Since the publication in 1980 by the American Psychiatric Association of its most recent Diagnostic and Statistical Manual (DSM-III), many psychoanalysts have experienced a mixture of bewilderment and pessimism. Is DSM-III truly an idea whose time had come? A set of researchable criteria? Or will psychodynamic concepts be swept away on a regressive tide of Kraepelinian criteria: descriptions proudly proclaimed to be adynamic, even though some of the concepts used actually refer to dynamically unconscious motives.
As reported in the last issue of this Newsletter, the Association was recently invited by the APA to evaluate DSM-III from the psychoanalytic perspective. A committee was appointed which submitted its report at the 1983 fall meeting. Its members were myself, Chairman, Drs. Kay Blacker (San Francisco), James Egan, (D.C.), Allen Frances (New York), Douglas Logue (Baltimore), Jon Meyer (Milwaukee), and Allan Rosenblatt (San Diego).
A revised version of DSM-III (DSM-IIIR) is targeted for publication in 1985. DSM-IV is scheduled to appear in 1992. The trustees of the American Psychiatric Association are attempting to broaden the base of participation by all segments of American psychiatry in the writing of these documents. The first step in this process was the formation in 1981, under the APA's Council on Research, of a Committee to Evaluate DSM-III chaired by Dr. Gary Tischler.
Dr. Tischler submitted questions concerning DSM-III's usefulness to many professional mental health organizations throughout the country. Our committee responded to all questions. As might be anticipated, however, our major interest was in the areas of diagnosis and treatment, where we made clear our case for the existence and importance of psychodynamically unconscious motives and conflicts.
We pointed out that DSM-III, in its enthusiasm for careful description, introduces an artificial separation between descriptive and psychodynamic data. DSM-III generally assumes that psychodynamics refers primarily to etiologic or metapsychologic concepts. Our position was that clinical observations concerning the dynamic unconscious are often no more inferential than observations considered descriptive in DSM-III. We demonstrated instances in DSM-III where such concepts were already present, despite the authors efforts to exclude them.
The concept of “splitting” is an example of such a useful diagnostic concept. The DSM-III criteria for its presence uses psychodynamic data which can be specified at the levels of clinical observation and clinical inference. Another example of psychodynamic data that are important in diagnosis and treatment is defined and described operationally in the psychoanalytic section of the APA's own Peer Review Manual under the heading, “Quality of Interaction.”
In the clinical interview, the patient unknowingly attempts to mold the relationship to fit the requirements of his or her disorder for certain qualities of relationship. These unconscious efforts by the patient evoke feelings in the clinician which motivate behaving, thinking, or feeling in certain ways. The clinician will have learned to detect
these efforts of the patient, to “read” his or her own evoked feelings and tendencies to act, and then to use them diagnostically as sensitive indicators of various forms of psychopathology and their underlying patterns of unconscious conflict and conflict resolution. We emphasized that these data are descriptions of the clinician's emotional responses, which are as valid as are the descriptions of the patient's emotional responses as they are described to the clinician.
The committee was mindful of the need not to grind any axes (pun intended) in considering recommendations for change in the present multi-axial approach that DSM-III uses. Currently, Axis I includes clinical syndromes and Axis II includes personality disorders. Even though the guidelines in DSM-III state that Axis II can be used to indicate personality traits when no personality disorder exists, the examples given (compulsive and paranoid traits) make it clear that pathological phenomena are being described.
Hence the committee recommended combining all pathological phenomena now in Axes I and II into a new Axis I, and replacing the current Axis II with a new Axis II. The latter would then include assessments of types of character structure, object relations, ego functions, and defensive organization. When possible, our report used examples from the Psychoanalytic Peer Review Manual (written by a committee of the Association) because it has an extensive glossary and was written for general psychiatrists as well as psychoanalysts. For example, operational language is provided in the glossary to describe rigid or fragile-brittle defensive organizations.
In other instances, we tried to use conventional psychoanalytic wisdom to make psychodynamic concepts operational. For example, in the assessment of object relations, we proposed that at a “low” level, relationships are fragmented, temporary, and shifting; people are used to satisfy the patient's own needs or merely to bolster exalted or depreciated self-concepts. At an “intermediate” level, relationships may be committed and devoted, but rigid: there is little room for other than routinized discourse or sexual exchange. New ideas or new objects are not permitted in a free and easy fashion. At a “high” level, relationships are committed, steady, loving, and devoted; there is the flexibility that allows the introduction of new friends, new ideas, and new methods of exchange between the partners.
