Georg Ivanovas From Autism to Humanism - systems theory in medicine

6. Systemic Medicine

previous -- home -- content -- next


6.13 Control and requisite variety

What is medicine good for? What is its aim? In order to come to some final conclusion it is necessary to refer to this basic question. The Hippocratic oath takes a stand on this, although it is more concerned with the social position of the physician. In the modern and slightly adapted form, in the Declaration of Genova, the oath says: “The health of my patient will be my first consideration” (Hippocratic oath 2008). I personally prefer the more strict statement of Hahnemann. He claims in the first aphorism of his ‘Organon’: “The physician's high and only mission is to restore the sick to health, to cure, as it is termed” (Hahnemann: 90). Some consequences of this point of view, to which probably everybody will agree, are elaborated in the footnote to this aphorism. Hahnemann says there:

His mission is not, however, to construct so-called systems, by interweaving empty speculations and hypotheses concerning the internal essential nature of the vital processes and the mode in which diseases originate in the interior of the organism, (whereon so many physicians have hitherto ambitiously wasted their talents and their time); nor is it to attempt to give countless explanations regarding the phenomena in diseases and their proximate cause (which must ever remain concealed), wrapped in unintelligible words and an inflated abstract mode of expression, which should sound very learned in order to astonish the ignorant - whilst sick humanity sighs in vain for aid. Of such learned reveries (to which the name of theoretic medicine is given, and for which special professorships are instituted) we have had quite enough, and it is now high time that all who call themselves physicians should at length cease to deceive suffering mankind with mere talk, and begin now, instead, for once to act, that is, really to help and to cure.

Although this statement is rather old, it refers to certain issues, still valid. As said before (chap. 2.1.g), reductionism resembles Celsus ‘theoretical medicine’ (appendix II). Reductionist research regards all kinds of single mechanisms as the ‘cause’ of a disease, including genetical, pathophysiological or biochemical traits. This implies that the correction or the manipulation of these traits will lead to health. It is a kind of bottom-up approach. In contrast, Hippocrates and Hahnemann come from an empirical medicine which is more top-down oriented. But such distinctions as bottom-up vs. top-down or theoretical vs. empirical remain somehow vague. Much more precise is the distinction between central and distributed control (chap. 6.5).

The central approach gives the physician a more important position. He becomes the main figure deciding on the end points to be reached. But in how far does the physician (the observer) understand what is necessary for the human and his organism? Or in other words: how smart is the human organism and how smart is the physician?

This crucial question has been formulated in the law of requisite variety (chap. 4.7). It says that in order to control a system the observing system has to have the same complexity as the observed system. This implies that the physician has to have tools of observation which correspond in their complexity to the complexity of the human. But this is, as we all know, not the case. Actually it is an unattainable ideal. So what can be done with the complexity and variety encountered in medicine?

Reductionism, as its name says, reduces variety, and it does it on different levels. First, observation is reduced to reproducible mechanisms or to statistically significant probabilities. Second, the variety of therapeutic means is reduced in excluding most non-specific measures, even the therapeutic relationship (chap. 6.1). Third, the variety in the patient is reduced by holding variables more or less stable. Research under such conditions does not correspond to the nonlinear nature of the human. Results are valid, but it is a little bit like hunting Easter eggs oneself has hidden (chap. 2.5.h).

According to the law of requisite variety it would be necessary to increase the variety of the observer. This is rarely done. Reductionist scientists often show a certain concern that allowing a different sight would imply that ‘anything goes’. But this is not necessarily the case. There is a middle way between the Scylla of reductionism and the Charybdis of arbitrariness. Concepts like ‘optimal variety’ (Paritsis 1992: 35-39) might be appropriate. That is, scientific strictness is also possible in complexity. Or better: Only the variety of epistemological tools is scientific when it comes to complex situations as in medicine.

Every medical problem can be described in different ways, from different points of view, on different levels. This has been demonstrated with osteoporosis (chap. 4.6), asthma (chap. 5.5.b) and with the relation alcohol-health (chap. 2.1.c). No formal model has been provided, yet, to connect all the different findings and theories. However, this would be necessary to overcome the autistic-undisciplined thinking. In my opinion the following preconditions have to be met, if one tries to establish a correct model of therapy:

  • First, it is necessary to understand the context of the problem and of a possible intervention (chap. 3.6).
  • Second, one has to formalize the level of abstractions (chap. 3.2).
  • Third, it has to be clear what kind of control is intended (chap. 6.5).
  • Fourth, the consequences of each intended intervention onto all levels has to be assessed, or at least assumed.
  • Fifth, the individual control of effectiveness has to be established according to clearly defined principles.

