The Cycle of Classification: DSM-I Through DSM-5 - part 1

Blashfield, R. K., Keeley, J. W., Flanagan, E. H., & Miles, S. R. (2014). The Cycle of Classification: DSM-I Through DSM-5. Annu. Rev. Clin. Psychol, 10, 25-51.‏

The struggles and controversies surrounding mental disorder definitions in the various DSMs remind me of the struggles surrounding learning disability definition.  This is one of the reasons it's good to know them.

DSM = Diagnostic and Statistical Manual of Mental Disorders

This paper reviews DSM development. The paper uses the term "patients" to describe people in need of psychiatric intervention and, thus, DSM definitions.  I'll use the term "patients" in this post following this paper.  The way we name or refer to our customers reflects a value judgement, to my opinion (I prefer to call them "clients").  This is also the case with the way we name our clients' problems (this paper names them "mental disorders").

Here are some interesting points from this paper.

We begin with this fascinating table:

NAME	PUBLICATION YEAR	NUMBER OF PAGES	NUMBER OF DIAGNOSTIC CATEGORIES	PRICE IN DOLLARS	REVENUE FOR THE AMERICAN PSYCHIATRIC ASSOCIATION IN MILLIONS OF DOLLARS
DSM1	1952	132	128	3$	UNKNOWN
DSM2	1968	119	193	3.5$	1.27
DSM3	1980	494	228	31.75$	9.33
DSM3R	1987	567	253	NOT WRITTEN IN THE PAPER	16.65
DSM4	1994	886	383	$48.95	120
DSM4TR	2000	943	383	$74.95	UNKNOWN
DSM5	2013	947	541	$199	UNKNOWN

DSM1

Following the Second World War, there were four classification systems of mental disorders in the US.   The American Psychiatric Association (APA) decided to overcome this “Tower of Babel” situation by creating a classification that would be acceptable to all members of its organization and that could unify the diagnostic terms of its psychiatrists. The result was the DSM1.    DSM-I had a hierarchical system in which the initial node in the hierarchy was differentiating organic brain syndromes from “functional” disorders. The functional disorders were further subdivided into psychotic versus neurotic versus character disorders. This organization roughly followed the decision-making process of clinicians.

 The DSM-I descriptions of disorders were prose paragraphs that incorporated behavioral and trait-like criteria. The terms in the description were relative and left to the interpretation of the clinician, leading to problems with reliability across professionals.

DSM2

DSM2 published in 1968 was the result of an effort to unite the classification systems in the world.  It was organized in a similar way as the DSM1.  Many of the new categories added in the DSM-II were categories of relevance to outpatient mental health efforts.

In 1971 Kendell et al conducted a study in which a set of eight videotapes of patients from the United States and Great Britain were shown to groups of American and British psychiatrists. For all eight videotapes, the modal diagnosis by the American clinicians was schizophrenia. In contrast, some of the videotapes, in the opinion of the British psychiatrists, represented patients with manic-depressive disorders, schizophrenia, and personality disorders.   The results, that showed there are still differences in diagnoses between professionals in different countries, were considered as evidence that Americans tended to be over inclusive in their use of schizophrenia as a diagnosis.

In 1973 Rosenhan published a provocative paper in Science about how a group of colleagues went to different inpatient facilities in the United States requesting admission. They were truthful about themselves during the intake interview except for two things: (a) they gave fictitious names so that their admissions would not appear on their future medical records, and (b) they reported hearing a voice saying “Empty” or “Thud.” All were admitted with a diagnosis of schizophrenia. Their average length of stay in the inpatient facility was nineteen days (the total range was 7 to 52 days). When discharged, most of them were given a diagnosis of “schizophrenia, in remission.” Rosenhan and his colleagues noted that most of the patients in the facilities spotted that they were fakes, but none of the pseudopatients were detected by the hospital staff. Rosenhan concluded that inpatient facilities of the time could not differentiate the sane from the insane.  Rosenhan’s paper stirred up a firestorm of reactions

.

These and other studies were stimuli for changes in the DSM3.   

DSM3

Publishing the DSM-III in 1980 was part of a paradigm shift in psychiatry (and the mental health field in general). Prior to the DSM-III, psychiatry was dominated by psychoanalytically trained psychiatrists. These psychoanalysts saw little value to clinical diagnosis for working with psychotherapy patients. In contrast, the main authors of the DSM-III attempted to bring psychiatry back to its medical roots. Their ideas fit well with the transition in treatment focus from psychotherapy to the use of medications.  

The DSM3 authors will to drop the term "neurosis" from the DSM raised a lot of controversy among professionals.  A compromised was reached in which the word "neurosis" appeared in parenthesis following the word "disorder" in specific cases.

