The somatoform conundrum: a question of nosological values

Commentary and Perspective From time to time, the Journal receives manuscripts that can be thought of as opinion pieces, essays, or editorial comment on matters of topical interest. Such submissions will be refereed in the usual fashion and, if suitable, published in this section. The Editorial Board invites Letters to the Editor or rebutting commentary with the understanding that all submissions are subject to editing.

The Somatoform Conundrum: A Question of Nosological Values Robert D. Martin, M.D. Abstract: This paper critically reviews the values and perspectives evident in the formulation of the category of Somatoform Disorders in the widely used Diagnostic and Statistical Manual-IV (DSM-IV) [1] of the American Psychiatric Association. The conflict of values evident in the DSM Committee’s elimination of causal issues and including only descriptive criteria in the final classification is emphasized. In this approach, causation was dismissed and only a description of popular clinical entities was allowed. It becomes clear that there was a conflict between causal and descriptive factors in the formation of the current classifications. The dubious logic and the inconsistencies that underlie the linking of the various diagnoses under the Somatoform Disorder (SD) rubric is presented. A heuristic model of the interaction of psychological and somatic variables is described. These variables act in a causal fashion and should be the basis for any nosological system of the SDs. There is ample evidence to support the importance of causal factors in the development of SDs, and the current descriptive nomenclature does not do justice to the rapidly growing field of Psychosomatic Medicine. A new scheme is proposed, using the presence or absence of stress as the connecting link for the diagnoses offered. The old nosology enhances the perpetuation of the mind-body dichotomy. Stress, representing the way in which the environment impinges on the human being, allows the mind-body dichotomy to be eliminated. The preferred synthesis of mind-body interaction as partly the result of experience in the environment becomes the basis for the new nosology. © 1999 Elsevier Science Inc.

Albert Einstein College of Medicine, Long Island Jewish Medical Center Campus, Consultation and Liaison Service, Room C-46, New Hyde Park, New York Address reprint requests to: R.D. Martin, M.D., 40 Stoner Avenue, Great Neck, NY, 11021

General Hospital Psychiatry 21, 177–186, 1999 © 1999 Elsevier Science Inc. All rights reserved. 655 Avenue of the Americas, New York, NY 10010

Introduction Any consideration of Somatoform Disorders (SDs) conjures up images of the body in disturbance due to the release of improperly directed emotions. But, as occurs with many struggles designed to resolve the enigmas destined for official designation, The Diagnostic and Statistical Manuel-IV (DSM-IV) of the American Psychiatric Association [1] does little to explain the underlying arguments that substantiate its reasons behind this classification. That in itself is somewhat excusable, but the officious screen, behind which the presumed certainties are protected, suggests there is more revealed with the diagnoses in this category than is in fact the case. In the DSM committee’s ardent attempt to be “scientific,” the hard-won efforts of the various discussions and surveys that did so much for other classifications, failed when it came to SDs. It should prove useful, therefore, to see what was left after such deliberations. Since this is the defining diagnostic schema for residents, insurance executives, and that lately trampled upon hoi polloi, the office psychiatrists, the benefits of its close examination may indicate that there is a more logically satisfying alternate.

The Conundrum Any clinician who has dealt with the nosology of SDs is mystified and confused by patients who are diagnosed with SD. Patients beliefs in their symptoms, their willingness to accept extensive testing, even multiple surgeries, and their persistence in holding on to symptoms for which no reasonable

177 ISSN 0163-8343/99/$–see front matter PII S0163-8343(98)00081-4

R.D. Martin

explanation is forthcoming, impresses even the most cynical clinician. Rarely does one confront this disorder without wondering how the presentation comes about. It is never known if some unexplained biological process underlies the specific discomfort for which the patient seeks relief. Since every diagnostic category in DSM-IV under the SDs alludes to a patient’s profile in some form, the characteristics of each case are presumed to be distinct. This is far from the case. Because the classifications seem mutually exclusive and inclusive within themselves, but on closer inspection are not, the essence of what constitutes a SD and what it is that makes each of those designated as somatoform are quite different, becomes a conundrum, “. . . a problem having only a conjectural answer” [2]. The question is whether Somatization Disorder, a subcategory of SDs, can contain Conversion or Hypochondriasis; whether the Pain Disorder patient is best served by being viewed as a Somatization Disorder or a special kind of physical Pain Disorder. These issues make the use of this category an uncomfortable process. One is also struck by the complete absence of interest in what happens to the physiology of the body in this section of DSM-IV. The implication is that the symptoms being revealed can only derive from the psychological issues. Without realizing it, it seems, the authors of these categories perpetuate the mind (psychological)-body (physiology never included in the psychological) dichotomy most researchers and clinicians would prefer to avoid. The mind-body dilemma has been the subject of many articles and books, most of which agree that the separation of the two in understanding psychosomatic issues should not be a part of any diagnostic nomenclature. Yet the separation persists in DSM-IV. Most importantly, there is an implied abhorrence of the idea that psychosomatic causality may play a role in the patients seen under the somatoform rubric. Psychosomatic causality is the concept that mindbody interactions are responsible for disease production. Its opposite would be the contention that mind or body, alone, separate, apart, is operating in an individual manner to give its own group of disorders. Under this latter view, there are mind illnesses and body illnesses; they cannot be considered as intermingling. From their separate divisions each have no currently describable causes that lead to the other. Mind problems do not cause body problems. Body problems do not cause mind problems. Their interaction causally leading to problems does not occur. The diagnostic categories now available are then