Such a model would be a rather ambitious project. It has to involve an understanding of robustness in general and of an individual prognosis in special. The following structure tries to provide a first tool to think about such notions as context or level of abstraction in a medical frame. It cannot be called a model, as it is linear and thus not appropriate to describe the observed processes.

The most important factor for health is probably what is called lifestyle (Universitat Autonoma de Barcelona 2008). For example, over half of deaths in women from chronic diseases such as cancer and heart disease could be avoided if they would never smoke, keep their weight in check, take exercise and eat a healthy diet low in red meat and trans-fats (Dam et al 2008, Forman et al. 2009). And obesity is as harmful as smoking (Prospective Studies Collaboration 2009). Lifestyle concerns the whole person and a therapeutic intervention is not limited to one level. It can’t be, although the theory might suggest something else from time to time. But even in those cases, or especially in those cases, a clear logical structure might be helpful to overlook the different possibilities of intervention (1). This shall be demonstrated with the asthmatic disease.

a) local pathology

Asthma can be seen as a local problem of the lungs with bronchoconstruction, inflammation, hypersecretion of Becher cells, a disturbed remodelling process (chap. 5.5.c), and so on. Interventions on this level (local steroids) improve the symptoms, often quickly and impressingly. However, these therapies do not change the further development of the disease (chap. 5.5.c). They are a first order therapy or a suppression of symptoms (chap. 6.8). A stimulating and not suppressing therapy would be the inhalation of sea water, as performed in several spas, the relaxation training of the lungs, or a hypnotherapy with the intention to improve breathing.

b) mediators and receptors

Mast cells release histamine, cysteinyl leukotrienes and cytokines producing an inflammatory cascade leading to the asthmatic attack. But “it is hard to believe that the very large and rather selectively distributed number of mast cells in normal, uninflamed, non-infected, non-traumatized mammalian skin or mucosal tissue simply hanging around there lazily day and night, just wait for the odd allergen or parasite-associated antigen to come by so the mast cell can finally swing into action” (Maurer et al 2003). Mast cells have a meaning. For example, they regulate tissue repair (Conneely et al 2004). That is, as soon as the local regulation is taken into account the meaning of this regulation, the semantic context (chap. 3.1), has to be understood. The level of mediators and receptors is no longer restricted to the lungs. It is associated with the whole picture of the atopic disease.

Specific interventions on this level (like antihistamines) have a much wider impact, as mediators and receptors are usually widely distributed with quite different equifinal functions.

It is rather difficult to attribute an unspecific treatment to this level. May be the ‘neural therapy’ (chap. 4.8.c), a massage using the segmental nerval interaction or the SHU-MU interventions of acupuncture (chap. 5.6.c) could be seen as an attempt to improve the function and regulation on the mediator/receptor level.

c) general cybernetic regulation

Asthma is a type 2 disease, an imbalance of Th1 and Th2 helper cells. This is the level of the general immune defence. All kind of influences affect this function, including infections or the bacterial and viral flora (chap. 2.8). Interventions interfering with this level are vaccinations, probiotics, antibiotics, antipyretics, etc. The relation between therapeutic measures and the observed effect is often not so clear. Antipyretics change the set point of the parameter ‘body temperature’, but their impact on the function of the immune system has never been assessed appropriately (chap. 6.8). Moreover, antipyretics do a lot of other jobs besides lowering body temperature. For example, they might change apoptosis (chap. 5.3.a) and inner rhythms (chap. 5.3.b). Also antibiotics do not kill only defined specific germs. They interfere with the bacterial flora more generally and might also change the viral flora. The occurrence of Candida albicans after an antibiotic treatment is rather common and might induce a further allergisation (Savolainen et al 1993), although the data for this assumption are not overwhelming.

Considering the former finding that living systems are nonlinear and operate far from the equilibrium (chap. 6.4) it might be expected that the Th1/Th2 relation is not static but constantly shifting around a certain ratio. That is, a slight tendency to allergy and auto-aggression might be transient and in line with a healthy function of the immune system. The attempt to fix a certain ratio (which is probably impossible) would necessarily lead to a decline of robustness. But also all other attempts with a linear goal (a conscious purpose) have to be seen rather critical. ‘Specific interventions’ are not possible on the level of the cybernetic regulation, on the level of the ecological system. The system will react and the reaction is nonlinear, unpredictable and uncontrollable (chap. 2.8; 4.2). Interventions in order to change a certain function and/or a set point are prone to induce network pathologies as long as they fix a variable (chap. 6.7) and do not intend to stimulate a reaction. This has been seen with parathormone which provokes bone loss when given continuously, but fosters bone formation when given intermittently (chap. 6.4).

