Abstract
In the last two editorials, 1,2 we have established that “from bench to bedside,” the supposedly inevitable paradigm of pharmaceutical research, is not a given and that integrative and complementary medicine follows different rules: It is often not (purely) pharmacological. It is applied in the population based on traditional knowledge—independently of research findings or lack thereof. It is based on other models of the development of diseases and their treatment.
In this last of three editorials on research methodology in integrative and complementary medicine, let us focus on the last two of five points in which we have identified differences to biomedicine: It treats each patient individually and uniquely, even when the pathology is the same. It almost always includes multimodal interventions or uses “Whole Medical Systems.” 2
Whole Medical Systems describe a core aspect of integrative and complementary medicine. Contrary to a common misconception, integrative and complementary medicine is not about using willow bark instead of synthetic acetylsalicylic acid for headaches, acupuncture instead of lavage for knee osteoarthritis, or yoga instead of aerobic exercise for high blood pressure. Rather, it is about a unique approach to illness and health, a distinct patho- and salutogenetic concept of how illnesses arise, can be treated, and (usually even more importantly) can be prevented. 3
Accordingly, Whole Medical Systems are self-contained, comprehensive theoretical and practical concepts that have developed parallel to and independently of biomedicine. 4 A classic example is traditional Chinese medicine, which includes not only acupuncture but also a variety of therapeutic methods such as Tai Chi, cupping, and, above all, herbal medicine, as well as its own diagnostic approaches such as pulse and tongue diagnosis. This is all the case because it is based on its own theories and not those of biomedicine. Western traditional medicine systems such as traditional European medicine and homeopathy also have their own theories on the development and treatment of disease and unique diagnostic and therapeutic approaches.
And of course, these therapy systems are not developed in the laboratory and then tested in clinical practice; they have been used in clinical practice for centuries, if not millennia. But they are also not verifiable by the reductionist paradigm of the randomized, placebo-controlled, and double-blind trial. If only because the research questions are different. Instead, comparative effectiveness research is more suitable:
Comparative effectiveness research is research that identifies what approaches work best for improving health. This involves comparing at least two therapy systems in their entirety, instead of breaking them down into the smallest and most easily digestible individual parts that are far removed from clinical practice. 5
Let us look at two examples:
In January 2010, the regional public health system in South Tyrol, Italy, initiated an integrative and complementary medicine program at Merano Hospital, initially for 2 years. The extension beyond these 2 years was linked to positive results in a clinical trial. In this study, 275 women with breast cancer were randomized into one of two therapy arms: standard oncological treatment or integrative oncology, consisting of standard oncological treatment and complementary medical procedures. In the second group, a total of 15 different individualized complementary medicine interventions were used alone or in combination. Roughly half of the patients in this group were treated with acupuncture, homeopathy, or orthomolecular therapy each; procedures such as healing touch or neural therapy were used less frequently.
Compared to standard treatment, quality of life improved significantly and clinically relevantly in the integrative medicine group. 6 But what does this result mean? Does acupuncture, homeopathy, or orthomolecular therapy improve the quality of life of breast cancer patients? Of course not. It means that the tested therapy system, integrative oncology, has this effect. No statements can be made about individual procedures, but that was actually never the question.
Another example: in a randomized trial, classical individualized homeopathy was compared with standard therapy in 410 cancer patients. Phosphorus was prescribed most frequently, but numerous other homeopathic remedies were also used. The main outcome parameter was again quality of life, which showed a significant superiority of classical homeopathy. 7
Can an effect of homeopathic phosphorus be derived from this? Of course not, you would need a conventional comparison of phosphorus versus placebo. So an effect of individually selected homeopathic remedies? Neither, as this would also require a placebo-controlled comparison.
The actual conclusion of the study can be found in the German guideline “Complementary medicine for oncology patients,” which gives the recommendation: “Data are available from an randomized trial on the use of classical homeopathy. This specific anamnesis in combination with individual prescription of remedies can be considered in order to improve the quality of life in oncological patients in addition to tumor therapy.” But also: “There are no sufficient data from RCTs on the efficacy of individual homeopathic remedies registered in Germany (established indication and individual indication) alone on mortality and disease- and therapy-associated morbidity and quality of life in oncology patients. No recommendation can be given for or against the use of homeopathic single remedies registered in Germany alone in these patients.” 8 So you can try classical homeopathy because it is evidence-based. But nobody knows whether homeopathic remedies work.
And this conclusion is, of course, correct: the whole package of homeopathy works. We cannot deduce from the study whether it is really the homeopathic remedies that work, or the initial anamnesis, which is definitely similar to psychotherapy, or a combination of both. But we do not want that either; the study only asked about the total package, as that is what patients receive in reality. Homeopathic anamnesis as a single therapy is just as rare in the reality of care as placebo pills as a standard therapy.
In summary, integrative and complementary medicine often requires modified research methods. This is because it tests more complex interventions but also because it asks different, less specific, but more comprehensive questions.
The randomized, placebo-controlled, and double-blind trial can examine individual components of the whole medical system, but this usually only makes sense once the system itself has been recognized as effective. In order to investigate how much of this effect is actually specific and how much is context. Context, which, of course, is an essential part of the system. After all, is it really relevant for patients whether the relief of their symptoms is a specific drug effect or partly a placebo effect, a context effect, or an attention effect? Or is the more important question not WHETHER the patient feels better rather than WHY they feel better?
In this way, integrative and complementary medicine reverses the sequence of stages in medical research: it does not go from preclinical studies via phase 1, 2, and 3 studies into practice. It starts in practice, then scientifically examines the overall system, then individual components, and only when the effect is relatively certain and only out of academic interest, the mechanisms of action. 9 From practice to research. From bedside to bench. Whereby bedside is usually more important for patients than bench.
Stay healthy and stay curious.