Notes on holistic medical research

Prepared by John Heron for the Research Council for Complementary Medicine, London, 1983.

On general philosophical grounds we take the view that patients are persons; and that persons have the capacity to be self-determining - as agents they are causes of their own behaviour. Such causation through personal agency is sui generis: it cannot be reduced to some other, e.g. purely physical, sort of causality - although it may interact with and impinge upon the latter.

The capacity to be self-determining, we hold, means also something more than the ability to exercise choice in overt action. It also includes a capacity for physiological self-regulation. Persons, through biofeedback, autogenic training, relaxation, visualization and meditation, can directly influence physiological processes. The full range and extent to which physiological processes can be affected by such internal agency is at present unknown. But we know no reasons, philosophical or empirical, to suppose that the potential for such influence is not vast.

We require of any research method in medicine that it should respect persons (patients) as self-determining in these two fundamental respects. It should respect patients as persons whose external actions are freely determined by full and relevant information, and whose internal actions have great potential for physiological self-regulation.

We therefore object to conventional research protocols which take patients matched with respect to disease conditions and other variables and randomly assign them to a treatment group and a control group. Random assignment treats the patients as less than persons because it contradicts their right to fully informed self-determination in the selection of available treatments. And in a more subtle way it contradicts their right to develop their potential for physiological self-regulation in the direct management of their disease: it does this by keeping patients quietly in the dark about this potential, for if they started to practise such self-regulation this would be an additional and unscheduled treatment variable that might foul up the experimental design. In a nutshell, traditional research protocols necessarily by their nature and assumptions treat persons who are the subjects as other-determined, not as self-determining.

Nor will it do to argue that these objections can be met by the principle of informed consent. Such consent in conventional medical research is ill-informed or improperly informed. Thus it is not made clear to prospective trial subjects (a) that they have potential self-healing power through physiological self-regulation, and (b) that for purposes of the experiment they are being asked to abdicate the use of this. Thus ill-informed consent is obtained by exploiting the uneducated Cartesian passivity of current patient populations. Nor is full information given about all that is known about the treatment to be tested, for, from the standpoint of the conventional approach, this might generate a distorting mental set that could interfere with a proper appraisal of the "objective" effects of the treatment or it might undermine patient compliance.

The point about informed consent is that it could only be given by prospective trial subjects if they had been given full information on the developmental history of the treatment or drug to be tested, on the experimental design and the assumptions on which it is based, on the full range of alternative treatments available, on human potential for physiological self-regulation. But what seriously ill rational being would give consent to participate in a standard experimental protocol after due consideration of all this? And what conventional researcher would consent to give out such information knowing that it would both subvert patient compliance with the experimental design and distort the design itself?

Random assignment also may pose a real moral problem for clinician-researchers who want to research a treatment on seriously ill patients and who believe on the basis of available clinical evidence that the treatment is effective. They will find it ethically unacceptable to withhold the treatment from some of their patients by assigning them to the no-treatment control group or to a comparison group using conventional but ineffective treatment.

A research method which uses as a central protocol random assignment of patients to treatment and control groups sustains a culture of alienation among patients, in which they are conditioned to be cut off from what is going on in their bodies - from free and fully informed decisions in the selection of treatments, and from physiological self-regulation in the self-management of their disease. It encourages and sustains the Cartesian split: doctors and patients see bodies as alienated from the exercise of self-determination and the influence of mind. And it keeps the development of medical knowledge firmly in the hands of the medical researchers, and out of the hands of the patients to whom it is supposed to refer.

Conventional medical research does of course paradoxically acknowledge the direct influence of the mind on physiological states - in the so-called placebo effect - but does so only in order to control for it and separate if off from the "objective" physical effect of the treatment. The conventional paradigm is thus separatist and reductionist: it wants to separate off mental from physical effects and reduce analysis of treatment to independent physical effects.