Another example of a network pathology through the intervention into a cybernetic cycle was a trial with the substance TGN1412 which influences T cell function. This therapy lead to a massive cytokine storm and a multiorgan failure in all six healthy human volunteers (StClair 2008), a typical exponential runaway. Similar events might be expected or at least cannot be excluded with all other attempts to influence a cybernetic function in a linear way. These effects might not be as obvious if the development of the network pathology is rather slow.

d) inner environment

The whole inner environment is a further extension of the level of abstraction. It is not concerned with the performance of a certain task like the immune system. Its aim is the autopoiesis (chap. 4.8). The level of autopoiesis is beyond the specific concept and asthma or atopic disease can hardly be described on this level. In this view they are neither an impairment of breathing nor of the immune defence. A definition on this level has to involve statements about network pathologies and about the meaning of genetics and epigenetics. Thus, no specific treatment can be designed for the inner environment, in special. This does not mean that therapies do not work on this level. They do. In influencing sub-systems a therapy has always a general effect, mostly more than the theory suggests. A therapy with beta blockers, for example, might change the adaptation fundamentally (chap. 6.6).

As a consequence, interventions on this level are necessarily unspecific. A typical example would be the nutrition (chap. 2.8.a) or a therapy with probiotics (chap. 2.8.b). Another therapeutic strategy, which came into the focus of research lately, is physical exercise. A lot of detailed findings show that physical exercise is good for nearly ever function in the body. Even the neurogenesis in the brain is stimulated (overview Blech 2007b). Rather famous became a study showing that in stable coronary artery disease physical exercise is superior to stent therapy (Hambrecht et al 2004). Of course, exercise is also helpful in asthma (Juvonen et al 2008, Williams et al 2008). But the effects of such a therapeutic intervention cannot be assessed in measuring a certain trait or surrogate parameter. Just as it is impossible to assess the therapeutic power of choir singing by measuring cortisol and fibrogen levels (chap. 5.4.a). Every intervention has to be evaluated in its relation to the whole function, in relation to robustness and autopoiesis (chap. 6.2).

Many ‘unspecific’ methods of CAM have explicitly the aim to induce robustness and to facilitate distributed control, most apparently in balneology. On the contrary, orthodox medicine is rarely concerned with this issue, which is a major handicap of the current scientific concept.

e) psycho-social context

This is the level of the human in his context. Many disciplines have been concerned with the human interaction, its emotional implications and its influence onto health.

A forerunner in this field of ‘psychoimmunology’ was Norman Cousins (1915-1990). He provided a lot of examples how and in which way interhuman relations are able to improve or worsen the course of a disease (Cousins 1985, 1990). But Cousins’ approach was human and not scientific. This had been an obstacle in dealing with the scientific community in the sixties and seventies of the last century. It is still a handicap today. In the first place, reductionist research considers human factors as unspecific (chap. 5.4.b). Moreover, reductionism excludes ‘psycho-social’ factors as they are so numerous that the results of one research group often cannot be confirmed by another group with a slightly different angle of observation, a dilemma of psychosomatic medicine ever since. Psychosomatic medicine often relied on the description of single cases (Uexküll 1986) emphasising sometimes the beauty of a theory and not its effectiveness (Haley 1973: 12), which can hardly be established with statistical methods (chap. 5.7).

As a result, interhuman factors are neglected in the medical discourse and in the medical education. It is only consequent that medical students lose (according to the hidden curriculum) their idealism (chap. 3.10) and their empathy (Shapiro 2008) during their studies.

Anyhow, the relevance of the psycho-social level has been generally underestimated. But to understand its importance, a different approach is necessary. This shall be demonstrated with the following example: A lot of diseases is somehow ‘contagious’. They spread along social networks. For obesity it has been shown that this kind of network relation is much more important than any ‘biological cause’ such as genetic disposition (Christakis et al 2007). The same is true for suicide (Bohannon 2006). But also many other diseases seem to follow a similar pattern (Dworaschk 2008), quitting smoking (Christakis/Fowler 2008) and happiness (Fowler/Christakis 2008) included.

The therapeutic impact of such a finding is immediately clear. In order to improve the health of a patient, beneficial networks have to be enforced and the influence of harming networks have to be repressed. A rather straightforward application of this strategy is the method of the weight watchers. Their success in combating obesity is only partly due to the proposed diet (which would represent a therapy of the local or mediator level). The weight watchers strongly rely on the social interaction with weekly meetings, public weighing, mutual support, financial fine, etc. All this helps to achieve a self-set goal (Weight Watchers 2008).

Another example is the therapy of anorexia nervosa with means of structural psychotherapy. Here the aim is to change the family structure and interaction. The caloric treatment of the individual is reserved to life-threatening situations (Minuchin et al 1978). That is, the treatment is less concerned with the pathological trait. The same is true for psychotherapeutic methods which improve the family relation of an asthmatic child supposed to hold the family together (chap. 5.2; 6.8). But also choir singing might reveal its therapeutic power in asthma neither by changing cortisol levels (chap. 5.4.a), nor by promoting a better breathing, but by influencing the social networks and by creating friendships which seem to have a major health protecting influence (Parker-Pope 2009b, Shively et al. 2009).

