Objective: While doctors often increase the dose of an antipsychotic when there is insufficient response, there is limited evidence that this intervention is any better than waiting longer on the lower dose. We put the proposition to test.

Method: In this 4-week, double-blind, randomized controlled trial conducted in psychiatric care from September 2012 to March 2015, 103 patients with schizophrenia (ICD-10) who did not respond to olanzapine 10 mg/d or risperidone 3 mg/d were randomly allocated to a dose-increment or -continuation group. In the increment group, antipsychotic doses were doubled for 4 weeks, whereas in the continuation group, doses were not changed. Completion rate (primary outcome measure); changes in psychopathology, function, and extrapyramidal symptoms; and response rate were compared between the groups. The relationship between baseline plasma antipsychotic concentrations and changes in psychopathology was examined.

Results: The completion rate was significantly lower in the increment group than in the continuation group (69.2% [36/52] vs 86.3% [44/51], P = .038). No significant superiority was observed in any of the outcome measures in the increment group compared to the continuation group, except the Positive and Negative Syndrome Scale (PANSS) positive subscale score change in intention-to-treat analysis. Those with lower plasma concentrations of olanzapine on their initial treatment showed a greater improvement on the PANSS positive subscale when their dose was increased (P = .042).

Conclusions: As a general strategy, patients with schizophrenia failing to respond to moderate antipsychotic doses may not benefit from an increase in dose. The possibility of benefit in those whose plasma antipsychotic concentrations at baseline are still low cannot be ruled out.

Trial Registration: UMIN.ac.jp/ctr/index.htm identifier: UMIN000008667