Review of: Tang Y et al, Curr Neuropharmacol2023;21(2):424–436

Study Type: Randomized, open-label clinical trial 

How long must one wait before making changes after initiating an antipsychotic? Conventional wisdom suggests four to six weeks, but emerging evidence indicates benefits may be observed as quickly as two weeks. The current study tested if early efficacy on an antipsychotic predicts subsequent efficacy.

In a multicenter, randomized, open-label clinical trial, the authors randomized Chinese patients with first-onset or relapsed schizophrenia (n=3010) to seven antipsychotic drugs (risperidone, olanzapine, quetiapine, aripiprazole, ziprasidone, perphenazine, and haloperidol). Participants were monitored for six weeks, with symptom severity measured biweekly using the Positive and Negative Syndrome Scale (PANSS). Logistic regression and random-forest models were used to determine if reductions in PANSS at week two and/or week four predicted later efficacy (≥50% reduction in PANSS at week six).

Over the six-week study, 1,682 of the participants (56%) achieved response. Both prediction models revealed that week two was the earliest time point that predicted efficacy at six weeks. Results remained significant after adjustment for common sociodemographic factors (eg, age, education) and illness factors (eg, course, baseline severity). No differences in predictive performance were observed between the different antipsychotics; however, early response to first-generation antipsychotics better predicted later efficacy than early response to second-generation antipsychotics.

The strengths of the study included the large sample size and inclusion of non-European participants. Limitations included confounding due to the many participants with relapsing vs first-onset illness, arbitrary threshold for efficacy, permitted use of benzodiazepines, potential for placebo effects, and short follow-up period.

Carlat Take

Upon initiating an antipsychotic medication, evaluate response at two weeks. Should there be minimal or no improvement, consider switching medications, increasing the dose, or adding other medications, as the prospects for significant improvement later appear low. In the case of first-generation antipsychotics, lack of early improvements should be a strong indicator for switching as further dose increases may cause more serious adverse effects.