That is, the effect of interventions on this level is rather indirect and hardly describable with a ‘specific’ mould. These interventions are less concerned with the manipulation of ‘causes’. They are more concerned with solutions (chap. 5.2) which might have nothing to do with what has been defined as a cause before.

f) ethical values and religiosity

A further extension of the frame of abstraction is the question of general ethical and religious values. This subject plays no role in medicine (Mohr 2006). Spiritual values are rarely addressed as such, although newer research addresses the (mostly positive) influence of religion onto health (Nagourney 2008). Religious questions normally appear only in a negative context, when, for example, people refuse life prolonging therapies (Winter et al 2007), or when Jehovah’s witnesses deny blood transfusion.

From a Christian point of view a simple intactness and functioning is no aim as such. Jesus was rather radical about that. In Matthew 5:30 he says: “And if thy right hand offend thee, cut it off, and cast it from thee: for it is profitable for thee that one of thy members should perish, and not that thy whole body should be cast into hell.” Although this might be seen as a metaphorical statement, it shows that every action has to refer to a higher goal. This is true for all religions, although their aims and concerns might differ. But even an atheistic view does not lead to pure egoism. All social research shows that altruism is a basic trait in evolution (University of Leicester 2008). A medicine abstaining from ethical considerations is endangered to become an extension of a simple consumerism.

The physician is continuously confronted with questions concerning the meaning of life. This is not only the case in suicidal patients, but also in all kind of mental disease. Is the aim of the therapy to make the ill an average Joe or Jane? According to which principles is the patient accompanied through the maze of life? This issue is rather prominent in artists where the vicinity of creativity and psychotic experience is well established (Jamison 1993). The list of people who have been diagnosed as manic-depressive (appendix III) reads like a who’s who of Western art (Simon 1995: 192). Mostly it is beneficial for the artists when treated (Zuger 2008a). But the way we as physicians see and support the creativity of our patients plays an important role, even if we do not treat a lot of van Goghs or Pollocks. Creativity is one of the most important tools in the inner development of every person (Petzold/Orth 1996). The question, every physician faces is, in how far (s)he is concerned with such issues like inner development. How important was Proust’s asthma for his creativity (Miller 1956)?

The matter is not different in depression. A major depression has to be treated, but minor depressions might have some meaning in the life of the patient (Wolpert 2008). It might be the motor for compassion (appendix III) or even of spiritual fulfilment.

The physician faces akin challenges in patients with cancer, after cardiac infarction, in obesity and in practically in all problems of life. For families with asthmatic children the question might be crucial in how far it is necessary to change their life-style or the place of living (chap. 6.1). Such questions arise in every regimen, in every ‘therapy by order’ (chap. 5.4.b) a physician establishes. Otherwise the outcome will be poor. The adherence to a certain treatment depends much on its meaning for a certain person (Park 2007). The Alcoholics Anonymous are rather successful in treating alcoholism and they stress extremely the importance of spiritual values. However, mostly it is difficult to assess the health effects of religious/spiritual practice (Park 2007), as they do not translate into a certain defined behaviour.

*

This structure, ranging from very specific to very unspecific interventions, suggests certain traits which have been addressed in one way or another throughout this thesis. Some of them might be summarized as follows:

There is some evidence that

  • the more specific a therapy is, the more it is at risk to be only a first order therapy with a higher danger of inducing rigidity (exception: strategic therapies like parathormone given discontinuously)
  • the more specific an intervention is, the more obvious is the relation of cause and effect
  • the more unspecific a therapy is, the more equifinal mechanisms are induced
  • the higher the level of organisation, the less it is possible to assess a relation between cause and effect
  • the higher the level of organisation, the more structural an intervention has to be

The main problem is in how far such implications can be proved or disproved. The more equifinal processes are involved, the more difficult is efficacy control (chap. 5.7). Specific interventions might be highly effective but do not contribute necessarily to longevity (chap. 2.2). Instead, most important factors for longevity like a healthy lifestyle, a Mediterranean diet and regular physical activity (Universitat Autonoma de Barcelona 2008) never perform as effective in the framework of EBM. One has to be aware of this possibility, which is nothing else than the simple wisdom of the incompleteness theorem (chap. 3.3) that truth and proof are something different.


(1) It is always arguable, or better impossible, to attribute a certain intervention or stimulus to one of the here introduced categories. The aim of this model is more to give a certain feeling for a hierarchy, for the therapeutic pluralism.



previous -- home -- content -- next