DSM-III and the Infantile Psychoses
William Goldfarb, M.D., Ph.D.
Donald I. Meyers, M.D.
Judy Florsheim, M.S.
Nathan Goldfarb, Ph.D.
Professional interest in the classification of the psychoses of childhood extends back four to five decades. Despite this fact, diagnostic confusion has prevailed. The diagnostic labels that emerged in this period included, among others, well-known and frequently employed designations, such as childhood schizophrenia (Bender, 1947; Fish, 1977; Potter, 1933) and infantile autism (Kanner, 1943). Each of these labels referred to subclasses of children whose symptoms were diversified; and all these subclasses obviously overlapped in the kinds of children they subsumed and the varieties of psychopathy manifested. Perhaps for this reason there was a tendency to include all the varieties of psychotic children in one large, frankly heterogeneous population; and the most usual global diagnostic classification assigned to the wide variety of psychotic children was childhood schizophrenia. Indeed, Kanner's relatively precise criteria for the clinical diagnosis of infantile autism, a condition he pointedly distinguished from schizophrenia, were incorporated along with others into the nine criteria proposed by the British Working Party for the diagnosis of a category termed "schizophrenic syndrome of childhood" (Creak, 1961). These criteria were the basis for selecting a diversified group of children, including some who exhibited all the characteristic features of infantile autism as originally defined by Kanner and others who shared fewer and less severe autistic manifestations. The absence of precise, objective, and reliable criteria for the designation of nonoverlapping subclasses have been major hindrances to investigating the childhood psychoses.
Over the past decade, however, there has been a significant growth in information about psychotic children and consequent improvement in their objective classification into homogeneous subclasses of children. These classifications have been incorporated into the latest version of the American Psychiatric Association's (1980) Diagnostic and Statistical Manual—DSM-III.
The segment of DSM-III dealing with what are termed Pervasive Developmental Disorders of Childhood is a most significant effort to classify the childhood psychoses in a fashion that is objective, reliable, and universally acceptable. Already there is evidence that the DSM-III diagnoses of the Pervasive Developmental Disorders of Childhood are indeed reliable and usable, (Cantwell et al., 1979; Mattison et al., 1979; Russell et al., 1979). However, questions may reasonably be raised regarding the exhaustiveness of this diagnostic scheme and the degree of homogeneity of each of the subclasses of the Pervasive Developmental Disorders. We address ourselves here particularly to the capacity of this scheme to encompass the full range and diversity of infantile psychoses with onset below thirty months.
DSM-III provides operational criteria for the diagnosis of three subclasses of preadoles