CAROL S. NORTH,
SEAN H. YUTZY
The
diagnostic approach in psychiatry does not have a long and illustrious tradition in the United
States. Historically, Freud and his
influential psychoanalytic school in Europe during the late 1800s focused on symptoms that were felt to be of
‘‘psychogenic’’ origin. Subsequently, the
fundamental principles of psychoanalysis were imported into the United States,
where they predominated for almost three-quarters of the twentieth century.
The
American Psychiatric Association (APA) published the first Diagnostic
and Statistical Manual of Mental Disorders (DSM) in 1952 (1), moving American
psychiatry incrementally toward a descriptive
approach. Although that text included various psychiatric conditions with
diagnostic labels, it did not provide explicit criteria, and most symptoms were conceptualized in a
framework of a reaction to a situation
or stressor.
Consistent
with the predominating Freudian approach at that time in America, symptoms were
considered to originate from underlying dynamic conflicts, representing symbolic
meaning. Understanding a patient’s
symptoms required exploration of that individual’s own developmental history and life experiences (14).
Problematic
to scientific
advancement of the field within this approach was the flaw that such prevailing assumptions about the
etiology of psychiatric problems were scientifically untestable (18). Further,
problems of unreliability plagued symptom interpretation, a process
unique to the individual patient and inconsistent from
one clinician to the next.
Definition
of psychiatric problems using these approaches poorly differentiated healthy from
ill individuals and hampered communication among clinicians about their cases.
In 1968, the APA published the second edition
of the text (DSM-II) (2), which
embraced the disease concept for mental illness—but again without explicit criteria.
Diagnostic
reliability was not yet a feature of the proffered labels. To find its place
among the medical sciences, psychiatry
needed to develop an epidemiological system for classifying disease,
known as ‘‘nosology’’ (14).
Several
European physicians specializing in problems of psychiatry (Kraepelin, Kahlbaum, Bleuler, and
others) who predated or were contemporaries of Freud found it more useful to
describe the phenomenology of mental
illness as clustering in ‘‘syndromes’’ according to their unique symptoms, course, and outcome over time. These
physicians followed an earlier
and evolving approach to disease initiated by Sydenham in the late
1600s and further developed over the centuries by
Koch, Pasteur, Virchow, and
others.
The
relevant assumption by most of these investigators was that identification of a syndrome would
lead to a better understanding of the illness
and possibly etiology. Kraepelin considered mental disorders to be
commensurate with physical diseases. He followed
medical traditions of careful
observation of many cases to describe overt characteristics of illness,
rather than relying on unproven etiological theories.
Kraepelin
predicted that
empirical research would eventually provide evidence of biological
origins of mental illness (18, 20). The classification/diagnostic approach and attendant
scientific methods were
not systematically applied in American psychiatry until the advent of the
‘‘descriptive school’’ in the latter half of the
twentieth century, whose traditions were founded on the earlier work of
Sydenham, Kraepelin, and others.
During
the 1960s and early 1970s, the Washington University group in St. Louis, described as ‘‘an outpost of
diagnostically oriented thinking’’ (p. 259) (18), refined the descriptive approach to psychiatry
and distinguished themselves as radically different from the prevailing
psychoanalytic school. In American
psychiatry at that time, the belief that mental disorders were a subset of medical disorders was controversial (18) if
not heretical. Members of this
group, often referred to as ‘‘neo-Kraepelinians,’’ were described by some
as ‘‘organic’’ or ‘‘biological’’ in orientation, but
their position regarding the origin
of symptoms was simply ‘‘what the evidence or reproducible data demonstrated.’’
In
1970, Robins and Guze (17) at Washington University in St. Louis published their proposal for the
phases necessary to establish the validity and reliability of a psychiatric diagnosis. The five
phases included the following:
1. Clinical
description. The first step is to describe the clinical picture. Race, sex, age at onset,
precipitating factor(s), and other items may be used to define the disorder. The clinical picture
described in the United States should be the
same as in the United Kingdom, Russia, and
elsewhere.
2. Laboratory studies. These include any laboratory test of a chemical, physiological, radiological, or anatomical (biopsy or autopsy) nature that could reproducibly demonstrate a finding.
3. Delimitation from other disorders. Symptoms of the disorder of interest must be distinguishable from other potentially overlapping disorders.
4. Follow-up studies. The first diagnosis is reconfirmed at follow-up (usually years later), supporting the original diagnosis.
