Prediction of response to acute neuroleptic treatment in schizophrenia

Predicting the outcome of treatment with neuroleptics and understanding the factors that contribute to variability in the response to these drugs have preoccupied researchers and clinicians since the class was developed. Clinicians need empirically based prospective criteria in order to choose a treatment strategy that will lead to the best response from the patient while minimizing side effects. Thus the optimal strategy will match patient, illness and drug characteristics. Before a response can be predicted, the researcher needs explicit concepts and definitions on the course of the illness, drug responses, treatment outcome and predictive criteria. Whereas outcome refers to an arbitrary endpoint of an illness, whether spontaneous or modified by treatment, response is a treatment-related concept. Response refers to a change in the course of the illness which is defined either by pre- or post-treatment criteria and is caused by treatment. Apart from treatment, the spontaneous course of the illness is shaped and modified by various factors, which are referred to as potential outcome/response predictors sampled from a wide area of patient, illness and environmental characteristics. A superordinate concept of prediction must show how these potential determinants of (treatment) course and outcome relate to each other in biopsychosocial terms. Valid predictions of response can only be expected if recognizable rules are operative in the illness course and the treatment outcome. With increasing knowledge about the pathophysiology of the illness, its course and potential determinants, the development of valid predictive algorithms will be promoted. The application of vigorous research strategies in future pharmacological treatment studies will also be required to reach this goal.