Robert Spitzer, who was the head of the DSM-III and the organizing committee for the DSM-III took the bold step of proposing a tentative definition of the concept of mental disorder. They needed this definition because an explicit goal of the creators of the DSM-III was to avoid speculations about the causal mechanisms (especially theoretical concepts couched in psychoanalytic terms) that explained psychopathology. This definition was also in direct contrast to the antipsychiatry movement that attempted to define a mental disorder as society’s way of dealing with undesirable people—by labeling them with a mental disorder to keep them quiet and segregated.

The definition in the DSM-III was: Each of the mental disorders is conceptualized as a clinically significant behavioral or psychological syndrome or pattern that occurs in an individual and that is typically associated with either a painful symptom (distress) or impairment in one or more important areas of functioning (disability). In addition, there is an inference that there is a behavioral, psychological, or biological dysfunction, and that the disturbance is not only in the relationship between the individual and the society.

The DSM-III definition of mental disorder led to an interesting and growing discussion of psychiatric classification by philosophers, cognitive psychologists, social anthropologists, and historians.

The DSM-III contained diagnostic criteria to specify the meaning of the categories. In addition, for each category, there was a description of the typical demographic profile of patients experiencing this disorder, a lengthy prose explanation of what the category meant, a description of how to differentiate the target category from any other category with which it might be confused, and a brief discussion of what was known, if anything, about the course and onset of the disorder. Another innovation to the DSM-III was that the system was multiaxial. Each patient was expected to be diagnosed along five separate axes: (a) the descriptive presentation of the patient (i.e., the mental disorder categories), (b) the underlying personality and/or intellectual disorder, (c) any associated medical disorder that was relevant to the patient’s psychiatric presentation, (d ) the psychosocial stressors in the patient’s environment, and (e) the patient’s highest level of adaptive functioning in the past year.

 After the publication of the DSM-III, Spitzer and his colleagues created the SCID (Structured Clinical Interview for DSM-III-R).   By the year 2000 there were over 240 instruments to measure various aspects of psychopathology and mental disorders.  The reliability of diagnostic assessment using these new instruments generally was a distinct improvement over what had been found in the pre-DSM-III research. Spitzer & Fleiss’s (1974) review of pre-DSM-III reliability research showed estimates of interclinician agreement typically ranging from 0.4 to 0.6. Using structured interviews like the SCID, reliability estimates were distinctly higher, typically in the range of 0.75 to 0.90. Because of the clearly defined method for assigning psychopathology, along with improved reliability, structured interviews would soon dominate the research world although even today they are rarely used in clinical practice.

How reflective was the DSM3 structure of the "natural" way clinicians conducted diagnoses?

Cantor et al. (1980) had 13 mental health clinicians list the features that they associated with nine DSM-II diagnostic categories of psychosis. Any feature that was chosen by at least 3 of the 13 clinicians was kept for the final feature list. Then they took twelve case histories of patients that had been given one of four psychotic diagnoses (manic, depressed, paranoid schizophrenia, and undifferentiated schizophrenia). Four cases were considered to be quite prototypical of the four diagnoses (i.e., these four cases contain almost all of the features generated by the 13 clinicians), four were moderately prototypical, and four were not typical (i.e., these four cases had four or less of the defining features generated by the 13 clinicians). These case histories were given to the clinicians to diagnose. The reliability of the diagnoses varied as a function of the prototypicality of the case histories, with the least prototypical cases having the lowest reliability.  This research and others  suggested that clinicians did not use diagnostic criteria to make diagnoses.  Clinicians’ diagnoses tended to follow a prototype-matching model rather than a criteria-based model.

DSM3R

DSM3R that was published in 1987 was not structurally different than the DSM3 (it had the axis system, used diagnostic criteria and the mental disorders were organized in a similar way) but it contained new categories.

The political struggles concerning the DSM-III centered around a battle between a psychoanalytic faction of the APA and a biologically-oriented faction. When the DSM-III-R was being created, the focus of controversy shifted. Feminists were concerned with proposals by the DSM-III-R committees for new categories such as premenstrual syndrome and masochistic PD. As a result of the controversy, the DSM-III-R added a new appendix to its classifications called “Proposed diagnostic categories needing further study.” Contained in this appendix were three categories: late luteal phase dysphoric disorder (the new name for premenstrual syndrome), sadistic PD (to balance masochistic PD), and self-defeating PD (the new name for masochistic PD).  

In 1992, WAKEFIELD raised attention the fact that a value judgment is necessary to instantiate any definition of mental disorder.   The same symptoms might be judged as disordered in one context but not in another. Further, as societal and individual values change over time, some conditions that used to be disordered will no longer be considered abnormal (e.g., homosexuality), and others that were not disordered might become problematic (e.g., Internet use). Thus, there can never be a “final” version of the DSM.

To be continued in a following post…