5. Family studies. Most psychiatric illnesses have been shown to run in families, whether of genetic or environmental origin. Identification of the illness in the family supports the validity of the diagnosis.
Feighner
et al. in 1972 (10) outlined the diagnoses and criteria that met these phases. Included were the
following: primary affective disorder (mania/ depression), secondary affective disorder,
schizophrenia, anxiety neurosis, obsessive-compulsive
neurosis, phobic neurosis, hysteria, antisocial personality disorder,
alcoholism, drug dependence, mental retardation, organic brain syndrome, anorexia nervosa, and several others.
It
should be noted that with the possible exception of brain syndrome(s),
laboratory studies still have not been developed that reproducibly demonstrate
pathological lesions. However,
thoughtful and careful completion, with replication, of most of the
phases (1, 3, 4, and 5) firmly established the bases
for the validity of the major mental
illnesses.
To
gain a foothold in the published literature, this group published their early work in diagnostic psychiatry
in British and other European literature, outside the reach of American hostility to their ideas
(Samuel B. Guze, personal communication). In
1974, the first edition of Psychiatric Diagnosis, co-authored by Robert A. Woodruff, Jr., Donald W.
Goodwin, and Samuel B. Guze,
was published by Oxford University Press (19). Although written for
the education of medical students, it became an icon
for the field. At the time, this
text was the most authoritative source of diagnostic psychiatry, as the
current diagnostic manual of the APA had not yet
incorporated empirically based
criteria.
In
the late 1970s, Robert Spitzer, with remarkable political finesse, convened meetings (Samuel B. Guze, personal
communication) of members of the two formal
psychiatric schools, psychoanalytic and evidence-based/biological (18) to
generate
the third edition of the Diagnostic and Statistical Manual (DSM-III) (3).
The
text was generally intended to be atheoretical in orientation regarding the origin of most psychopathology. (Of
note, to appease an opposition that threatened to block the document’s
approval, the word neurosis, a concept
fundamental to Freudian psychoanalysis, was retained parenthetically after the
word disorder (18). Neurosis did not
survive the next revision of the diagnostic text.) DSM-III was the first of the APA’s diagnostic
manuals to embrace the Washington University (St. Louis) group’s emphasis
that medical diagnosis is the
keystone of medical practice and research (18). The
DSM-III text included practically en bloc the Feighner criteria for the
diagnoses of the major mental illnesses.
Unfortunately for validity and reliability considerations, the text also
included approximately 213 other ‘‘mental disorder’’
labels with essentially untested
criteria. Nonetheless, the creation of a common language utilizing
valid and reliable psychiatric diagnoses, as well as
defining explicit criteria, was an
important advance.
While
some considered the advances of DSM-III a revolution,
many data-oriented psychiatrists saw it as simply a move to align with the
substantially evolved
general medical model of the time. With DSM-III, psychiatric
diagnosis became practical and useful to clinicians as well as researchers.
Psychiatrists
now had psychiatric diagnoses that provided clear boundaries for major psychiatric illnesses
within dichotomies rather than dimensions that blurred boundaries between normal and abnormal.
These diagnoses were based on documented
observations of psychopathology rather than on unsupported theoretical etiologic mechanisms (18).
Of
particular note was the fact that diagnostic reliability (measured by the kappa statistic
designed to measure diagnostic concordance or agreement about
diagnosis among examinations) of the major diagnoses moved into the good to
excellent range, equaling the established
reliability of medical diagnoses made on clinical grounds.
The
data/evidence-based school continued describing its approach to psychiatric nosology simply as
‘‘medical model’’ (dating back to 1973). This term is broader than ‘‘biological’’ (implying
limitation within purely physical processes
such as genes and brain chemistry). The medical model is more descriptive of the categorical/diagnostic approach,
which encompasses all aspects
of illness, including environmental and social aspects, within medical
definitions and descriptions.
At
first, the academicians involved with the development of DSM-III and those embracing the new medical
model of psychiatry were alleged to constitute an ‘‘invisible college’’ of
professionals trained at Washington University dispersed to leadership positions at the University of
Iowa, the University of Minnesota, the University of Kansas, and Columbia
University in New York City, among other sites. As their
numbers and their publications grew and these clinician/scientists blended into the landscape
of American psychiatry under
the aegis of DSM-III and its
predecessors, the invisible college grew into the mainstream institution.
Prevailing
forces in medicine and psychiatry at the time facilitated and embraced the changes brought about
with DSM-III (18). Prior to DSM-III, psychiatry had largely fallen
outside the mainstream of the science and practice of medicine. Psychologists, social workers, and other
mental health professionals threatened to replace psychiatrists.
