Drug-related and illness-related factors in the outcome of chlorpromazine treatment: testing a model.

Patients who presented with acute psychoses and were treated with chlorpromazine were first divided into 2 groups with good (23) and poor (13) outcome. These outcome groups differed little in their initial clinical features and showed no difference in 2 indices of dopamine receptor blockade (extrapyramidal symptoms and plasma prolactin concentrations). The group which improved was then subdivided on the basis of evidence of dopaminergic blockade into 15 who had improved and also showed anti-dopamine effects ('responders') and 8 who had improved but showed no anti-dopamine effects ('remitters'). The remainder were eventually classified as 'resistant' to the effects of the drug. The group of 'remitters' contained no patients with nuclear schizophrenia; the 'responders' were mainly nuclear schizophrenics; and the 'resistant' patients were schizophrenic or schizo-affective. The 3 groups who were defined in this way also differed in their subsequent clinical course. It is suggested that this scheme for dividing patients may be useful in clinical work and could also assist research worker to identify the patients who can most appropriately be studied to determine mechanisms of drug action.