| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
SPECIAL ARTICLE |
From the Department of Psychiatry, Pennsylvania Hospital, and University of Pennsylvania, Philadelphia, PA.
Address reprint requests to: Donald Oken, MD, Department of Psychiatry, Pennsylvania Hospital, 800 Spruce St., Philadelphia PA 19107-6192.
ABSTRACT
Current medical diagnosis reflects the prevailing biomedical model of disease. A need exists for a new system of diagnosis that, instead, is based on the psychosomatic model. This article presents an analysis of the underlying framework of the multiaxial system developed in recent years for diagnosis in Psychiatry that indicates its relevance to the psychosomatic model. It goes on to describe a new multiaxial system of diagnosis derived from that analysis that allows diagnosis to be stated as a process of adaptation in the environment, which includes biological, psychological, and social factors. The practical application of this system to the broad range of medical illnesses is explained and illustrated. This multiaxial approach represents a first step toward, and a stimulus for, the development of a better diagnostic system that can provide one basis for the crucial transformation of medical care to reflect the psychosomatic model of disease.
Key Words: diagnosis • psychosomatic • disease model • multiaxial.
Abbreviations: Consultation/liaison = C/L; DSM-III = Diagnostic andStatistical Manual of Mental Disorders, third edition; DSM-III-R = Diagnostic and Statistical Manual of MentalDisorders, third edition, revised; DSM-IV =Diagnostic and Statistical Manual of Mental Disorders,fourth edition; ICD = International Classification of Diseases; MAS = multiaxial assessment system.
Diagnosis is a bedrock of modern medicine, and with apparent good reason. A diagnosis is a concise denotation meant to provide a practical guide to treatment, course, and prognosis, and a basis for prevention and research. Physicians consider diagnostic categories, for the most part, as firm, reliable entities. They may acknowledge that some have imperfections but tend to view these as close approximations limited by gaps in current knowledge, with the presumption that they will be perfected as science advances. There is limited awareness that many current diagnostic categories, particularly those for chronic diseases, do not provide a clear basis for treatment and are "prognostically worthless," as Feinstein (1) has noted. Indeed, there are major defects in the overall system in which our nosology is compiled. The ICD consists of a hodge-podge of etiological, morphologic, pathogenetic, pathophysiological, symptomatic, and syndromic rubrics that reflect an accretion of shifting historical notions about the nature of disease (1, 2).
Even when diagnoses reflect the cutting edge of knowledge, however, they remain inadequate because they are based on an extant model of disease that is fundamentally flawed, the "biomedical model" (3). As Engel (3) has noted, that model embodies a reductionistic view in which disease is "fully accounted for by deviations from the norm of measurable biological (somatic) variables." Furthermore, it relies largely on a linear, sequential unicausality, exemplified in the "one gene, one enzyme, one disease" simplism. In this traditional view, disease represents a state of abnormality discontinuous with health imposed on the organism: Disease is something one "gets." Within this model, it is diseases that are treated, not patients, and diagnoses are specific, nomothetic "entities" with fixed characteristics independent of the patient. The logical results of such typologies are stereotypes that ignore readily observable differences among patients with the same disease, so that "the better a clinician knows a patient, the harder it is to make a diag-nosis" (4).
Psychosomatic medicine provides a very different model of disease. Although its most evident feature is that it is "biopsychosocial," to use Engel’s (3) term, it has additional crucial elements that make "psychosomatic model" a more apt name.1 With the inclusion of these elements, the model states that health and disease fall on a continuum; each person is more or less healthy and more or less diseased in a variety of ways. Both represent a complex of adaptive processes of the human organism that involve biological, psychological, and social components, all in transactional (7) relationships with the environment—physical, social, and cultural. Each "disease" in each person reflects an ideographic process comprised of these complex transactions among internal and external variables that only partly reflects the nomothetic stereotype.
