Predictors of 'all-cause discontinuation' of initial oral antipsychotic medication in first episode psychosis

Introduction: Discontinuation of the initial oral antipsychotic prescribed for a first episode of psychosis (FEP) can derail outcome. Our objective was to examine the rate of and time to all-cause discontinuation of the first antipsychotic prescribed and the factors influencing such discontinuation.

Methods: In a sample of 390 FEP patients, we estimated the rate of and time to discontinuation of the initial antipsychotic over a one-year period. The effects of a number of putative predictors of discontinuation were estimated using regression analyses.

Results: Rate of discontinuation of the first antipsychotic was 72%, with no difference between the 3 investigated antipsychotics (olanzapine (73%), risperidone (68%) and aripiprazole (75%)), (χ² (2) = 1.89, p = 0.388). Mean time to discontinuation was 7.2 (4.6) months and was not different among the three antipsychotics (Log-rank χ² (2) = 0.257, p = 0.879). Binary logistic regression showed that higher positive and negative symptoms remission and baseline functioning were associated with lower rates of discontinuation (Nagelkerke R² = 0.36, χ² (10) = 66.9, p < 0.001). Multiple linear regression showed the same predictors, in addition to male gender and less weight gain per month of exposure to the initial antipsychotic, to be associated with longer time to discontinuation (adjusted R² = 0.336, F (9, 219) = 13.8, p < 0.001).

Conclusion: Discontinuation of the initial antipsychotic is a major concern in the course of treating FEP. Symptom relief, better functioning and lower side effects appear to be the major factors associated with continuing an antipsychotic medication.