Critics
of psychiatry claimed that definitions of mental illnesses did not have objective
criteria, asserting that psychiatric
illness did not even exist. DSM-III provided
objective diagnostic criteria and positioned psychiatric diagnosis within
the medical purview.
Psychiatry
re-asserted its place in medicine as a medical specialty. Evolving computer technology contributed
substantially to the investigation of empirical aspects of disease. Advances in diagnosis provided
powerful incentives for the pharmaceutical
industry and independent researchers to develop new drugs and other treatments targeted for treatment of
specific disorders.
Insurance companies sought diagnostic
structure to guide reimbursement for treatment of validated disorders with demonstrable evidence of
specified treatments for them.
The achievement of diagnostic reliability led to new research opportunities and
directions (18).
Subsequent
versions of the APA textbook (DSM-III-R (4), DSM-IV (5), and DSM-IV-TR (6)) further solidified and
reaffirmed the transformation of American psychiatry to an empirically based system of
classification of diagnosis that DSM-III initiated
(18). They also, however, further increased and expanded the number of labeled
mental disorders (5, 6).
Unfortunately,
the new
labels promulgated have not met the phases noted above, which over time have become the ‘‘gold standard’’ for establishing
validity and reliability of a psychiatric
diagnosis. In sum, the vast majority of diagnostic labels in the latest
version (DSM-IV-TR) have not
met the diagnostic gold standard. Not surprisingly, research on many of the
coded non-major ‘‘mental disorders’’ or labels during the 1990s and later have led
to significant questions about comorbidity (problems with delimitation). (For specific examples
of these problems, posttraumatic stress disorder and borderline personality
disorder.)
The
effects of the medical model movement in psychiatry have been substantial. The National Institute
of Mental Health embraced the DSM-III and subsequent versions of this text as the authoritative
text for psychiatric disorder classification. Shortly after the publication
of DSM-III, students and physicians from American medical
schools and residency programs were expected
to pass qualifying examinations based on APA diagnostic criteria.
Scholarly
journals and federal research granting agencies similarly assumed conformation to current diagnostic
conventions as delineated in APA criteria. Colleagues in different locales, even different
countries, could finally collaborate using a common diagnostic language (18).
Through
this history, the Psychiatric Diagnosis textbook
stayed alive—and grew—not
in numbers of diagnoses, but in size. Accumulated data in support
of the valid major mental illnesses over half a
century were slowly and painstakingly gathered into the second, third, fourth,
fifth, and now sixth editions. The original
authors have all passed away: first Bob Woodruff (1976), later Don
Goodwin (1999), and finally Sam Guze (2000). The
text’s new authors were trained
by the original members of the Washington University group, Eli Robins and Sam Guze, and, as students, read Psychiatric Diagnosis
along with the
hundreds and thousands of other students educated with this text over the
years.
The
text reviews the accumulated wealth of data over the years relevant to the
diagnoses included in it. The information is not meant to comprise a cutting-edge review of the latest
findings in research. Rather,
it is meant to provide fundamental and time-tested evidence for psychiatric
disorders to teach and support the principles of psychiatric diagnosis so
eloquently constructed through established empirical
research.
Newer trends that are buffeting the field of psychiatry.
Evolving
developments in biological technology are increasingly providing new directions for scientific
exploration. Many such leads, holding promise of fundamental advances in the understanding of
psychopathology, have disappointingly not delivered as anticipated.
A
classic example is the dexamethasone suppression test, introduced in the early
1980s, which was hoped to be psychiatry’s first laboratory test (8). This putative
marker generated great interest
and hundreds of articles; however, low sensitivity of the test limited
its clinical utility (7), although it has remained a
useful research tool.
Psychiatric
genetics is another area that was hoped to translate into nosologic insights that could
further revolutionize conceptualization and utility of psychiatric diagnosis. It has been
predicted that increasing elucidation of genotype–phenotype relationships may
eventually force the abandonment of traditional Kraepelinian dichotomy in the
classification of psychiatric illness
(9).
Continuing
research, however, has yielded growing frustration and no indications that genetic
research findings will fundamentally alter empirically based classification of psychiatric
illness any time soon. First, no ‘‘Mendelian-like’’
genes have been found for the major psychiatric disorders, and it is generally recognized that none will be
found.
Second,
efforts to carve nature
at its joints are not destined to cleave psychiatric disorders using purely genetic tools, because the psychiatric phenotype is
more than a product of its genes.