By contrast we believe medical research needs to be inclusive and holistic, adopting a contextual or ecological approach to inquiry into treatment. Treatment is always in a context. It is a dynamic part of a system, and it needs to be explained and understood in terms of its interconnectedness within that whole system. Take acupuncture treatment. The context or system in which it occurs includes at least the following parts: the location and depth of insertion of the needle; the way in which the needle is manipulated by the practitioners; the behavioural cues and sensations of the patients during and after this manipulation; the physiological states of the patient before, during and after insertion; the state of the morphogenetic field of the patient's body (and of the practitioner's body) before, during and after treatment; intentions, expectations, emotional states and unconscious mental processes of the patient and of the practitioner before, during and after treatment; current and other sorts of interaction between patient and practitioner, such as counselling or patient self-help contracts; the psychosocial and physical history of the patient and current psychosocial situation; the norms, values and belief-systems of the wider culture within which the treatment is set.

Such a contextual system within which treatment occurs is highly complex, including orders or levels of being - the physical, the morphogenetic, the psychosocial. A contextual or systemic inquiry would presumably have two basic stages. The first stage would be to formulate some hypothetical model of the basic sorts of patterning within the system. Such a model might derive from a consideration of such questions as: Is the system hierarchical? If so, which parts are included in which? In what respects are the parts functionally autonomous, and in what respects functionally interactive? Is the interface between the autonomous and the interactive variable, and if so, what factors might influence this variability? What's the relationship between a hierarchy within a level of being, and a hierarchy between different levels of being? And so on. See paragraphs 19 and 20 below.

The second stage of a contextual inquiry would be to test the fit of this conceptual model in the context of actual treatment. This is the empirical part of the inquiry and we would hold that it is necessarily participatory: those who are in the system and constitutive of it - the practitioner and patient - are necessarily front-line researchers since they alone have adequate access to central parts of the system, such as intentions, beliefs and expectations. Observers relatively external to the system may provide some pertinent data that throws further light on what is going on within the system: but what the system is and how this light is thrown can only be defined firstly from participation within it. And if external. observers join the system sufficiently to be able to discriminate it from within, then they are no longer external observers but are contributing to and participating in the treatment process.

Contextual inquiry, then, converts the practitioners into a researcher, the treatment process into a concurrent inquiry process, and the patient into a co-researcher who is invited to internalize and co-operate in the inquiry. This means that treatment becomes also a form of action research in which practitioner and patient collaborate in generating knowledge about the whole treatment context or system through collaborating in seeking to bring about change in the patient's ill health. Such a collaborative action research model follows too from the requirement that medical research should respect the patient as a self-determining person: the model can honour both informed free choice of treatment, and the exercise of physiological self-regulation.

Patient collaboration can vary on a continuum from the patient being fully involved as creative and imaginative co-researcher, to the practitioners sharing full information at all stages and seeking patient assent. It is clearly important to acknowledge this continuum, since patients vary enormously with respect to age, intelligence, education, and debility due to disease process.

In contextual inquiry into complementary medicine, the question is not simply: Does acupuncture work? Is there an active principle in homoeopathic remedies? It is rather: In the context of what whole treatment system does acupuncture work? When acupuncture "works" what is the patterning of elements or parts of this system, and what is the pattern of interacting factors that contribute to the treatment outcome?

Now a basic objection to this from the conventional research standpoint will be that it provides no way of separating out whether the acupuncture per se has any effect independent of all the other variables: as long as it is studied only when embedded in the midst of a host of other treatment variables, you will never know whether it is having any effect at all as distinct from other parts of the treatment. An equally basic answer to this objection is that there is no such thing as a physical treatment per se: physical treatment is always given in a psychosocial context, and in the context of the formative field of the patient's body. If the context is hierarchically influential in determining how the physical treatment works, it's no good selecting physical treatments from the bizarre contexts of conventional research designs: rather we need to understand physical treatments in dynamic interaction with their normal fully fledged person to person context.