178

only descriptions of what is seen or experienced. These categories are derived by consensus. There is no particular thread connecting them other than the link that they seem to fit, and there is general agreement within the DSM-IV committee that it would be useful to include them in the diagnosis list. This is the descriptive approach. It becomes apparent that causal ideas and descriptive ideas are viewed as incongruent. By this argument, two arenas of thought are created. A conflict arises between causal and descriptive, and the winner is chosen by committee. The battle is then decided for descriptive in a most arbitrary and inexplicable manner. What is worse, though the two values of description and cause are at war, the war is not revealed. The observer and reader of DSM-IV is left to believe that those who created this classification have considered causality in all its ramifications and have carefully concluded causality does not (or cannot) apply. All evidence available for those who follow the development of psychosomatic research and concepts suggests otherwise. It appears to them as if the purely descriptive nature of the current nosology is confusing, inconsistent, and has no underlying associative schema.

Patient Examples It is instructive to consider, for example, a fictional though typical and representative patient. Mrs. Jones is a 38-year-old woman who presents with pains throughout her body. Her internist tells her it is Myofascial Syndrome. His drugs do not work. She is sent to the psychiatrist for his medicines because the internist told her, in somewhat cryptic terms, that the drugs the psychiatrist uses are sometimes more effective and the psychiatrist knows how to use them best. The internist secretly thinks she has a depression or her symptoms have no “medical” basis, but does not tell her this. He is also puzzled by the seemingly legitimate punctate pain spots he has elicited on multiple occasions. The psychiatric interview shows she has some depression but there are also many stress factors of long duration. Where is this woman placed nosologically? In at least six different categories to start. She can be shunted aside as an Axis III presumptive; she can be called Somatization Disorder, defying the number of symptoms required for this classification; Conversion is a possibility even though the immediate history reveals no specific psychodynamic issue; Pain Disorder could be used as long as her

The Somatoform Conundrum

Depression is not considered the causal agent; and, if she is worried, Hypochondriasis. There is always the catch-all Psychological Factors Affecting Physical Illness, completely aside from the Somatoform category. The only one she is clearly not is Body Dysmorphic Disorder. It is clear, then, that as her diagnosis is considered, the complexities mount. If one remains purely descriptive and concedes to an official somatoform diagnosis, causal factors are presumed irrelevant. Unknown does not mean irrelevant. If causal factors are relegated to the “other doctors” in the medical profession to explicate by being placed on Axis III, then it is implied that there is no somatoform interest in the case. If the presentation is physical, the psychological need not be considered a factor in the presentation itself. If it is psychological, then the physical aspects of this syndrome can be avoided, diminished, or simply discounted. One is not allowed to have it both ways. It is a pressing question as to why the design of DSM-IV does not allow it both ways. The unsatisfactory nature of DSM-IV becomes clearer and the diagnostic procedure more frustrating. Each diagnosis insists on exclusory criteria to eliminate the other diagnoses. The maze of uncertainty mounts. Here is the conundrum. Why can one not diagnose the patient as suffering from intertwined psychological and physical factors? Why is it the physical and mental aspects of Myofascial Syndrome cannot co-exist and even be highlighted in the diagnostic nomenclature? At the outset of DSM-IV we are thrown into this type of quandary: “The common feature of the SDs is the presence of physical symptoms that suggest a general medical condition (hence, the term somatoform) and are not fully explained by a general medical condition, by the direct effects of a substance, or by another mental disorder (e.g., Panic Disorder)” [1]. A physical symptom not fully explained by a medical condition may be the pain, if you accept Fibromyalgia Syndrome as a medical syndrome (which some do not). So the physical symptom in this case may or may not be “medical.” If it is medical, does one use “somatoform,” or does one merely place it on Axis III? Does Axis III exclude somatoform? Can both exist? The manual implies that the pain and the medical condition should be exclusive. Anyone who has studied this condition knows this is, more often than not, impossible. And this is just the beginning. In another part of the same cited DSM-IV [1], Pain Disorder is considered somatoform if “Psychological factors are judged to play a significant role in the onset, severity, exacerbation, or maintenance of the

