Predictors of clozapine response in patients with treatment-refractory schizophrenia: results from a Danish Register Study

Abstract

We aimed to identify factors associated with greater clozapine response to guide targeted clozapine use. The study was based on data from the Danish Psychiatric Central Research Register and the National Prescription Database including schizophrenia patients initiating clozapine from 1997 to 2006. Cox regression was used to identify predictors of time to psychiatric hospitalization and all-cause discontinuation from first clozapine prescription. In a 2-year mirror-image design, multiple logistic regression models were used to identify predictors of psychiatric hospitalization. Among 633 schizophrenia patients starting clozapine, shorter time to admission was predicted by increasing number of different antipsychotics (hazard ratio [HR], 1.08/trial; confidence interval [CI], 1.01-1.15/trial) and admissions (HR, 1.04/admission; CI, 1.03-1.05/admission) before first clozapine prescription, earlier onset of schizophrenia (HR, 0.98/y; CI, 0.96-0.99/y), and lower clozapine dose (HR, 0.07/100 mg; CI, 0.03-0.13/100 mg). In the 2-year mirror-image model, during clozapine treatment, there was a significant reduction in bed-days (269.9 days [CI, 238.3-287.8 days] to 64.2 days [CI, 53.0-79.3 days], P < 0.001) and admissions (3.4 [CI, 3.1-3.6] to 2.2 [CI, 1.9-2.5], P < 0.011). Being admitted during clozapine treatment was also associated with more antipsychotic trials (odds ratio [OR], 1.11; CI, 1.00-1.22) and admissions before clozapine initiation (OR, 1.08; CI, 1.04-1.11) and female sex (OR, 1.84; CI, 1.31-2.58). Although the study design does not allow any causal inferences, all 3 models suggested a lower number of psychiatric hospitalizations and antipsychotic trials before clozapine initiation to be associated with greater clozapine response.