Debate continues about the best way to subcategorize schizophrenia. The classic subtypes (undifferentiated, catatonic, paranoid, and hebephrenic) have given way in the literature over the past decade to the current research schema (McGlashan, 1992) which delineates symptom domains which are probably semi-independent of each other: positive symptoms (hallucinations, delusions) and negative symptoms (e.g., flat affect and poverty of speech). More recently a third domain, thought disorganization, has been added to cover the cognitive and attentional impairment (Carpenter, 1993). This sub-categorization, which has been validated by studies of the natural history of the disorder, some brain imaging studies, neuropsychological testing, and differing drug response, seems to indicate that there may be different physiological processes involved in the symptom domains, and certainly different time courses for the development of these symptoms (figure 1). It is thought that a patient with enduring negative symptoms (flat affect, social difficulties) has an increased risk of poor prognosis, chronicity, neurological defects, and possibly increased genetic loading. Positive symptoms (hallucinations and delusions) correlate with early phases of the disorder. Nevertheless, the DSM IV (see DSM IV for diagnostic criteria) has not fully adopted this in its classification schema, and relies on subcategories of paranoid, disorganized, catatonic, undifferentiated (those not meeting criteria for the first three subcategories), and residual types (in which there are no prominent positive symptoms). The later is an attempt to build a bridge between the old and the new.
Where are we heading in the future classification of schizophrenia? If one reviews the history of sub-categorization of other diseases, such as diabetes (a state of excessive urinary output), one can see that we are in the early phases of the process. With the advent of technologies such as brain imaging, animal models, and neurochemical studies, we are moving toward the next step in sub-classification. We will move from a sub-classification system based on phenomenology (symptom profiles and course of illness) to one based on the ability to document abnormal physiology. In diabetes, this step was represented by the realization that in most cases excessive urine volume was accompanied by elevated blood sugar, but in other cases the blood sugar was normal, and in these cases the cause of the increased volume resided in kidney malfunction. The outward appearance of the two disorders was very similar, but effective treatment was quite different. Whether schizophrenia turns out to be the result of one abnormal physiological process or several is yet to be seen (see discussion below).