pain.” Who is to say in this case how much was due to the physical aspects of this controversial syndrome (fibromyalgia pain) and how much to the emotions of the patient. This can be stated even more emphatically in another example, multiple sclerosis. Here there is a known lesion, scattered white matter degeneration, but the psychological factors that can enhance or suppress the symptoms and signs are often discernable. Clinical experience demonstrates that life stresses contribute to the onset or severity of this largely physical illness. One wants to call forth both the somatoform and the biological. The overlapping between the two is brought into relief as there is no controversy regarding the physical basis for multiple sclerosis. It is possible, in practice, to see a patient with genuine multiple sclerosis with numerous contributing psychological factors. The presence of pain, for example, may be physical or psychological. It follows that there should be a classification in which multiple sclerosis would properly, by itself, be a classification in the Somatoform category. It would seem that most of the problems in the nosology of SDs arise from the attempt to have categories that demand exclusory criteria. The implication of the “this or not this,” approach is that overlaps and intertwines cannot exist. This creates an underlying fallacy inconsistent with the knowledge and research of the moment. In official psychiatry’s desperate attempt to be “scientific” in the face of its being perceived as having inadequate knowledge, it has leaped to a state of confidence, using a logic of exclusory criteria that simply is not viable when the world of psychosomatic findings and the clinic is surveyed. Psychiatry’s uncertainties over considerations of both causal and descriptive values must be declared. A more forthright and sophisticated approach than that in DSM-IV would be to consider the limitations of current knowledge, but use admittedly descriptive and causal nosology, allowing as many possibilities to emerge as needed. This should be done so that the diagnoses used are based on as consistent principles as can be attained. The issue of consistency in nomenclature becomes an immediate problem when considering the classifications grouped under Somatoform Disorders. There is value in surveying the logic of including these diagnoses in DSM-IV. By so doing, the arbitrary nature of the groupings will become clear and the basis for recognizing an alternative form can be established.

179

R.D. Martin

Other Authors A brief review of the somatoform literature is most relevant. Kirmayer and Robbins are outstanding in this regard. In “Functional Somatic Syndromes,” [3] they discussed “Latent-variable models of somatic distress,” in which it is argued that the overlapping symptomatology of somatoform disorders may be the cause of arbitrarily creating discrete categories “. . . by discounting co-occurring symptoms in other bodily systems.” The “clustering” of symptoms allowed a classification to be contrived. This, of course, is a statistical approach that defies conceptual forms. Symptoms do cluster, therefore the cluster has reality. This is the basis for DSM-IV’s Somatization Disorders and does not easily lend itself to all the somatoform diagnoses. This paper illustrates well the problems inherent in classification, even clarifies how one might better appreciate multiple factors in psychosomatic syndromes, but it does not directly address the persistent problems of DSM-IV. Kirmayer [4] views “Psychosomatic explanations . . . [as] . . . one form of metaphoric elaboration, but does not elucidate in this paper an overall form of classification. He emphasizes the limitations of “. . . the imposition of authoritative meaning through diagnosis,” and the constrictive nature of these diagnoses. This, of course, is true. However, he does not address the problems of inconsistencies within one diagnostic schema. That “. . . epistemic constraints of the clinical setting result in indeterminate diagnoses” [4] may be true, but the problem of DSM-IV is another matter. In this DSM-IV case, the nature of the nomenclature is restrictive to the extreme extent that current integrative psychosomatic theory is displaced inconsistently and without restraint of logic (as will be demonstrated). If it is true, as Bass and Murphy declare [5], that Somatization is best conceived as a process in which many symptoms evolve out of a complex mixture of relevant variables, then this leaves Somatoform as a higher classification not addressed. Surely, all the somatoform diagnoses should be viewed as part of a whole. Nevertheless, these authors uphold the established view that clumps of descriptions are superior to underlying causal connections. Is hypochondriasis a process? Certainly. Is Conversion? Of course. If so, should not the particular processes that lead to these diagnoses be addressed? Should cause be considered? Is description sufficient. The answers to these questions is the purpose of this paper. Pilowsky [6] attempts to reclassify somatoform problems based on “abnormal illness behavior.” He

