Placebos and Psychosomatic Disease Part 2

An interview with Dr. Bob Ader

back to part 1

Engel’s Biopsychosocial Model, Psychosomatic Medicine And Stress

PJR: I’m glad you mentioned George Engel’s influence for several reasons. He was your immediate superior, and it is hard to think of anyone better equipped to objectively criticize your hypotheses or provide support if he thought they were correct. A towering figure in medicine as well as psychiatry, George is best known for proposing a biospsychosocial model of health in a 1977 article in Science. It posited that health was best understood in terms of a combination of biological, psychological, and social factors, rather than purely biological terms. This was in sharp contrast to the traditional medical belief that every illness or disease was due to some pathogen, injury, genetic or developmental abnormality. I had the pleasure of meeting him when you invited me to give a talk on Stress and Cancer at the University of Rochester, which was attended by Art Schmale and some of the other pioneers in this field. George Engel drove me back to the airport, during which we discussed mutual friends, like Stewart Wolf, who was his classmate at Johns Hopkins, my views on Hans Selye’s contributions, and so many other topics that the time flew and we continued our conversation for another five or ten minutes after we reached our destination. He eagerly accepted my invitation to do a Newsletter interview and I always regretted that his untimely death prevented this. You became the George L. Engel Professor of Psychosocial Medicine at Rochester, and I always thought this was particularly appropriate, since so many of your views were similar, if not congruent. For example, back in 1974, before biopsychosocial and psychoneuroimmunology were invented, both of you published separate papers in different journals, which essentially argued that psychosomatic research does not deal with psychosocial factors as a cause of disease, but rather in altering the individual’s susceptibility to disease. The term psychosomatic was introduced into American medicine by Flanders Dunbar around the same time as Hans Selye’s initial publication on stress and I was privileged to work and coauthor articles with both of these luminaries. Dunbar founded what would later be the American Psychosomatic Society as well as its journal, Psychosomatic Medicine, and served as its editor for the first eight years. Her research focused on an attempt to relate different disorders to specific emotional conflicts and/or personality patterns. Like Selye’s concept of nonspecific responses to stress as a cause of disease, her views, such as the notion that rheumatic fever was largely due to autoeroticism and homosexuality, were also later rejected. This led to considerable controversy, especially when psychosomatic illnesses began to be viewed as a form of malingering. Psychosomatic was replaced by somatoform, neurotic and stress-related disorders by the World Health Organization and  Psychosomatic Medicine is now often referred to as Behavioral Medicine. Among your many honors, like George Engel and Stewart Wolf, you also served as President of the Psychosomatic Society, and I wondered what your current views were on psychosomatic disease and the role of stress.


RA: There is no such thing as psychosomatic disease! Modern psychosomatic medicine deals with the role of psychosocial factors (including stressors) that contribute to (but not by themselves, cause) the development and/or progression of disease. If I were to accept the argument that a particular disease was psychosomatic, I would, by definition, have to agree that some diseases were not psychosomatic. If, however, it can be shown that psychosocial factors can influence one particular disease, I must allow for the possibility that the interaction among biological, psychological and social variables could, to a greater or lesser extent, influence the course of all diseases. In the search for single causes for single effects, H. pylori was identified and labeled as the cause of duodenal ulcers. However, most people harbor H. pylori but only a fraction of these develop duodenal ulcers. Thus, H. pylori, the ostensible cause of ulcers, may be a necessary component, but it is not sufficient—its mere presence does not mean that an ulcer will develop. Psychosocial and biological factors can also be essential ingredients. For the most part, the mechanisms underlying the effects of stressors or stress on immune function and disease can only be outlined in general terms. Affective responses to what are perceived to be stressful circumstances are accompanied by autonomic nervous system and neuroendocrine changes capable of influencing immune function and thereby altering susceptibility to a variety of diseases. The majority of stress research emphasizes the common responses to stressors (e.g., adrenocortical responses). If these were the changes responsible for changes in disease susceptibility, we would expect that all stressors would exact the same effects. In fact, different stressors have different effects on the same experimentally induced disease process in animal studies and the same stressor can have different effects on different disease processes. It would seem, then, that it is not the common, nonspecific effects but the uncommon, idiosyncratic effects of stressors that are responsible for the alterations in susceptibility to different disease processes.