The problem with such an overarching model is its very generality. What is required are operational concepts derived from the model that are applicable to the practical tasks of medicine. Until these are developed, the model remains of academic interest with limited acceptance by clinicians. This need is particularly acute with regard to diagnosis as the linchpin of clinical practice. The development of a functional diagnostic system based on the psychosomatic model of disease would not only enhance the credibility of the model but would have far-reaching consequences if incorporated into the daily thinking of physicians. It is widely known that a conceptual model affects the choice of terminology. But, conversely, language structure also affects how one thinks (8). The persisting use of standard diagnostic nosology reinforces the traditional biomedical model, whereas its substitution by a diagnostic system based on the psychosomatic model can reorient physicians’ thinking toward the latter.
MULTIAXIAL DIAGNOSIS IN THE DSM
In 1980, American psychiatry adopted DSM-III as a basis for diagnosis (9). Much of the attention generated by this document focused on its new taxonomy, nosological entities, and especially its criterion sets. Less noticed by medicine generally, but the most innovative feature of this new approach, was its MAS. Stemming from a proposal by Essen-Moller and Wolfahrt (10), similar systems had been used already in other areas of medicine (1). Many variations of a multiaxial approach were suggested over subsequent years, tailored to the needs of specific age groups, clinical groups, and psychiatric ideologies (11). Various modifications of the axes adopted were suggested, and some were incorporated into two subsequent revisions, DSM-III-R and DSM-IV (12, 13).
The expansion from single-item diagnosis to several axes was intended to provide additional "domains of information assumed to be of high clinical value" (11) for psychiatric practice. The particular version adopted was designed to provide a method for including biological, social, and psychological factors related to diagnosis, reflecting the official espousal of Engel’s biopsychosocial model by the American Psychiatric Association. In fact, part of the reason for this inclusion was to appease each of the major ideological subgroups in American psychiatry who, for the most part, continue their separate reductionistic positions and pay only lip service to the model. But even among those psychiatrists who embrace the official model, the predominant understanding is that it permits the inclusion of the three elements in an additive manner, as a more complete, "comprehensive" approach, a far cry from their transactional integration or the inclusion of other key elements of the genuine psychosomatic model. With rare exceptions (14–16), what has been written about the MAS reflects this simplistic, additive understanding.
Nevertheless, deeper examination of the psychiatric MAS reveals that its elements do provide a method for communicating diagnosis as an adaptive process in relation to the environment. Hence, it offers a first step toward the development of a meaningful, operational system of diagnosis based on the genuine psychosomatic model. To discern this requires an analysis of the conceptual framework underlying each of the DSM2 axes that looks beneath the specifics of their concrete definitions provided for psychiatry.
Axis I is straightforward. Formally, it represents "clinical syndromes" or other psychiatric "conditions ... that are a focus of attention or treatment." (9). More specifically, it represents the relevant choice from the disorders or conditions included in the official taxonomy of the DSM that represents the primary focus of clinical attention. This is the traditional "diagnosis" that brings the patient to psychiatric (or other mental health) attention. Up to a point in time before that happened, the patient had been functioning more or less adequately. Traditionally, one would say that the patient had been well and became sick. From a psychosomatic perspective, this can be restated to indicate that the patient had been in a condition of stable "adaptedness" (17) that was disrupted, resulting in more or less acute "maladaptation" of sufficient significance to require psychiatric (etc.) attention. The specific entry on Axis I simply represents the particular variety (diagnosis) of the maladaptation that has ensued.
A bit more effort is required to discern the underlying nature of Axis II. Its explicit definition in DSM-IV indicates that it includes "Personality Disorders and Mental Retardation," although it also allows the inclusion of "prominent maladaptive personality features and defense mechanisms." (DSM-III and DSM-III-R contained different variations of other developmental disabilities to be included, but these are of no immediate relevance.) The word "maladaptive" points to what is involved here. The key aspect of all the disorders included is that they are enduring, relatively immutable conditions that represent a baseline substrate of impaired adaptedness, of deficiencies or distortions that limit the capacity to adapt successfully to the demands of life. Kendell’s (18) term for these, "handicaps," is apt. They impose a Procrustean bed beyond which the affected individual cannot adapt. Although their influence may be unobtrusive over a long period of time, clinical significance arises when new or changed environmental demands require specific adaptive responses that are precluded by the handicap.