180

gives many helpful suggestions as to how this may be done. This approach depends greatly on patient reportage, various measurements, and so forth. There is merit in this approach, but it does not emphasize the mind-body interaction. Rather, questionnaires play a large role in determining the presence of illness behavior. This meritorious approach avoids the problems present in ulcerative colitis, arthritis, or other “psychosomatic” illnesses in which physical processes are receiving interference or influence from psychological factors. Kirmayer and Taillefer [7] present a similar view to the one herein, that these diagnoses “. . . reify patterns of illness behavior that cut across other psychiatric disorders as discrete conditions.” However, their concern is focused on the problem of biological and psychological reductionism, and does not address the problem of underlying constructs. They fear diagnoses will underestimate social issues and steer the clinician away from addressing psychosocial problems. This perspective, as with Pilowsky [6] and so many others, does not fully appreciate the integration of physiological reactivity with individual-social interaction. There is a tendency to diminish the importance of any parameter in the complex mind-body mix, and a concomitant tendency to emphasize one over the other (e.g., “social contingencies”). The future of understanding biopsychosocial complex interactions would be better served by an appreciation of their varying valences in illness production, not by espousing one variable over the other in a global sense. It follows, then, that many authors have puzzled over the meaning of somatization and somatoform. Kimball [8] struggled over how to view the multiple variables of causality in psychosomatic diagnosis, sadly contributing to the present-DSM-IV ambiguities. By writing of causal factors whose “exact relationship is unknown” he unknowingly offered up the petard to causality entirely. Pilowsky, Kirmayer, Robbins, and others, have tried to avoid mind-body interactions by contriving what they tend to call the “patient behavior” paradigm. This certainly has value. Abnormal behavior no doubt plays a large role in illness formation. But this ignores how the psychological variable enters into body physical processes and alters disease formation. This must be a part of any somatoform schema. Statistical methods and conceptual groupings have assisted the categorizations. However, no one has tried a schema that integrates various diagnoses under a rubric that will facilitate research and understanding of the particular mind-body

The Somatoform Conundrum

framework which occupies so much current research. The DSM-IV Somatoform category is simply a grouping of favorites, not a conceptual framework. What follows is a criticism of DSM-IV and an attempt to construct a reasoned and reasonable framework in which all psychosomatic processes can be conceptualized. It is a modest attempt which no doubt will require many modifications as others contribute to its refinement, but it is a step away from the avoidance of causality, regardless of complexity, that has marred the psychiatric somatoform nomenclature to date.

Somatization Disorder Under Somatoform Disorders there are six subcategories besides the “Not Otherwise Specified” which ends the group. Each of these has a slightly different logic basis. The first, Somatization Disorder, depends on a kind of symptom listing based on work done by Perley and Guze [9] at the Washington University research group in St. Louis. In their enthusiasm, these researchers and others felt confident that noting women’s multiple symptoms (e.g., four pain symptoms, two gastrointestinal symptoms, one sexual symptom, and one pseudo-neurological symptom) plus the criteria of chronicity beginning before the age of 30 in a woman, would make a reliable classification [1]. The statistics of reliability obtained when this strategy was used in the clinic belied the problems inherent in the conceptual confusion it created. Some authors, such as Kirmayer and Robbins [10] seemed to feel comfortable in this arena, and this style of nosology was included in DSM-IV. However, usage was another matter. Depressive patients have multiple somatic symptoms. There is no place in DSM-IV for Somatization in Depression, or Anxiety, or Conversion, for that matter. If the number of symptoms in Depression qualify, what is more salient, the diagnosis of Depression or Somatization disorder? If the number of symptoms goes below the criteria in DSM-IV, does this become Not Otherwise Specified? Is that valuable? So, the number of overlaps with other diagnostic categories is considerable, and the idea of diagnosis by number (however “reliable”) is distasteful to the conceptual minded. It remains a fact that this is the thinking, the diagnosis by number, of this category. If the patient has the statistic, the patient has the diagnosis. When we move to Conversion Disorder our logical basis for diagnosis must shift completely. We are, then, in an entirely different realm of consideration.

Conversion Disorder The primary, controversial point in the category of Conversion Disorder is evidenced by criterion B. “Psychological factors are judged to be associated with the symptom or deficit because the initiation or exacerbation of the symptom or deficit is preceded by conflicts or other stressors” [1]. Aside from the intellectual deficiency of this statement (too lengthy to refute here), it can be briefly stated that the inclusion of stressors and conflicts in the same criterion is confounding. Conversion is a term used by Freud and Breuer [11]. In “Studies on Hysteria,” Breuer writes: “The excitation arising from the affective idea is converted into a somatic phenomenon.” Whether one agrees with the psychoanalytic concepts that gave rise to the use of “conversion,” that the term still remains in use cannot exclude an understanding of its origins which are based on the mind-body interaction and not on stress as an environmental experience producing discomfort or disturbance. Stress may be associated during the developmental process with conflict; stress does not require conflict for its influence on the individual. Conversion does. The two cannot be included in the same phrasing without considerable explanation, because they indicate two very different mechanisms. Their both being included dissolves the meaning of the word “conversion” as it originated and undermines any new interpretation. This allusion to epistemology seems most important. The argument regarding inconsistencies aside, it is clear then that the category of Somatization has a different thinking base than that of Conversion. Both being considered as Somatoform Disorders creates many logical problems if the DSM-IV spirit of exclusory criteria is sustained. Somatization is regarded as a statistical, noncausal category, and conversion is a psychodynamic, nonstatistical category in which stress is inserted as an added factor without explanation of its confounding contribution. In the view of this author, stress is an important part of current thoughts about all somatoform processes. It plays a role in the statistical or cluster categories as well as in the conceptual-dynamic. The point being made here is that DSM-IV is unclear. It makes descriptions that overlap with categories in which mere labels or descriptions do not apply. Conversion is conceptual. Somatization is purely descriptive. Nowhere does it take physical complaints which do not apply to a known medical condition and distinguish these complaints in the various categories. A conversion pain may be a