Kendler (12) recently concluded that even genes known to be involved
in the generation of psychopathology cannot provide
the basis for construction of psychiatric diagnoses. He predicted that
molecular genetics, like traditional
genetics, will offer disappointingly little in terms of advancing
conceptualization of psychiatric diagnosis.
Advances
in biological science continually challenge the field of psychiatry to evolve. Although an empirical
approach to classification in psychiatry was revolutionary in the 1970s when a small group of
academicians and their text Psychiatric Diagnosis were
instrumental in changing the field, the current classification system has now been generally accepted
for many years.
Pressures
to change have persisted, however, during the period of exponential growth in research since the first
edition of this text was published. Not surprisingly,
critics of the current classification system have replayed and amplified suggestions of past eras for a ‘‘dimensional
construction’’ of diagnostic concepts.
Although
past and some new voices continue to advocate for the inclusion of dimensional methodology in
definition of psychopathology, no clearly compelling reason has been advanced
to eliminate the categorical process of diagnostic assessment. Academic
proponents of dimensional components of psychiatric classification continue to champion the
categorical system of diagnosis while emphasizing advantages of adding
dimensional measurements to diagnostic
taxonomies (11, 13).
In
many contexts, categorical and dimensional measurements are interrelated and are sometimes
equivalent. Further, while a dimensional
approach may, to a degree, enhance data, statistical power, and measurement of psychiatric change, it does not
contemplate in many forms the actual
validity of the diagnostic assessment. Categorical considerations remain
unarguably vital in establishing inclusion criteria
for research studies and for making
clinical decisions about treatment (13).
Critics
of the categorical diagnostic approach rightfully point to the inflation of
number of recognized psychiatric diagnoses as problematic (15). Others who are troubled by the inability
to classify symptomatic individuals not meeting
diagnostic criteria promote concepts of ‘‘partial’’ or ‘‘spectrum’’ diagnoses.
Continuing
concerns about lack of clear demarcation between diagnostic cases and others
(16) have not recognized the original authors’ specification that approximately
one out of five cases assessed in the Washington
University tradition is not cross-sectionally classifiable in a diagnostic
structure. Thus, uncertainties in estimation of boundaries were built
into the diagnostic classification process,
circumventing otherwise unsolvable problems
of diagnostic decision making.
In
addition to movements to dimensionalize psychiatric classification systems, the field has encountered
significant external pressures to assess psychopathology
in short time frames and develop outcome measures to justify pharmaceutical/psychotherapeutic interventions. These
trends have encouraged the use of brief screening instruments for
identification of psychiatric disorders
and self-report questionnaires in both clinical and research settings.
These shortcuts, however, cannot substitute for
careful establishing of a psychiatric diagnosis.
Although
some screening tools may have utility in overall patient assessment, they do not
rely upon or yield adequate information for a formal diagnosis. Screening instruments may be
appropriate and useful for identifying
a high-risk subgroup when the population of interest is too large to
permit full diagnostic assessment of all its members
and to identify a subset of individuals
at high risk for illness who can then be further assessed using the
full diagnostic criteria.
Screening
instruments without formal interview, however, should not be relied upon for
assigning a psychiatric diagnosis that can be the basis for treatment decisions.
Further, screening instruments are inadequate for assessing general population
prevalence rates of psychopathology. Unfortunately,
the research literature is fraught with examples of studies inappropriately using screening instruments to
diagnose cases and estimate population
prevalence.
Overall,
psychiatric diagnosis is of paramount importance, and there is no shortcut to diagnosis using
screening tools to estimate likelihood within individual patients or entire
populations. Further, while dimensional aspects of illness may add to our understanding of psychiatric
disorders, they cannot replace psychiatric diagnosis.
For
the foreseeable future, psychiatrists and other medical doctors must continue to rely on
established psychiatric diagnoses and gathering information the old-fashioned
way—through clinical interviews
with patients—to guide assessment and treatment decisions.
Finally,
the current edition of this text flows from the original and seminal work by its predecessors to
establish and validate psychiatric diagnoses that support the mainstream activities of medical practice
and research. We do not entertain
the recently popular screening tools, notions of ‘‘partial’’ diagnosis, dimensional classifications, or disorders outside
those considered sufficiently validated
and useful within empirical traditions of psychiatric diagnosis.
Clearly,
the diagnostic traditions based on empirical findings over approximately five
decades—sufficiently revised in this text to accommodate new knowledge and the changing
field—will instruct future generations of students in the fundamental concepts of psychiatric diagnosis.
