Schizophrenia: syndromes and diseases

Abstract

A century of research suggests that schizophrenia is a clinical syndrome and not a single disease. From the time of Kraepelin (1971) and Bleuler (1950) onward, unifying theories have been put forward, but no documentation of schizophrenia as a single disease entity has yet been compelling. Studies comparing a schizophrenia group with a comparison group often show di€erences without clarifying whether the di€erence relates only to a subgroup of subjects. If a fairly generalized schizophrenia di€erence is observed, it may not be clear if it relates to disease pathology (e.g. psychosis) or a common phenomena (e.g. dysphoric a€ect) which is not a direct disease manifestation. Clinical syndromes are de®ned on the basis of certain shared signs and/or symptoms. It is usually understood that this heuristic grouping is a ®rst step intended to facilitate the investigation of etiopathophysiology and treatment. The hope is that such groupings will eventually lead to the identi®cation of separate diseases with distinguishing treatment, etiology, and, ®nally, preventive measures. To the extent that psychopathology and treatment are based on ®nal common pathways, distinct diseases will still share features common to the syndrome, for example, antipsychotic ecacy of neuroleptic drugs across etiologically distinct classes. Despite the general acceptance of the syndromal status for schizophrenia, theoretical models and study designs routinely conform to the single disease paradigm. The major consequence of addressing a syndrome as though it were a single disease is that progress in understanding the diseases that cause the syndrome is greatly impaired. Consider the problem of investigating the genetics or neuropathology of Alzheimer's disease, if study cohorts comprised all cases of dementia. While multi-infarct dementia and the dementia due to B12 de®ciency both have an abnormality in memory, including such cases in research studies would undermine the study of Alzheimer's Disease (McHugh and Slavney, 1983). A parallel consideration may account for many of the small-e€ect size studies and the nonreplications that are so common in schizophrenia research. It is easy to imagine the adverse e€ect in a gene linkage study if the probands with a diagnosis of schizophrenia actually have di€erent diseases, with di€erent genetic predispositions. We consider it likely that the syndrome of schizophrenia will be resolved into several diseases, paralleling progress in mental retardation over the past 50 y. The task then becomes to discover the speci®c diseases, so that progress can be made in understanding the pathophysiology of each, and thereby improve treatment and devise preventive measures. Fortunately, a strong empirical basis for making this e€ort has emerged from the factor analytic studies of the last 20 y (Buchanan and Carpenter, 1994). A relatively consistent ®nding in these studies has been the existence of (at least) three factors, representing positive symptoms (hallucinations and delusions), negative symptoms (blunted a€ect, poverty of speech, and a loss of motivation), and a disorganization of thought and behavior. Because of these ®ndings, we concluded that at present the best heuristic for dissecting the heterogeneity of schizophrenia was to identify a major psychopathological component of the syndrome which behaves difJournal of Psychiatric Research 33 (1999) 473±475