181

R.D. Martin

somatization pain and vice versa. There is no conceptual system in which the two can be distinguished. The practitioner frequently notices the overlapping of diagnoses, but DSM-IV takes no account of the overlapping. If one drew a Venn diagram of physical complaints in different diagnostic categories and tried to separate diagnostic categories with the circles, the result would be an unsuccessful jumble of nearly congruent overlapping. Again, the argument made in the “DSM-IV Options Book” [12] regarding the substitute for DSMII’s Psychophysiological Disorders, which is called Autonomic Arousal Disorder, is particularly noteworthy. It says: “Reasons for not adopting this newly suggested diagnosis included the limited research supporting it, and the possible creation of a wastebasket diagnosis that is better covered by more specific Axis I or Axis III diagnoses” [12]. How can one help but smile after reading this thought. The forgoing argument and Table 1 showing the different logic used for the over-inclusive categories already in DSM-IV could be the reference for this quote. Psychophysiological Disorders or its DSM-IV equivalent is no more unresearched than anything now present in the nosology. Further, there is enormous research available regarding psychosomatic processes which is the parent concept of psychophysiology. One need merely investigate the psychosomatic-psychophysiological material extant, see its application, and work hard to use it. Nothing approaching this was in evidence. It would seem that a simple unrevealed bias was at work here. The question is what nosology would be acceptable. Frankly, even that which has been criticized above is acceptable if its pseudo-scientific, practical, hodge-podge basis is confessed. Its potpourri quality should clearly be placed on the table and its tentativeness and uncertainties should not be disguised.

Concepts A careful reading of DSM-IV and Current Concepts of Somatization by Kirmayer and Robbins [10] would

suggest an important theoretical dichotomy: “Whereas psychosomatic theory is concerned with disease causation, somatization focuses attention on the experience and expression of illness.” Somatoform (and Somatization) becomes descriptive, whereas psychosomatic becomes causal. The descriptive, anticausal nature of DSM-IV nosology obfuscates and misleads. It may be helpful here to use a classroom heuristic tool regarding psychophysical symptoms. Psychiatric residents in training are asked to imagine a line on which all diseases are placed. It is a finite line, and at one end, on the left, let us say, is the Greek letter psi (C) which stands for psychological or emotional factors. At the other end of the line is the English letter P for Physical factors, the operative biology of the human being. Along this line or spectrum of possibilities are placed all diseases known to humanity. The psi and P are viewed as variables which operate in various ways in the development of diseases (the two variables psi and P are, of course, super variables in which many intricate other variables are operating.) As diseases were assigned, large clumps of illnesses would occur at the P end and others at the psi end. There would also be insertions of diseases at various points, diagnosed with different proportions of the two variables. Migraines, for example, would have a large psychological component, but also are determined by physical diathesis. Pneumonias, on the other hand, would have less psi and more P. It might look something like Figure 1. From this line of reasoning, it would follow that any somatoform disorder would be an excerpt from this spectrum. The various elements of psychological or physical variables would be distinct or overlap depending on their operant functions within the disease under scrutiny. There are situations when a Venn Diagram of the two variables would be congruent, but in most instances there probably would be some discernable differences allowing the variable functions to be distinguished. Other Venn Diagrams of Diseases would show little or no overlapping. Therefore, the mind-body interaction is more completely viewed as part of the disease spectrum,

Table 1. Decision determinants in Somatoform diagnoses Somatization Statistical Psychodynamic Clinical presentation

182

Conversion

Hypochondriasis

Pain disorders

Dysmorphia

X X

X

X X

X

X

The Somatoform Conundrum

Figure 1. Spectrum of Somatoform Disorders (Psychosomatic Disorders).

and not a peculiar focus for those of special interests. DSM-IV attempts to deny such a perspective and tries to keep these factors as extremely distinct as possible. That is not tenable. The confusion that follows from such an effort of variable isolation is considerable. Research in Post Traumatic Stress Disorders and in the general area of stress makes it clear that people vary in their reactions. Some give no sign of physical illness and evidence a good deal of unhappiness, whereas others express physical symptoms such as headaches, fatigue, and other bodily complaints [13]. Reiser [14] and others have tried to explain this problem using psychoanalytic, psychodynamic, and cognitive schemes intermingled with a maze of physical variables. Engel [15] emphasized the biopsychosocial model in an attempt to encourage the integration of the psychological variable in working with general medical diseases. Weiner [16] aggressively attempted to show how biological and psychosocial variables might interact. He postulated a transducer system which integrates environmental stimuli with organism biology to create altered functioning and possible disease states. In his book, he argued forcibly that interaction is the key concept in much disease formation, and that it is no longer defensible to discuss the purely psychogenic development of illness. There is no longer a question that stress in the mammal can cause illness. Holmes’ and Rahe’s [17] work pioneered the idea that daily tribulations and miseries may increase the likeliness of illness. Henry et. al. [18] demonstrated that laboratory stress on rats, if the animal was genetically susceptible, could cause blood pressure elevation. Finally, among the many startling discoveries is the finding that rats can be trained to suppress their immune system by learning to associate saccharine with cyclophosphamide. This stands out as a dramatic illustration of how sensitive the mammal is to environmental learning (and it follows, of course, to environmental stress as a learning experience) [19]. This brief survey in the paragraph above is

