as distinct from the “therapeutic results” of the practitioner’s point of view, regardless of biomedical or alternative backgrounds. The observers of the treatment evaluation are accounted for in the “therapeutic efficacy” of the biomedical scientist and the “therapeutic success” of the medical anthropologist, who takes into account the social, cultural, political and economic circumstances in which the treatment takes place (Hsu 1996:38), thus enlarging the thrust area of treatment efficacy within the disease-illness context.
Several studies proved that the doctor’s personality and empathic understanding of the patient affects the treatment outcome (Moermann 2002:36-7). He also proved that merely the physician’s knowledge during a RCT can bring about pain relief (Moermann 2002:32-3), which shows the influence of thought in therapies that are not accounted for in RCTs, but are often used in traditional medical systems.
Moermann could be criticised for neglecting the influence of the cultural and social status of the medical profession. Medical anthropologists would argue that the doctor’s social status and authority has an even stronger placebo effect. This would apply to various medical cultures. Also the shaman in his special costume or the priest on his temple seat act from a place of authority from the patient’s point of view.
3. Critical Appreciation of RCT in Traditional Tibetan Medicine
In Ayurveda and Tibetan medicine, the diagnosis of illness is made in terms of the balance between five elements (Sanskrit panca mahabhuta; Tibetan `byung balnga) and three humoural principles (Sanskrit dosa, Tibetan nyes pa). This balance is assessed through skilled questioning about the patient's symptoms, behaviour and state of mind, feeling their pulse, and observing their urine, tongue, and external appearance. “Feeling” and “observing” constitute culturally valid epistemological means to arrive at therapeutic conclusions.
Tibetan pharmacology is based on the ability of the doctor to ‘taste’ the right combination of plants. The Tibetan laboratory, so to speak, is on the tongue. The six tastes (sweet, salty, bitter, astringent, hot and sour) determine how the ingredients will affect the five elements. Sensitising the individual taste perception is part of the medical education. As part of their examination, students have to identify plants with blind-folded eyes, simply by their taste. The combination of the ingredients (varying from 4 to 100) is carefully calculated according to the taste of the substances and their consequent effects on the humoural balance.
Moreover, “a medicinal plant has to be picked at a certain time of the year at which the configuration of the five elements and three humoural principles in the environment maximises its medicinal value. Other factors to be considered in gathering medicinal plants include the compass direction of the slope on which they are found, the position of lunar and planetary constellations at the time of gathering, the time of the day, the signs of specific birds and animals in the natural surroundings as well as the state of mind of the person collecting the plant. Different plants have to be washed, dried and preserved in different ways.” (Gerke, Jacobson 1996:17). All of these factors are understood to effect the medicinal power of the final drug. Spiritual practices such as consecrations through the recitation of mantras are often seen as essential contributions to the final drug. Consecration can be understood as an ‘intentional suggestion’ and a ‘projection of concentrated thought’ onto the drug, involving ritual practice. ‘Blessed’ medicines by a ‘high lama’ are considered ‘highly effective’ among Tibetans. The principal appears similar to ‘suggestion’ in intended placebo effects.
Interestingly, RCT has been used to evaluate the efficacy of a Tibetan medical drug, Padma 28, produced by a Swiss company (Drabaek et al 1993, in Aschoff 1996, no. 525):
"36 patients with a median age of 67 years and a median duration of intermittent claudicatio of five years were randomised to either active treatment with Padma 28 or placebo. The effect of treatment was quantified by measurements of systemic and peripheral systolic blood pressures and by measurements of the pain-free and the maximal walking distance on a treadmill. ...The group randomised to active treatment received two tablets bid containing 340 mg of a dried herbal mixture ... of Padma 28. After active treatments, administered over a period of four months in a double-blinded, randomised design, the patients allocated to this group attained a significant increase in the pain-free walking distance. The patient-group receiving placebo treatments did not show any significant changes in either the pain-free or the maximal walking distance."
This trial suggests that RCT can be used in evaluating traditional Tibetan medicine for classified biomedical diseases and thus make it available to the European market, which would otherwise remain closed due to legal restrictions. In fact, adopting RCT and Western medical research methodologies is the only legal and possible way for Tibetan drugs to be produced and accepted in most of Europe and North America.
I have to add here that all 28 ingredients of Padma 28 have been botanically identified and the chemical active components have been isolated and classified. The drug has been manufactured under Western standard laboratory conditions in Switzerland. Considering the earlier exposition on traditional Tibetan pharmacology, it becomes clear that the manufacturing process as well as the RCT does not account for any of the principles of the Tibetan pharmacological epistemic system and that the trial does not measure the therapeutic effect on the balance of the humours and elements.
Moreover, the individual constitution is ignored in the RCT. But it is considered important in the traditional system and includes disease classifications of ‘hot’ and ‘cold’, which implies that patients having the same disease could be treated differently because of their ‘hot’ or ‘cold’ nature. We also need to consider that none of the isolated active principles is relevant in the cultural setting and knowledge system of a Tibetan pharmacologist, who is concerned with the taste of the raw ingredients and the final drug for its effects on the humours.
Still, RCT proved that the recipe Padma 28, which is of Buryat origin, is effective in cases of stable intermitted claudicatio. Whether this is ethically, culturally, medically or anthropologically an acceptable path to introduce Tibetan medicine to Western patients is still widely debated. Medical anthropologists argue that medicines should be evaluated within the framework where they were established. For Tibetan medicine this would mean that it would never be introduced to the West on a large scale, unless the biomedical legal establishment changes. This brings us back to the first note of how biomedicine established its supreme authority and decision power in Europe and America through illegitimatising the ‘placebo effect’ which has a direct link between knowledge/mind and healing. Consequently, all medical traditions which are characterised by direct links between knowledge and healing have been illegitimatised in the same way by the biomedical definition of ‘therapeutic efficacy’.