For those analysts still concerned with the place of the concept of “neurosis” in subsequent DSMs, the report considered neurosis in terms of the descriptive symptom complex proposed for Axis I; our proposed Axis II considers neurosis or neurotic disorder in terms of the developmental level of organization. Thus, a patient may have a set of obsessive-compulsive symptoms on Axis I, but it will make a great deal of difference diagnostically and therapeutically how these symptoms occur on Axis II. For example, similar symptoms may appear in a neurotic individual with a relatively high level of development of object relations and a relatively stable defense organization, in one with a borderline personality organization, or in a psychotic individual with severe difficulty in reality testing. The implication is that the signs and symptoms on Axis I may be generated either primarily by intrapsychic conflict and/or by developmental arrest, including some by one and some by the other in the same patient.
To develop Axis II in the way we propose—for the assessment of personality organization—it will be necessary to operationalize and simplify psychodynamic concepts. This will require the support of the APA committee developing DSM III-R and DSM-IV. Field trials will be necessary to test the usefulness of these operationalized psychodynamic concepts.
We hope that Dr. Tischler's committee might wish to recommend to the Research Council of the APA, to which it reports, that the APA would be wise to invest some of the profits from the sale of DSM-III to support such research, in anticipation of the increased value of DSM-IV to the profession. Such an approach
could be comprehensive, organized, and encyclopedic. It could, in turn, perhaps be followed by a pilot field trial, in which we would attempt to teach relatively unskilled interviewers such as third year medical students how to elicit this additional data.
We recognize that the ability to tolerate and therapeutically use strong evoked emotional reactions requires specific training. However, unskilled clinical interviewers, even outside the mental health field (such as people in the humanities) are capable of assessing this quality of interaction.
There was an important minority opinion on the committee which disagreed with the suggestion that Axes I and II be merged because of concern that the clinical research interest in the relationship between disorder and personality stimulated by the current multi-axial system would end. They would advocate an axis for defenses or an axis for type of personality organization, but believe that we should not try to integrate the former into the latter, though they accept the concept that defenses are part of the personality organization.
Their conviction is based on the fact that unpublished research indicates a current inability to obtain agreement among clinicians to define clear boundaries among various personality organization categories. They point out that there is not yet a decision-making tree (algorithm) available for such distinctions. By contrast, the results of research on defenses, while far from conclusive, are in the literature and are farther along. These committee members prefer to keep personality disorders, in whatever form, on Axis II and to devise a new axis—Axis VI—for defenses.
Regarding the usefulness of DSM-III for making treatment decisions, the committee's position was that psychoanalysis and psychodynamic approaches in psychiatry contribute to the treatment planning process in three important ways.
First, they provide the basis for a method of generating and collecting data—the psychodynamic interview. This is an interactive process between patient and psychiatrist that provides access not only to the facts that the psychiatrist is able to elicit, but also to the implicit, often unconscious, patterns that shape the patient's inner experience and outer behavior. The psychodynamic interview can be thought of as an experiment, with the interviewer actively intervening and testing hypotheses, creating possibilities, and observing how the patient (and the interviewer) responds to them.
This is in contrast to less psychodynamic approaches in psychiatric data gathering, in which the information is presumed to exist only in the patient, and the interviewer's task is to carefully collect and organize it, but not to utilize the interview situation itself as the setting in which new data can be created.
Second, psychodynamic approaches point to important areas of data—the patient's subjective world, inner representations of self and others, patterns of human relationships, characteristic modes of defense and coping, and early life experience, and the persistence of that experience in the patient's contemporary adaptation. These data are all central to a psychodynamic assessment, although they have only varying relevance to a descriptive Psychopathologic diagnosis. They are vital both in describing psychiatric treatments and in predicting treatment responses. This is particularly true for treatments that involve psychodynamically informed interventions, but also holds in broader terms for any treatment that requires even the patient's minimum participation in a relationship with the therapist.
Finally, psychoanalysis and psychodynamic understanding have led to the development of a wide array of therapeutic methods, including some of the most widely accepted treatments in contemporary psychiatry. These treatment approaches cannot be prescibed rationally without a psychodynamic assessment and evaluation process, in addition to a descriptive Psychopathologic diagnosis. The two types of data and the two types of assessment are fully compatible; in fact, they tend to strengthen and enrich each other. Together they provide the basis of a scientific psychiatric treatment approach.
It is our hope that this approach to treatment will be enhanced by our proposed revisions of DSM-III. It may be that history is on our side. The majority of the psychiatrists who participated in the DSM-III field trials identified themselves as psychodynamically oriented. Hence, if the APA is serious about wanting to broaden the base of participation in future DSMs, they may take our suggestions seriously.