merely meant to emphasize the alteration of thinking that causal psychosomatics has bestowed upon science. None of this was utilized in the nosology of DSM-IV. In fact, as already mentioned, there was a fear that “not enough was known . . .” [12] to include this material in a nomenclature of classification. There are more psychosomatic or mind-body interactional diagnoses possible than those that have been offered in DSM-IV. Making categories as if this is not the case is no longer acceptable. If it does persist, it will exclude research devoted to clarifying the intermixture of the biological, social, and physical spheres. It is proposed, then, that a classification that uses criteria in which the environment and the organism interact would prove more useful. In searching for the theme that might underlie such a diagnostic structure, stress appears to be a healthy candidate. The presence or absence of stress may be viewed as an environmental stimulus that can operate as a connecting thread through all the diagnoses to which the somatoform concept may be applied. The validity of the concept of stress was discussed at length in a former paper by this author [20] and will not be addressed here. Suffice it to say, the concept is viewed as valuable for a nosology. With some diagnoses, stress may be absent as a factor and others might be present. Each instance gives a new category for diagnostic consideration.

A Proposal It may well be that a completely consistent and inherently logical classification of psychosomatic disorders is not feasible at this time. However, it is entirely possible to go beyond merely labeling differences in a collection of metaphorical pebble textures and then denying that the labelings are arbitrary. One proposed format is made for inspection (Table 2). Though this attempt at classification is certainly preliminary and less than ideal, it does illustrate major differences over that in DSM-IV. The importance of environmental factors as a dif-

183

R.D. Martin

Table 2. An alternative classification of Somatoform Disorders I. External Disorders (environment working on the susceptible individual) Stress Induced Acute Disorders Somatization Disorders (due to intense, stressful social situations which are acute). Acute Insomnias (included here and under Sleep Disorders) Acute traumas (e.g. PTSD—physical or psychological) Direct stimuli causing distress (e.g., extreme noises, electric shock, dangers) Chronic Disorders Somatization Disorders (due to stressful social situations over time. Though controversial, this might include Chronic Fatigue Syndrome, Myofascial Syndrome, or Fibromyalgia. Chronic Insomnias (included here and under Sleep Disorders) Psychiatric Disorders that lead through body interaction to Physical Illness Depression (grief reactions with physical complaints) Depression (Deprivations States, Hospitalism, Love lost—all with physical complaints. Anxiety States (including panic) that have physical symptoms Hyperarousal States (e.g., terror in the face of real danger. Rage States) Psychophysiological States (altered activity of the autonomic nervous system and other physical body systems that give rise to symptoms. These would derive from known stresses.) Medical Diseases in which stress or other emotional issues are thought to be a contributing factor. (e.g., asthma, colitis, arthritis, irritable bowel, immune suppressant disorders.) These may be placed on Axis III, but also deserve their own place in the classifications. Pain Disorder not clearly due to Medical Disease II. Internal Disorders (no immediate environmental influence) Non-Stress Induced Somatization Disorders (Replacement Disorders: physical symptoms used in place of emotional symptoms) Conversion Disorders (historical issues evident) Hypochondriasis Self-Induced Illnesses (e.g., self inflicted wounds, suicide attempts with injuries, accidents with psychological factors involved, subjecting oneself to illness inducing situations) Symptom complaints due to unresolved personal issues (developmental or acquired in adulthood. These would not be conversion symptoms. They arise from conscious conflicts which might generate physical symptoms.) Depression with somatization Pain Disorders not due to a clear medical disease (due to internal individual issues) Misinterpretation Disorders Disorders in which minor physical symptoms are misinterpreted in a way that they suggest serious major physical disorders (e.g., dizziness suggesting brain tumor or middle ear problems; minor pain suggesting cancer or arthritis.) Dysmorphic Disorder

ferentiating element in all diagnoses is immediately apparent. As a corollary, the developmental process is recognized as a significant factor that helps to create somatization. The Perley and Guze criteria of assorted symptoms required for somatization is eliminated. Somatization is a body expression of emotional factors. This schema therefore includes the patient illness-behavior factor referred to previously in other author citations. Somatization is repeated in I and II of this classification and is dependent on the stimulation or lack of stimulation by environmental factors, not on numbers of symptoms. Its meaning as a diagnosis with multiple