Axis II conditions thus heighten vulnerability to the occurrence of the acute maladaptations of Axis I. In some instances, this is general: There is broad vulnerability to a variety of adaptive breakdowns. In others, the relationship between the two is more specific: Certain Axis II disorders are linked to particular Axis I diagnoses. One example is the association of Obsessive-Compulsive Personality Disorder with Melancholic Depression (19). Although the basic data that document such relationships are epidemiological, psychoanalysis provides a conceptual base for understanding many of the links between specific personality distortions and corresponding clinical disorders.
Axis III is used to indicate any "current physical disorder or condition that is potentially relevant to the understanding or management" of the individual who has developed the Axis I diagnosis. The universe of such factors may seem to be restricted by its limitation to diagnoses found in the nonpsychiatric sections of the ICD. The latter, however, includes a host of conditions of a wide variety, such as headache, pruritus, constipation, and pregnancy. Although this axis may be used for conditions relevant only to treatment, its major thrust is clearly etiological. Its essential purpose is to reflect those biological stressors relevant to the breakdown in adaptation denoted in Axis I.
Axis IV parallels its predecessor. It serves to report "psychosocial and environmental problems that may affect the diagnosis, treatment, and prognosis of mental disorders," that is, of the Axis I disorder. Included are the range of interpersonal, social, environmental, educational, occupational, economic, and legal problems. "Prognosis and treatment" take us astray, and a further explication that the psychosocial problems may also be those that develop as a consequence of psychopathology does so even more. It is evident that the central quality of this axis is simply the counterpoint to Axis III, shifting from the biological to the psychosocial sphere. It represents another facet, the psychosocial, of the relevant stressors that have upset the prior adaptive balance. Perhaps there is practical value in separating biological from psychosocial stressors to ensure that both are considered, but conceptually the two are merely different aspects of the forces acting to undermine adaptedness.
The final Axis, V, is designated to provide a "Global Assessment of Functioning." The exact scale used is a petty detail. What is clear is that this represents an effort to quantify the effectiveness of adaptation at given relevant points in time. The time chosen is of some importance. When DSM-III was created, it was decided that the assessment was to be made of "the highest level of adaptive functioning (for at least a few months) during the past year." The choice of a year was arbitrary and somewhat controversial, but the intent is clear. This was to be a measure of the preexisting level of effectiveness of adaptedness, that is, that level present before the acute maladaptation of the Axis I disorder. It was to represent the baseline of function, which included the effects of enduring (Axis II) disabilities, and offered a realistic goal for treatment outcome. With DSM-III-R, the decision was made to include a second occasion, the time period of the acute illness (and this became the sole focus of DSM-IV). What was thus also quantified was the extent of the acute maladaptation. From our standpoint, the dual inclusion is ideal, because it provides a quantification of both the prior level of adaptedness as well as that reached during the impact of the illness.
MEDICAL MULTIAXIAL DIAGNOSIS
By intent, the preceding analysis has been abstract, dealing with the MAS in general rather than specific applications. Plentiful examples of applications of the reformulated MAS to psychiatric problems could be given, but this would be of little interest to those other than psychiatrists. Even with the analysis provided, psychiatric diagnosis is a sideshow for the rest of medicine and of only limited interest to those working in psychosomatic medicine. Our concern is not with psychiatric diagnosis but with a diagnostic system applicable to all that reflects the psychosomatic model of disease. What is important is that the dissection of the underlying structure of each of the axes can serve as the basis for a multiaxial configuration consonant with psychosomatic perspective. Before that can be demonstrated, it is necessary to put the newly configured MAS together.
