Abstract
A study on 243 schizophrenic inpatients treated with neuroleptics was performed. Using standardized rating instruments a large set of potential prediction variables was tested. Several predictors, known from the literature, could be confirmed, also by cross validation according to the split half technique: among others premorbid disturbances of social adjustment, insidious onset of illness, duration of symptoms before admission, intensity of symptoms at admission, duration of previous hospitalisation. The prognostic scale of Strauss and Carpenter, originally developed for long term prognosis, proved as the most relevant predictor. By stepwise multiple regression analyses an optimal combination of five predictors, explaining 30% to 40% outcome variance was found. The latter result is being tested in a prospective study.

This publication has 0 references indexed in Scilit: