The primary outcome was efficacy as measured by overall change in symptoms of schizophrenia. In terms of overall efficacy:
- Olanzapine was more efficacious than quetiapine and haloperidol. By how much? The standardized mean difference (same as effect size) was 0.29 for both comparisons. A standardized mean difference of 0.2 is considered to be a small effect size, 0.5 is moderate, and 0.8 is large. So this superiority of olanzapine was relatively small.
- Clozapine was more efficacious than haloperidol. By how much? Standardized mean difference 0.22.
That’s it. Two comparisons that showed a small advantage of one antipsychotic. For all other comparisons, the pairs of antipsychotics compared were not different in their efficacy for “treatment-resistant” schizophrenia.
Next, rather than overall efficacy, the meta-analysis also looked specifically at improvement in positive symptoms.
- Risperidone was more efficacious than quetiapine (standardized mean difference 0.43)
- Clozapine was more efficacious than quetiapine (standardized mean difference 0.40)
- Olanzapine was more efficacious than quetiapine (standardized mean difference 0.33)
- Risperidone was more efficacious than haloperidol (standardized mean difference 0.29)
- Clozapine was more efficacious than haloperidol (standardized mean difference 0.27)
So, risperidone, olanzapine, and clozapine stood out as more efficacious while quetiapine was the antipsychotic that stood out as less efficacious.
What about negative symptoms? Olanzapine was more efficacious than clozapine (very small effect size, though), risperidone, haloperidol, and chlorpromazine.
Bottomline: overall, a general pattern of superiority for olanzapine, clozapine, and risperidone was seen, though these results were not consistent and effect sizes were usually small.
Relatively few studies in “treatment-resistant” schizophrenia were available for antipsychotics other than clozapine, haloperidol, olanzapine, and risperidone. Thus, many antipsychotics (e.g., aripiprazole) were not included in this meta-analysis.