Psychophysiological predictors of drug treatment response.

Few of the psychophysiologic findings reviewed above are diagnostically specific. Not all persons given the same psychiatric diagnosis are likely to share the biological characteristic while others with different diagnoses do. The first and most simple explanation commonly offered is to refer to inadequacies in clinical diagnostic procedures or in the particular diagnostic system that was used in the study. It is true that the reliability of psychiatric ratings are typically lower than the reliability of psychophysiologic measures, but the situation appears to be more complex. In considering the issue of the relationship between psychiatric diagnosis and psychophysiological data, it is important to realize that psychiatric diagnosis may have several purposes only one of which is predicting drug treatment response. In addition, the degree to which a biological characteristic is present before the illness and between episodes, or in relatives or normal subjects having traits suggestive of vulnerability to the disorder, are important aspects of psychophysiological investigations. The state vs trait nature of the measure is of frequent concern, and changes in a psychophysiologic measure as a function of drug treatment are considered a part of this broader issue. Furthermore, psychophysiologic measures seem to be tapping underlying dimensions which cross-cut current diagnostic boundaries to a greater or lesser degree depending upon the measure and the subject samples. Candidates for such underlying dimensions include illness severity, anxiety, arousal, attention, cognitive impairment, neuronal loss, intelligence and mood, to mention a few. Scores on these dimensions may predict drug response to a higher degree than diagnosis. The use of drug treatment response itself to validate subgroups of individuals identified by diagnosis is a common assumption in psychophysiology. Thus, patients who are responsive to a drug are assumed to have a different underlying illness than patients who are resistant. In clinical nosology, this assumption is not usually taken for granted. However, Brown and Hertz have recently argued for the need to pay more attention to identification and classification of patients along a neuroleptic response dimension. Diagnostic systems based on psychophysiological measures have been developed. One such system is "neurometrics" which involves comparing individuals on electrophysiological measurements, such as EEG power spectra, against a data base of normative values previously obtained from normal subjects. John et al. have shown that classifications based on neurometrics corroborate clinical diagnostic categories to a high degree, and can be used to independently validate current psychiatric nosology.(ABSTRACT TRUNCATED AT 400 WORDS)