Predictors of risk for relapse in patients with schizophrenia or schizoaffective disorder during olanzapine drug therapy

PurposeTo evaluate the relationship of dose decrease, symptom worsening, and baseline covariates on subsequent relapse during olanzapine treatment in patients with schizophrenia or schizoaffective disorder.MethodsIn two 28-week, randomized, double-blind clinical trials, a Cox proportional hazards model was used to determine potential correlates of relapse (defined as ⩾20% worsening on PANSS total and CGI-Severity ⩾3) among patients (N = 271) who responded to 8 weeks of olanzapine treatment (10–20 mg/day). Variables examined included: demographics, illness characteristics, baseline symptoms, symptom change, dose, adverse events, and functioning.ResultsPatients with a lower last dose relative to the preceding visit interval were 4 times more likely to relapse during that visit interval than other patients (p < .001). A similar finding was observed for a decrease in interval modal dose, although this variable was more predictive of relapse in the visit interval immediately following dose decrease (p = .027). In a subgroup analysis by gender, there was a significantly greater incidence of relapse in men with a dose decrease, whereas a dose decrease in women did not correlate with relapse. Relapse was also correlated with the emergence or worsening of a psychiatric adverse event during the same (p < .001) and preceding (p = .007) visit intervals, and with increased rating scale measures of psychopathology. The occurrence of a non-psychiatric adverse event was not associated with relapse.ConclusionDose decrease is a significant predictor of relapse in male but not female patients. Psychiatric adverse events also predicted relapse. Patients should be periodically reassessed to determine the need for maintenance treatment with appropriate dose.