What has been developed is a new multiaxial model in which:
With this reformulated general MAS in hand, we can move beyond the abstract to a real clinical application. Consider the following example. A 33-year-old man with known insulin-dependent diabetes mellitus of 15 years’ duration is hospitalized with fever and an obviously infected toe. His pulse and respiratory rates are elevated, and respiration is deep and labored. Other physical findings include weakness; flushed, hot, dry skin with dry mucous membranes; red streaking of the foot; and stupor. Urinalysis reveals 4+ sugar and ketones. Blood measures include a glucose concentration of 748 mg/dl; low Na, Cl, and CO2; elevated K; and a pH of 7.21. Blood and toe cultures later reveal coagulase-positive Staphylococcus aureus. He is treated with methicillin, insulin, fluid, and electrolytes, and he recovers over the next 3 days. A psychiatric consultation is then requested because this is the fourth similar admission in 6 months after a long history of relatively stable diabetic control. The C/L psychiatrist notes that the earlier delirium has cleared but that the patient looks and acts moderately depressed. In the interview, the patient reports intermittent sadness and loneliness since the death of his wife 7 months earlier. Childless and living alone, he admits that since her death, he "cheated" on his diet, inconsistently tested his blood glucose, and sometimes "couldn’t be bothered" to take his insulin. Although he had decreased his social activities, he had continued to work regularly in his long-standing job as a paralegal, and he found considerable relief and satisfaction in the work as well as in the company of coworkers. After discussion, he declined to take antidepressants or to accept a referral to a bereavement support group but understood the need to increase his use of social support and, especially, to take better care of himself. He was discharged to be followed by his internist.
In accordance with usual medical practice, this patient was discharged with a principal diagnosis of diabetic ketoacidosis, and additional diagnoses of infection of the toe with S. aureus and Insulin-Dependent (Type I) Diabetes Mellitus. No mention was made of the psychiatrist’s diagnosis (Adjustment Disorder with Depressed Mood). Had this been done, it would have simply been listed as an additional diagnosis.
Consider, instead, a diagnostic formulation for this patient using the reformulated multiaxial system:
Admittedly, this formally organized series of statements has a structured, somewhat stilted quality. But we are always faced with the need to provide diagnoses in codified format. Moreover, the multiaxial formulation not only portrays the situation accurately and reasonably succinctly while maintaining the broad picture, but also does so as an adaptational process involving biological, psychological, and social variables.
DISCUSSION
The traditional internist might object to this example as supposedly aberrant, with a contrived inclusion of an unusual "psychiatric complication." But this merely reflects entrapment within the biomedical model. In fact, cases of this type are typical occurrences on every hospital medical service. The experience of liaison psychiatrists who see medical and surgical patients free of psychopathology is that psychological and/or social factors play significant roles in the illnesses of most patients. That is precisely why there is need for a psychosomatic diagnostic system. One very common behavioral pathway through which these factors operate is noncompliance, a factor that has been documented as a major basis for the decompensations that require hospitalization or other acute medical care (20).
Observations over years of consultation/liaison psychiatry practice reveal that the proposed system is applicable to most medical situations. For example, it has no trouble dealing with such situations as arthritis where structural changes may arise after one or more acute attacks. Here, the solution is to include the structural abnormality, having once appeared, as an Axis I diagnosis (eg, arthritic deformities of the hand), reduce the level of adaptive functioning appropriately, and use Axis I for acute attacks.
Admittedly, a relatively simple, straightforward clinical situation was chosen to make the exposition clear. The MAS presented does have greater problems dealing with more complex medical situations. Some of the limitations that arise are not intrinsic to the system, however, but are imposed by shortcomings in the current medical nosology. For example, although our patient did not (yet) have these factors present, Type I Diabetes is commonly associated with hypertension, obesity, and/or renal insufficiency, and the relationships among these factors are complex transactions that cannot be expressed in a mere sequential listing of these conditions without doing violence to the reality. One solution for this would be to add such whole complexes as single entries in the ICD so that they can be reported as such in our Axis II. A nosological mechanism of that type would provide a better fit with the transactional character of psychosomatic conceptualization, although it pertains only to biological factors.