Incidentally, human ‘ignorance’ (Tib. ma rig pa) is mentioned as primary cause of illness in Tibetan medical texts. Having realised the ‘power of mind’ as causing illness, its therapeutic qualities have also been utilised effectively in Tibetan healing rituals.
4. Proposals for the Evaluation of Efficacy of Ethnomedicinal Plants
Bernard Ortiz de Montellano made an interesting proposal to evaluate Aztec medicinal plants as early as 1975. He was interested in the pure native view on medicine, but used modern knowledge to evaluate the effectiveness of Aztec medicinal plants. Montellano argued that with 16 of the 25 tested plants producing the effects claimed in Aztec sources, that – apart from magic and religion - there must have been a strong empirical knowledge among the Aztecs, which should receive more attention (Montellano 1975: 220). In anticipation of his early research, Montellano (1981) and Browner et al (1988) proposed “four levels of confidence” according to which herbal medicines could be evaluated. Their data showed that “two distinct ethnomedical systems can reach similar therapeutic understandings through different analytical routes” (Browner et al 1988:686-8). The authors claim that their approach can be utilised to organise culture-specific material into standardised comparable units, and that biomedicine - despite all its limitations and culture-constructed categories - still offers useful methodologies to analyse and compare ethnographic materials. (Browner et al 1988:689).
This approach of traditional treatment evaluation was heavily criticised by Byron Good who accused Browner et al for “mapping local cultural phenomena onto a biomedical grid”. Good argued that “the task of medical anthropology can never be simply to discover links between medical cultures and biology”, and that “illness realities are never reflections of biology. Biology becomes human illness only as it is given meaning....” (Browner et al 1988:693-4).
Conclusion
Evaluating treatment efficacy is a very complex undertaking. Biomedicine has developed approaches, such as double-blind randomised clinical trials, which are useful to evaluate the efficacy of single drugs in a biomedical context. Apart from being expensive, trials only take into account isolated active substances and their effect on disease, ignoring patient-doctor relationships and socio-political aspects of illness (Sullivan 1993, Hsu 1996).
The deconstruction of the role of the ‘placebo effect’ in history (Kaptchuk 1998) revealed its abuse in the design of clinical trials, undermining and illegitimatising its therapeutic effect. Recent studies on placebos (Moermann 2002) led again to an increasing integration of ‘suggestion’ into medical practice, though keeping an extrinsic biomedical view. The effect of intentional thought and suggestion, apart from use in placebos, has not yet received sufficient attention, especially in traditional medical systems which often have a detailed knowledge system on the healing power of the mind.
The selective ’disease-drug view’ of RCTs is necessary for arriving at an unbiased evaluation, and perhaps useful for comparison and classification of data, but seems highly inappropriate for evaluating traditional medical drugs from their indigenous cultural view point. The examples of Aztec medicine (Montellano 1975) and cross-cultural ethnomedical research (Browner et al 1988) show that biomedical methodologies can be used to evaluate the efficacy of medicinal plants to a certain extent if the native aetiology of the drug use is taken into account. However, this applies to single herbs only and still uses a biomedical restricted model.
The epistemology of the Tibetan multi-compound pharmacology, for example, is much more complex. Any conducted RCT will account only for the efficacy of the ‘traditional’ drug according to biomedical disease categories, which is essential if the tradition wants to find a legal entry to the Western health market. However, native aetiologies of the individual constitution, the three humours, five elements and six tastes are ignored in such trials. Medical anthropologists may be able to create more awareness on these evaluation issues and may take part in designing new methodologies to evaluate therapeutic efficacy involving categories of indigenous knowledge systems for assessing and evaluating the therapeutic effect on the principles used in the traditional medical system itself.
References
Drabaek, Hans, Mehlsen, Jesper, Himmelstrup, Holger, and Winther, Kaj 1993: “A Botanical Compound, Padma 28, Increases Walking Distance in Stable Intermittent Claudicatio. Angiology. The Journal of Vascular Diseases Baltimore/New York:44:863-867. Quoted from Aschoff Juergen 1996: Annotated Bibliography of Tibetan Medicine (1789-1995). Ulm: Fabri Verlag.
Browner C. H. et al. 1988: “A Methodology for Cross-cultural Ethnomedical Research”. Current Anthropology 29(5):681-702.
Duden, B. 1991: The woman beneath the skin: A Doctor's Patients in Eighteenth-Century Germany. Cambridge MA, London: Harvard University Press.
Gerke B. Jacobson E. 1996: “Traditional Asian Medical Cultures Encounter Biomedical Research”. AyurVijnana 1. Kalimpong: International Trust for Traditional Medicine:15-21.
Hsu E. 1996: “The Polyglot Practitioner: Towards Acceptance of Different Approaches to Treatment Evaluation”. Olsen S. Gosvig & Hoeg E. (eds.) Studies in Alternative Therapy III. Odense: Odense UP: 37-53.
Kaptchuk T 1998: “Intentional Ignorance: A History of Blind Assessment and Placebo Controls in Medicine”. Bulletin of the History of Medicine 72, 389-433.
Moerman D. 2002: Meaning, Medicine and the ‘Placebo Effect’, Cambridge University Press.
Ortiz de Montellano B. 1975: “Empirical Aztec Medicine: Aztec medicinal plants seem to be effective if they are judged by Aztec standards”. Science: 188(4185):215-220.
Sullivan M.D. 1993: “Placebo Controls and Epistemic Control in Orthodoxy Medicine”. Journal of Medicine and Philosophy 18:213-31.
Young Allan: “The Creation of Medical Knowledge: Some Problems in Interpretation”. Social Science & Medicine: 15B:379-86.