184

symptoms with or without the Perley-Guze criteria is retained. Its new position takes into account environmental as well as internal issues that might contribute to its formation. Medical Diseases are recognized as potentially genuine psychosomatic disorders and are not placed apart. There are those medical diseases that will deserve to be included under Somatoform Disorders, and others will be viewed as purely physical. This considers an important and neglected part of medicine and includes medical diseases under the psychiatric rubric where it is sometimes felt they belong. This widely held view derived from

The Somatoform Conundrum

clinical work has long been neglected nosologically, and, in some circles, is a fervently denied truth. The gastroenterologist on many occasion has great difficulty considering Irritable Bowel Syndrome as a mind-body interaction issue. The cardiologist refuses to accept that family stress or even depression may trigger arrhythmias. This classification sends a signal that the disease spectrum includes both the somatic and the psychological variables. There is no reason why the somatization, so common in Major Depression, should be subsumed under that eponym. Even if Major Depression deserves its own special place in the overall classification of mental disorders. Recognizing somatization as an offspring of depression has value that is not now regularly appreciated. Placing it here as a separate category would highlight its importance to nonpsychiatrist physicians who work in primary care. It would also eliminate the confusion that arises when depression and somatization are the patient’s presentation. Since somatization often presents as a symptom of endogenous depression arising from presumed genetic factors, it may be a part of depression without a stress stimulus. Autonomic arousal occurs with anxiety and panic. Though the limbic system and papez circuit is recognized as an important part of the emotional brain, there is a reluctance to include autonomic and emotional elements in the diagnosis. As the brain structure is clarified further, the importance of these aspects of behavior can no longer be ignored. The amygdala is now viewed as an important, if not elemental, contributor to emotional and, then, autonomic activity [21]. We must make sure this increasingly important aspect of witnessed clinical conditions is not excluded. Psychophysiological States must therefore no longer be an abandoned diagnostic category. Pain Disorders not due to a medical disease can arise from either acute or chronic stress, derived from external stimuli or internal depressive states. For this reason it is repeated in Table 2. This repetition emphasizes the multiple possibilities for pain, avoids the “psychogenic” confusion associated with pain in DSM-III and DSM-IIIR, but keeps it as a separate entity. Pain certainly might be part of a muscular tension syndrome. There is no reason to exclude Pain Disorder when making other medical or somatoform diagnoses. The multiple uses allows a full range of intellectual consideration and research when the topic is treated in this manner. Pain Disorder is considered part of a Medical Disease, but is also an aspect of a somatization process,

for example. There may be, in addition, exaggerated pain complaints within the context of known pathology. The use of Medical Disease under StressInduced Disorders should cover the matter. The drugseeking patient with a genuine pain problem can be included here and also be placed under SubstanceRelated Disorders and Axis III medical disorders. Section II of Table 2 deals with problems for which external stress plays a minor or no role in the formation of the presenting clinical condition. The consideration here is largely psychodynamic. That is, long-standing developmental issues cause internal adaptive problems that arise at any point in life after the developmental period has long past. Another example would be Depression with Somatization in which generic Major Depression may exhibit somatic symptoms with no stimulation from either the environment or from provocative stress issues. Conversion Disorder should be included here since developmental and psychodynamic issues are prominent. Conversion is largely a result of historical conflicts emerging later in development. Though stress may precipitate the syndrome, its full configuration is not viewed as environmentally related. It could be argued that some conversion arises out of life experiences and it is only the form of the symptoms that justifies the diagnosis. This argument is in contradistinction to the psychodynamic view of conversion. The entire issue will be addressed in a future paper. For now, the psychodynamic model is held as primary in this diagnosis and it is placed in the nonstress category. The Misinterpretation Disorders in Section II is a new category which requires further discussion. Its value is to include those unexplained singular symptoms that are frequently encountered in clinics, such as unexplained vertigo, minor muscular pain, and gastric upset.

Conclusion The major problem being confronted in this paper is how to deal with the overlapping of psychosomatic issues with other psychiatric and medical problems in the diagnostic nomenclature. A secondary problem is the confusion that arises from using exclusory criteria, eliminating mind-body interaction causal factors, and creating a diagnostic structure that suggests absolute delimitation of categories. Though it is comforting to imagine that Somatization and Conversion and Pain can be distinctly and clearly separated at all times from agreed upon medical diseases, in practice this is not the case. The