Numerous additional clinical examples could be provided to document the applicability of the model to a wide variety of medical and surgical illnesses. But such detail would serve no point. It would be foolish to suggest that the model presented here is a perfect one with universal applicability to medicine. At the least, modification will be required. This MAS is not the solution to our need for a diagnostic system consonant with the psychosomatic model of disease. Rather, it is a first step toward such, one that seems to have real potential with further development.
In the current era, when economics has become the major force shaping medical practice, "efficiency"-driven fragmentation of care based on ever-briefer contact with patients has become the norm. This climate makes it all the more important to have a diagnostic system that promotes eliciting the psychological and social data along with the biological, which converts the facade of "doctoring" to genuine medical care and which provides a way of recording that data so that it can be transmitted to the next healthcare provider to whom the patient is referred.
Should the present MAS prove an unsatisfactory lead, that will not be important. What is important is that we move to develop a diagnostic system for medicine as a whole that eschews static pseudoentities and instead provides a dynamic statement of the adaptive processes taking place in sick people that incorporates biological, psychological, and social variables: in short, a psychosomatic diagnostic system. It seems reasonable to consider that the MAS described here, which is based on the underlying properties of that recently developed for psychiatry, may provide an initial basis for such a system, as I have tried to demonstrate. At the least, this effort may serve as an impetus for the development of a better alternative. If psychosomatic medicine is to continue to serve as a vital force within medicine, it must provide practical tools for dealing with its clinical tasks, of which a system of diagnosis is central. We should move ahead to do so.
NOTES
As I have previously indicated (5), this term also has the advantage of historical grounding that takes account of the rich, relevant extant research literature, the bulk of which has been published in this journal. In no way, however, is this to suggest that Engel’s model is limited to the mere inclusion of the three systems (3, 6); in fact, in its full exposition, it is the psychosomatic model, differently labeled. 
Because the purpose of this analysis is to expose its underlying structure, focus is placed on the commonalities of the three existing versions of the DSM, although some reference will be made to differences to clarify their relationship to the analysis.
Received for publication April 22, 1999.
Revision received August 19, 1999.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  D. Oken Evolution of Psychosomatic Diagnosis in DSM Psychosom Med, November 1, 2007; 69(9): 830 - 831. [Full Text] [PDF]
|
|
|
|
 |
|
 D. H. Novack, O. Cameron, E. Epel, R. Ader, S. R. Waldstein, S. Levenstein, M. H. Antoni, and A. R. Wainer Psychosomatic Medicine: The Scientific Foundation of the Biopsychosocial Model Acad Psychiatry, October 1, 2007; 31(5): 388 - 401. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Picardi, P. Porcelli, P. Pasquini, G. Fassone, E. Mazzotti, I. Lega, L. Ramieri, E. Sagoni, D. Abeni, A. Tiago, et al. Integration of multiple criteria for psychosomatic assessment of dermatological patients. Psychosomatics, March 1, 2006; 47(2): 122 - 128. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. M. Galeazzi, S. Ferrari, A. Mackinnon, and M. Rigatelli Interrater Reliability, Prevalence, and Relation to ICD-10 Diagnoses of the Diagnostic Criteria for Psychosomatic Research in Consultation-Liaison Psychiatry Patients Psychosomatics, October 1, 2004; 45(5): 386 - 393. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. H. Novack Realizing Engel's Vision: Psychosomatic Medicine and the Education of Physician-Healers Psychosom Med, November 1, 2003; 65(6): 925 - 930. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Afari and D. Buchwald Chronic Fatigue Syndrome: A Review Am J Psychiatry, February 1, 2003; 160(2): 221 - 236. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