Has Psychoneuroimmunology Research Resulted In Any Clinical Benefits?


PJR: The discovery that H. pylori was the cause of peptic ulcers, led many to conclude that the prior widespread belief in the role of stress was hogwash, despite abundant anecdotal support and Stewart Wolf’s direct observation of this. But stress lowers resistance to bacterial infections, so its contributory role seems quite plausible. Similarly, tuberculosis cannot occur unless the tubercle bacillus is present, but many individuals who harbor this organism do not develop clinical tuberculosis, or do so only after exposure to an 10 increase in stress related hormones like cortisone. Stress has also been shown to be associated with an exacerbation in autoimmune inflammatory diseases like psoriasis. As you may recall, I had invited you to organize a session on “How Can Basic Psychoneuroimmunology Research Be Put To Practical Use” at our 1995 International Congress on Stress in Switzerland. Unfortunately, you could not attend due to illness but did arrange to have Nick Cohen chair this session. It included presentations on the effect of psychological intervention on immune and inflammatory responses, psychoimmune factors in juvenile rheumatoid arthritis and the impact of emotional status on cancer. Although these confirmed the important effects that the mind and emotions could have on immune system function, how could these assist physicians in their efforts to prevent or treat disease? As Nick noted, “despite the large body of evidence validating psychoneuroimmunology as a bona fide interdisciplinary field with potential clinical relevance, the current applications of research in this area are still more in the realm of wishful thinking than in reality.” I was therefore pleased to see your recent paper in Psychosomatic Medicine dealing with psoriasis and wondered if you could tell us about this and any other future clinical applications.


RA: In the paper you referred to, we hypothesized that psoriasis patients treated under a partial schedule of pharmacologic (corticosteroid) reinforcement would show less severe symptoms and relapse than patients given the same amount of drug under standard conditions (continuous reinforcement). To paraphrase our abstract, this was a double blind, simple randomization intervention conducted with 46 patients from Stanford and Rochester. Initially, lesions were treated with 0.1% acetonide triamcinolone under standard treatment conditions. Thereafter, a Standard Therapy group stayed on continuous reinforcement (active drug every treatment) with 100% of the initial dose; Partial Reinforcement patients received a full dose 25-50% of the time and placebo medication other times; Dose Control patients received continuous reinforcement with 25-50% of the initial dose. Severity of disease was rated weekly on a 9-point Psoriasis Severity Scale. Severity scores in California neither supported nor refuted the hypothesis. In New York, partial reinforcement effected a greater reduction in lesion severity than Dose Control conditions and did not differ from Standard Therapy patients receiving 2-4 times more drug. For the entire population, the incidence of relapse under partial reinforcement (26.7%) was lower than in Dose Control patients (61.5%) and did not differ from full-dose treatment (22.2%). It appeared that a partial schedule of pharmacotherapeutic reinforcement could maintain psoriasis patients with a cumulative amount of corticosteroid that was relatively ineffective when administered under standard treatment conditions. It is conceivable that corticosteroid administration only one quarter or half as frequently as currently prescribed is sufficient to treat psoriasis and that possibility will be addressed in future 11 studies. We posit, however, that these preliminary observations implicate conditioning processes in (and for the design of) pharmacotherapy regimens. Adding a behavioral dimension to the design of drug treatment protocols changes the equation for understanding drug effects and is likely to stimulate new interdisciplinary research in neuropharmacology and behavioral pharmacology. Partial schedules of reinforcement might:

• Reduce the total amount of drug required for treating various disorders

• Reduce deleterious side effects and thus increase adherence to treatment

 • Extend pharmacotherapeutic effects (increase resistance to extinction)

• Reduce very substantially the cost of long-term drug treatments

Reducing costs may be the most important benefit and in some instances this and the other rewards noted above have already been achieved.


Why Do You Believe That Placebo Effects Are Conditioned Responses?