REFERENCES
IN
1. American Psychiatric Association. Diagnostic and
Statistical Manual of Mental Disorders, 1st edition. Washington, DC: Author,
1952.2. American Psychiatric Association. Diagnostic and
Statistical Manual of Mental Disorders, 2nd edition. Washington, DC: Author,
1968.3. American Psychiatric Association. Diagnostic and
Statistical Manual of Mental Disorders, 3rd edition. Washington, DC: Author,
1980.4. American Psychiatric Association. Diagnostic and
Statistical Manual of Mental Disorders, 3rd edition, revised. Washington,
DC: Author, 1987.5. American Psychiatric Association. Diagnostic and
Statistical Manual of Mental Disorders, 4th edition. Washington, DC: Author,
1994.6. American Psychiatric Association. Diagnostic and
Statistical Manual of Mental Disorders, 4th edition, text revision.
Washington, DC: Author, 2000.7. American Psychiatric Association Task Force on
Laboratory Tests in Psychiatry. The
dexamethasone suppression test: an overview of its current status in
psychiatry. Am. J. Psychiat., 144:1253–1262,
1987.8. Carroll, B. J. The dexamethasone test for
melancholia. Br. J. Psychiat., 140:292– 304, 1982.9. Craddock, N., O’Donovan, M. C., and Owen, M. J. Genes for schizophrenia and bipolar disorder? Implications for psychiatric nosology. Schizophr. Bull.,
9. Craddock, N., O’Donovan, M. C., and Owen, M. J. Genes for schizophrenia and bipolar disorder? Implications for psychiatric nosology. Schizophr. Bull., 32:9–16, 2006.
10. Feighner, J. P., Robins, E., Guze, S. B., Woodruff, R. A., Winokur, G., and Mun˜oz, R. Diagnostic criteria for use in psychiatric research. Arch. Gen. Psychiat.,
10. Feighner, J. P., Robins, E., Guze, S. B., Woodruff, R. A., Winokur, G., and Mun˜oz, R. Diagnostic criteria for use in psychiatric research. Arch. Gen. Psychiat., 26:57– 62, 1972.
11. Helzer, J. E., Kraemer, H. C., and Krueger, R. F. The feasibility and need for dimensional psychiatric diagnoses. Psychol. Med.,
11. Helzer, J. E., Kraemer, H. C., and Krueger, R. F. The feasibility and need for dimensional psychiatric diagnoses. Psychol. Med., 36:1671–1680, 2006.12. Kendler, K. S. Reflections on the relationship between psychiatric genetics and psychiatric nosology. Am. J. Psychiat., 163:1138–1146, 2006.13. Kraemer, H. C., Noda, A., and O’Hara, R. Categorical versus dimensional approaches to diagnosis: methodological challenges. J. Psychiat. Res., 38:17–25, 2004.14. Mayes, R., and Horwitz, A. V. DSM-III and the revolution in the classification of mental illness. J. Hist. Behav. Sci., 41:249–267, 2005.15. Mullen, P. E. A modest proposal for another phenomenological approach to psychopathology. Schizophr. Bull., 33:113–121, 2007.16. Neese, R. M., and Jackson, E. D. Evolution: psychiatric nosology’s missing biological foundation. Clin. Neuropsychiat., 3:121–131, 2006.17. Robins, E., and Guze, S. B. Establishment of diagnostic validity in psychiatric illness: its application to schizophrenia. Am. J. Psychiat., 126:983–987, 1970.18. Rogler, L. H. Making sense of historical changes in the Diagnostic and Statistical Manual of Mental Disorders: five propositions. J. Health Soc. Behav., 38:9–20, 1997.19. Woodruff, R. A., Jr., Goodwin, D. W., and Guze, S. B. Psychiatric Diagnosis. New York: Oxford University Press, 1974.
20. Young, A.
20. Young, A. The Harmony of Illusions: Inventing Post-traumatic Stress Disorder. Princeton, NJ: Princeton University Press, 1996.
References
CAROL S. NORTH,
SEAN H. YUTZY, 2010, Goodwin and Guze’s Psychiatric
Diagnosis, Sixth Edition, Oxford University Press, Inc
Read Also
The Psychiatric Evaluation
The Clinical Interview
Reality Testing and Cognitive Functioning
Completing the Diagnostic Assessment
The Clinical Interview
Reality Testing and Cognitive Functioning
Completing the Diagnostic Assessment
No comments:
Post a Comment