185

R.D. Martin

psychological variable moves, willy nilly, into all aspects of medicine (see Fig. 1). Full recognition of this fact has been delayed and resisted for too long. The autonomic, endocrine, and immune systems are now acknowledged to be influenced by the brain which, in turn, is affected by the environment. Mazure and Druss [22] make it clear that matters are clearly more complicated than the simple “fight or flight” of Cannon, and the intricacy of brain-body interactions must be included somehow into the diagnostic nomenclature. It is widely agreed that stress is a major factor in illness formation and it, too, cannot be excluded. The classification proposed here tries to take these two factors into account. This proposal does not presume to answer the questions created by the conundrum. It is merely a suggestion for discussion. It is hoped that these alternatives to DSM-IV will be seen as a point of healthy departure. The battle of causal vs descriptive nosology must be put aside. Avoiding causal factors because of fear of prematurity seems as bad an error as excessive hubris regarding the extant data of causality. There is no reason why the two value-laden approaches of descriptive and causal cannot be integrated. The extreme position of DSM-IV warrants significant modification. There are those who would declare that “. . . there are no psychosomatic disorders, only situations where psychological (and social factors) play an important [predominant?] role in causation, exacerbations or maintenance of disease and distress” (a personal communication). This position seems to introduce vague diagnostic categories with no foundation in physiological science. The question is not whether there is “only” psychosomatic. Rather, it is whether “psychosomatic” can continue to be excluded from our conceptual framework in the diagnostic nomenclature. Though the framework offered here cannot possibly be the last word, it is hoped that it could be a provocative source for further discussion, if not an improvement over what is widely used at this time—DSM-IV.

References 1. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders DSM-IV, 4th ed. Washington, DC, American Psychiatric Association, 1994, pp xvi, 445–461 2. Websters Ninth New Collegiate Dictionary, Frederick C. Mish, (ed). Norwalk, Connecticut, The Eastern Press, 1990, p 286 3. Kirmayer LJ, Robbins JM: 1991. Functional somatic syndromes. In Kirmayer LJ, Robbins JM (eds). Current Concepts of Somatization, Research and Clinical

186

4. 5.

6. 7. 8. 9. 10.

11. 12.

13.

14. 15. 16. 17. 18. 19. 20. 21. 22.

Perspectives. Washington, DC: American Psychiatric Press, p 91 Kirmayer LJ: Improvisation and authority in illness meaning. Culture Med Psychiatry 18(2):183–214 Bass CM, Murphy MR: 1990. Somatization disorder: critique of the concept and suggestions for further research. In Bass CM, Cawley RH (eds), Somatization: Physical Symptoms and Psychological Illness, Oxford, Blackwell Scientific Publications, pp 301–332 Pilowsky I: The concept of abnormal illness behavior. Psychosomatics 31(2):207–213, 1990 Kirmayer LJ, Taillefer S: 1997. Somatoform Disorders. In Turner S, Hersen M (eds), Adult Psychopathology, 3rd ed. New York. John Wiley and Sons, pp 333–383 Kimball CP: 1978. Diagnosing Psychosomatic Situations. In Wolman BB, (ed), Clinical Diagnosis of Mental Disorders. New York, Plenum Press pp 677–708 Perley MJ, Guze SB: Hysteria—the stability and usefulness of clinical criteria. N Engl J Med 266:421–426, 1962 Kirmayer LJ, Robbins JM: 1991. Introduction: concepts of somatization. In Kirmayer LJ, Robbins JM (eds), Current Concepts of Somatization, Research and Clinical Perspectives, Washington, DC: American Psychiatric Press, p 2 Freud S, Breuer J: Studies on Hysteria. New York, Avon Books, p 249, 1966 American Psychiatric Association Task Force on DSMIV. 1991. Chapter I: Somatoform Disorders, In DSM-IV Options Book: Work in Progress 9/1/91. Washington, DC, American Psychiatric Association, pp I:5–I:11 Bremner JD, Southwick SM, Charney DS: Etiological factors in the development of posttraumatic stress disorder. In Does Stress Cause Psychiatric Illness? EM Mazure, Progress in Psychiatry Series, No. 46. Washington, DC, American Psychiatric Press, Inc, 1995 Reiser MF: Mind, Brain, Body, Toward a Convergence of Psychoanalysis and Neurobiology. New York, Basic Books, 1984 Engel GL: The need for a new medical model: a challenge to biomedicine. Science 196:129–136, 1977 Weiner H: Psychobiology and Human Disease. Elsevier Scientific, New York, pp 585–624, 1977 Holmes TH, Rahe RH: The social readjustment scale. J Psychosom Res 11:213, 1967 Henry JP, Meehan JP, Stephens PM: The use of psychosocial stimuli to induce prolonged systolic hypertension in mice. Psychosom Med 29:408, 1967 Ader R, Cohen N: Conditioned immunopharmacologic responses. In Ader R (ed), Psychoneuroimmunology. New York, Academic Press, p 281 Martin RD: A critical review of the concept of stress in psychosomatic medicine. Perspect Biol Med 27(3): 443–464, 1984 LeDoux J: The Emotional Brain. The Mysterious Underpinnings of Life. New York, Simon and Schuster, 1996. p 17 Mazure CM, Druss BG: A historical perspective on stress and psychiatric illness. In Mazure CM (ed), Does Stress Cause Psychiatric Illness? Progress in Psychiatry Series, No. 46. Washington, DC, American Psychiatric Press, Inc.