 PJR: At our 2000 International Congress On Stress in Hawaii, we devoted a session to “The Power Of The Placebo” that was chaired by Karen Olness and featured presentations by Stewart Wolf, Herb Benson, Wayne Jonas and other notables. You were again unable to attend but submitted an abstract entitled “Are Placebo Effects Conditioned Responses?” in which you wrote If the response to placebo is a conditioned response, there is an alternative to the typical administration of drug or placebo; namely, administration of drug and placebo — a partial schedule of reinforcement. In effect, reinforcement schedule, or the “active drug:placebo ratio,” represents an additional dimension of drug treatment protocols and an alternative means of titrating cumulative drug dose that may enable one to maintain some physiological responses within homeostatic limits using lower cumulative amounts of active medication. Karen’s presentation was also entitled “Are Placebo Effects Conditioned Responses? (The Macadamia Chocolate Decaf Effect)”. In it, she described an 11-year-old girl with systemic lupus and severe complications, for which she was to receive intravenous Cytoxan. Her mother, who was a psychologist, had read the study you and Nick did on applying conditioning in a mouse model of systemic lupus being treated with Cytoxan paired with a saccharin solution as the conditioned stimulus. You showed that the saccharin solution alone could delay the onset of disease and reduce the dose of Cytoxan needed to have a therapeutic effect. The mother wanted the doctors to use a similar protocol on her daughter when she received the intravenous Cytoxan. In this case, the conditioned stimuli were cod liver oil (taste) and the scent from a rose perfume Contém 1g (smell). As I recall, the conditioned stimuli were given over a 15-month period during which it was possible to present only the conditioned stimuli and thereby reduce the frequency of intravenous Cytoxan infusions, and her daughter did well for 8 years. However, I find it hard to believe that all placebo phenomena fall under the category of conditioned responses. For example, how would this apply to Stewart Wolf’s study demonstrating the antinausea effect of ipecac 12 in pregnant women with morning sickness? Has your position on placebos and conditioning changed over the past decade?


RA: No, my position has not changed. Well, actually, it has changed: I’m now more convinced that placebo effects are learned responses and that some placebo responses reflect conditioned pharmacotherapeutic effects. This hypothetical statement is not restricted to placebo responses involving the immune system—nor by the sensory modality of the conditioned stimuli, all of which are, by definition, nervous system stimuli. Indeed, we may have to distinguish between different kinds (and “explanations”) of placebo effects such as faith healing, belief systems, verbally-induced expectations, direct instruction, authority pronouncements, observation and conditioning, all of which can induce expectations. While all placebo responses do not involve conditioning, it seems to me that, in the final analysis, they all involve learning; they are derived from experience. Who on this earth, for example, has not been helped by somebody (physician, shaman, witch doctor, teacher, parent) at some time? The attempt to attribute a particular placebo response to one or another of these explanations is complicated by the fact that more than a single kind of learning may be involved. Studies that attempt to pit one explanation against another can be difficult to unravel because there is no way to equate, for example, the amount or value of the information communicated by a verbally-induced expectation with the information value of prior conditioning. As you pointed out, clinical research and drug evaluation studies have, for the most part, adhered to the model in which a placebo is administered in order to evaluate the efficacy of pharmacotherapies or to define the pharmacologic (as opposed to the psychologic) action of a drug. Thus, research has been directed to characterizations of beliefs and expectancies, including those induced by the instructions to subjects, and characterizations of the subjects who respond to placebos. Much placebo research also derives from an effort to define the “true” unadulterated action of a drug, rather than an effort to understand the nature of the placebo effect and its therapeutic actions. There have been repeated but unanswered calls for studies of the placebo effect as a phenomenon that may have clinical implications in its own right. And that’s what we’re trying to do by exploring the clinical implications of placebo responses from a learning perspective. The conditioning model of placebo effects challenges the very definition of a placebo response as a nonspecific response to an inert agent. Perhaps the response to a placebo is a two-stage process. The initial response satisfies the definition of a placebo response as a nonspecific response to a therapeutically neutral stimulus based, perhaps, on the individual’s experiences with healers of one sort or another. As a second stage, I would suggest that some placebo responses are neither nonspecific nor are placebos (conditioned stimuli) inert. If a conditioned stimulus can evoke a response that approximates the results seen with the unconditioned 13 stimulus, you could hardly call it neutral or inert. And if that conditioned stimulus, over time, elicits conditioned physiological responses that resemble the responses unconditionally elicited by a drug or other therapeutic intervention, the response can hardly be called nonspecific. I can refer to it as a conditioned pharmacotherapeutic response. The clinical community may or may not want to call it a placebo response. They may wish to retain an entrenched concept and definition that has not, however, clarified our understanding of the placebo effect or its therapeutic potential and has, I believe, misdirected the search for models amenable to experimental analysis and from which new, testable hypotheses can be derived.


