Prolegomenon to the Clinical Prerequisite; Psychopharmacology and the Classification of Mental Disorders
(Volume 1)

Conceptual Development of Current Psychiatric Nosology

Thomas A. Ban and Antonio Torrez Ruiz

Theoretical Issues

There are two major theoretical controversies relevant to psychiatric classification.

Categorical Model

The controversy whether diagnostic classification in psychiatry should be based on a "dimensional model" or a "categorical model" began with the dialogue between the adherents to Griesinger's (1845) concept of "unitary psychosis," a "dimensional concept" -- which implies that "psychiatric disorders" are based on a "large number of continuously distributed measures on one or more dimensions in a homogenous space" -- and the followers of Kraepelin's (1896) "categorical nosology," which implies that "psychiatric disorders" are based on a "large number of discontinuously distributed elements of two or more categories in a heterogeneous space" (Pichot, 1986). Essentially the same controversy was reflected in the dialogue between the "separatists" and the "gradualists" (Stengel, 1959). For the "separatists,” "psychoses" were "autonomous" disease entities, i.e., categories, qualitatively different from the "neuroses" and "character disorders," whereas for the "gradualists" mental pathology was distributed on a "continuum," i.e., on a "dimension" from the "normal" to the "psychotic," that was considered to be only quantitively different from the "neurotic" (Lehmann, 1971).

One of the most powerful arguments against the "categorical" model and in favor of the "dimensional" model in psychiatric classification was put forward by Eysenck (1960). His argument was based on the findings that factor analytic and related techniques have consistently failed to replicate the clustering of the features implicit in "categorical nosologies." After failing to demonstrate "bimodality" between "affective disorder" and “schizophrenia" (by "discriminant function analysis"), a similar argument against the "categorical model" was raised by Kendell and Gourley (1970). Nevertheless Roth (1978) maintained that with consideration to the "clear, and in some instances qualitative differences (which) have been demonstrated between these two conditions in respect of heredity, fertility, measures of personality, course, outcome, response to treatment" among many others," it is the validity and relevance of the particular dimension (they have chosen) that fails to reveal discontinuity." Because of this, according to him, the only valid conclusion one can derive from such a study is that the dimensions chosen for differentiation have been wrong and "not the independence of schizophrenic and affective psychoses as nosological entities."

The dialogue was reopened by Kendell's (1982) suggestion that a "dimensional classification" of depression offers at least "theoretically" some advantages for biological research (Philipp and Maier, 1986). According to him, "transforming the graduation in symptomatology between two syndromes into a linear variable allows to select groups with high diagnostic homogeneity" (Philipp and Maier, 1986). But even such an "exercise would be (considered as) impractical" by Roth (1978), because "starting from a clinical tabula rasa, which ignores all existing categorical entities, the number of dimensions required for categorizing patients would be without limit." Consequently, "to reduce the task to manageable proportions, a clinical diagnosis would have to be made first."

In recent years, Loranger (1981) has presented findings on the "genetic independence of manic-depression and schizophrenia," in support of the “categorical model.” In spite of this, on the basis of complex genetic "categorical model" considerations, Crow (1986) put forward a speculation on "the continuum of psychoses" in favor of the "dimensional model" of psychiatric disorders.

Hierarchical Order

In the late 1970s Pope and Lipinski (1978), Koehler (1979) and Berner (1982) challenged Jaspers (1962) contention that in case of the presence of both, "schizophrenic psychopathological symptoms" outweigh "affective psychopathological symptoms." According to these authors they are the "affective" and not the "schizophrenic psychopathological symptoms" which have differential diagnostic significance. This is to the extent that not only current, but also prior "affective psychopathology" and/or family history of "major affective disorder" may be of relevance in ruling out a diagnosis of "schizophrenia."

One of the arguments used in favor of the "reversed hierarchical principle" is that all original concepts relevant to "affective disorder" have remained by and large unchanged, whereas all the original concepts relevant to "schizophrenia" have "continually been the subject of scientific disagreements and new attempts at formulation" (Berner et al., 1983). Another argument in favor of the "reversed hierarchical principle" is based on Janzarik's (1949, 1959) contention that Schneider's (1957) "first rank symptom of schizophrenia" are the result of "dynamic--affective--instability," one of the three forms of "dynamic derailment" that may arise in a variety of disorders, e.g., "transitory psychosis," "mixed psychosis" and even "manic-depressive psychosis." Because of this, Berner et al. (1983) suggested that "schizophrenic symptoms in general are fully unspecific phenomena which may appear under various psychotic conditions." A third and possibly the most powerful argument used in favor of the "reversed hierarchical principle" is the significantly greater predictive validity of the "endogenomorphic cyclothymic (affective) axial syndrome" than of the "endogenomorphic schizophrenic axial syndrome" (Berner et al., 1986).

The controversy regarding the "hierarchy" of disorders in psychiatric classifications has not affected diagnostic practice. Although some of the recent diagnostic classifications allow for multiple diagnoses, in general the "traditional hierarchy" -- "organic dementias," "functional psychoses," and "personality disorders" -- still prevails. One of the main reasons for this is that it is difficult to diagnose a "functional psychosis," i.e., a schizophrenic disorder or an affective disorder in case of the presence of an "organic disorder"; and impossible to diagnose a "personality disorder" in case of the presence of any of the other disorder (although the DSM-III-R requires an Axis II - personality disorder diagnosis regardless of the Axis I - clinical syndrome diagnosis.)