Short-term remission in schizophrenia as a combination of several outcome measures

Clinical, cognitive, metabolic and functioning variables have been evaluated in patients with schizophrenia in an 8-week trial with Ziprasidone. The aim of this post-hoc analysis is to investigate how these variables interact in determining short-term remission. Baseline values or the variation from baseline to endpoint were considered predictors. 262 schizophrenic patients were recruited. Two logistic regressions were conducted to determine which variables predict remission. The first was performed on baseline values as predictors. The second used the variation from baseline to endpoint (delta) of the outcome evaluations as predictors. Using literature reported criteria for remission, we distinguished 124 subjects (47.33%) in remission, and 138 not in remission at the end of the trial. The first logistic regression does not show a good fit. The second logistic regression, with delta scores as predictors, reports instead an overall good fit (71.8% of the predicted cases assigned to the right category). The analysis reveals that general score from Positive and Negative Syndrome Scale (PANSS), cholesterol LDL, subjective well-being under neuroleptic, Simpson–Angus Scale (SAS) delta scores, drug dosage and premorbid intelligence entered in the equation. These results suggest that indexes such as cognition, metabolic status, other than symptoms, have to be taken into account in order to refine the short-term remission prediction.