Some Parallels With Selye And Stress And Psychoelectroneuroimmunology?

PJR:I see certain intriguing similarities between you and Hans Selye. Both of you described your discoveries as “serendipitous”, which implies that they were simply lucky accidents. However, as Pasteur emphasized, “Chance only favors the prepared mind”. Both of you introduced novel concepts that launched such a huge avalanche of research that it became impossible to keep up to date. Stress became a popular buzzword that was used to describe very different things and psychoneuroimmunology is at risk for suffering a similar fate, as the allure of its cachet is abused by charlatans who want to profit from its credibility and scientific patina. In 1972, Selye developed a reticulosarcoma, a normally fatal malignancy, from which he completely recovered. He refused chemotherapy, and attributed his good fortune not to any other treatment received, but rather his very firm determination to continue living so that he could complete his important research activities. Based on reports of similar experiences and spontaneous remissions, he was convinced that a firm faith and fierce determination could retard or reverse cancer growth. I vividly remember getting the news about your pancreatic cancer but it’s hard to believe that this was over five years ago since normal life expectancy is a year or less after detection. Before that you suffered a heart attack from which you were not expected to recover, and subsequent by-pass surgery. Since then, you have had one or more defibrillators implanted, several bladder cancers removed, a pulmonary embolism and most recently, a gastrectomy for a stomach tumor. Could your amazing ability to overcome all these obstacles also stem from a strong desire to continue your research, similar to Selye’s? How such effects are mediated is not clear, but as you may recall, I used the term psychoelectroneuroimmunology to refer to subtle energy communication pathways in a book we both contributed to. As Thoreau wrote, “To know that we know what we know, and that we do not know what we do not know, that is true knowledge.” I suspect that the second half of this aphorism is particularly pertinent to the present situation, and in the limited space available, wondered what your speculations might be on the future of psychoneuroimmunology. 


 RA: I could go on and on reciting what we have yet to learn, what needs to be done and how to do it. But that would be my conservative self. Given the space that remains, let me look further ahead and speculate very broadly, but briefly, on the potential health implications of psychoneuroimmunology. First, I would repeat the central premise of psychoneuroimmunology, which is that immunoregulatory processes are part of a single complex interacting network of adaptive responses. As such, psychoneuroimmunology does not recognize the arbitrary and illusory boundaries of the disciplines into which we have divided the biomedical sciences. We know a great deal about the actions of different components of the immune system and about the neuroendocrine responses to stressors, but we are only just beginning to understand how these systems interact in health and disease. Or, as Lewis Thomas put it: “You’ll never understand how bees make honey no matter how carefully you dissect a single bee.” If you really accept the proposition that there’s only a single, integrated defense system, new questions emerge. Central among these may be the need to reconsider the “cause” or development and the treatment of what we now speak of as “neurological” diseases, “endocrine” diseases or “immune system” disorders. How do early life experiences, including stress, influence the development of the immune system? Conversely, how do immunological challenges early in life or prenatally influence subsequent behavior and neuroendocrine function? Research in these areas has already begun. And what about a focusing on neuroendocrine interventions in the treatment of autoimmune disorders or the applicability of immunological interventions in the treatment of neuroendocrine disorders? It seems to me that basic research on the interactions among behavior, neuroendocrine and immune processes has a bright future that promises new developments in our understanding of adaptive processes with profound consequences for the maintenance of health and for the treatment of disease.


PJR: Many thanks, Bob, for this trip down memory lane and sharing your thoughts for the future.