Validity can be sought in various ways. Firstly, as in all action research, the patient-practitioner dyad is its own control on a serial basis. The dyad has a working conceptual model of the treatment context with its different levels, applies this in treatment actions, and uses data from this application to amend, refine and correct the model. So treatment model and treatment action reciprocally control and modify each other over time. Secondly, the practitioner can use data from his own previous dyads as a further basis for comparison and control. Thirdly, peer groups of practitioners, with or without some of the patients involved attending, can compare and contrast their current and past dyadic findings. What is essentially involved in all this is that a whole variety of perspectives on the dynamic of the treatment context are used to clarify a valid account of it. Over the whole sub-culture of such treatment dyads there will be trial and error, experimentation, appropriate attempts at falsification - a criss-cross of findings used to inform each other and provide starting points for future inquiries.

A research culture of contextual inquiry involves some major shifts of attitude both for patients and practitioner. Instead of patients being those who have been arbitrarily afflicted by a disease process and passively await the expert delivery of care, they become persons challenged with an opportunity for increased self-determination through active participation in a process of inquiry and learning. And practitioners, instead of seeing themselves as delivering a theoretical and technical system of physical treatment, see themselves as participants in a whole treatment context of which their particular physical specialty is but one dynamically interacting part. It may be that for some practitioners, or indeed for some whole school or modality of complementary therapy, principles enshrined in the theoretical and technical system of this physical treatment will need to yield to somewhat different principles born out in action research contextual inquiry into the use of that physical treatment.

The first stage in contextual inquiry, mentioned above, is to formulate some conceptual model for the way in which components of the treatment context interact to form a dynamic system. As a prolegomenon to this stage, it may be useful to look at the explanatory notion of causation. This concept of causation is used as a way of explaining and understanding how it is brought about. But there are many different sorts of causation (sorts of change), and one important business for an holistic, ecological, contextual sort of inquiry is to separate them out and understand something about their systemic interaction. For such interaction will have a fundamental bearing upon both the aetiological and therapeutic aspects of treatment. The following sorts of causation have a prima facie claim to be considered as not obviously reducible to each other:

1. Energetic (efficient or mechanistic) causation: linear in time, the cause precedes the effect; involves some kind of energy transaction; used in the physical sciences.
2. Formative causation: involves the notion of non-physical fields with patterns that determine the development, the spatial ordering, the maintenance, of physical forms; used (by some) in biology.
3. Mutual causation: involves a circuit of interactions in which no one part is the cause of what happens in any other, each part acts on and is acted upon by each of the others, persons in relation with each other provide a basic example.
4. Subintentional causation: unconscious and other mental states in a person unawarely alter overt behaviour or manifest in overt physical symptoms.
5. Unintentional or belief causation: as in the so—called placebo effect, when believing that something will have a physical effect produces that effect.
6. Intentional causation: the conscious exercise of choice or will in causing changes in overt behaviour or in physiological states.
7. Normative causation: the influence of the norms and values of a culture, of social tradition, in shaping thinking, attitude and behaviour.
8. Archetypal causation: the influence (if there be such) of archetypes, imaginals, Platonic forms, "the Gods".
9. Divine causation: cosmogenesis and cosmosustenance; emanation from the divine beyond, emergence from the divine within.

For everyday practical purposes this list can be boiled down to five:  (a) energetic causation;  (b) formative causation;  (c) mutual causation;  (d) mental causation (including subintentional, unintentional and intentional);  (e) normative causation. Is there a contextual hierarchy here? The norms, values, beliefs of society (e) provide the necessary condition for the development of intentional behaviour (d) which provides the necessary condition for human communication and interaction (c) which in turn provides the necessary conditions for bringing about changes in the physical body (b) and (a). Maybe there is a descending causal hierarchy too. Thus the more enlightened the norms, values and beliefs of a social system, the more self-determining intentional behaviour is, which improves the quality of human interaction and communication, which strengthens the formative field of the body, which becomes physiologically healthier. And maybe there is also a return ascending causal sequence. Perhaps, further, there is a relative autonomy about the functioning of the causal mode at each level, yet the limits and fluctuation of this autonomy are set by the causal modes of higher levels and are subject to influence from the causal modes of lower levels. All these speculations are intended merely to suggest ways of starting to conceptualize in systemic terms the whole context of treatment.