Sex differences in the efficacy, acceptability, and tolerability of antipsychotics in people with acute schizophrenia: individual participant data pairwise, dose-response, and network meta-analyses

Antipsychotics are the mainstay of schizophrenia treatment, but their efficacy and tolerability might differ between females and males, and previous evidence on sex differences has been inconsistent and methodologically limited.

We aimed to evaluate sex differences in the efficacy, acceptability, and tolerability of antipsychotics in adults with acute schizophrenia. We performed individual participant data pairwise, dose-response, and network meta-analyses of antipsychotics versus placebo or another antipsychotic, using data from the Vivli and YODA trial data-sharing platforms (inception to April 9, 2025); the main analyses were restricted to fixed-dose trials.

The primary outcome was overall symptoms on the Positive and Negative Syndrome Scale at week 6 (or closest). Secondary outcomes included all-cause dropout and other efficacy and tolerability outcomes.

We performed one-stage pairwise meta-analyses to quantitatively test sex-treatment interactions, and two-stage sex-stratified dose-response and network meta-analyses to visually examine potential sex differences. Credibility of effect modification was assessed with the Instrument to assess the Credibility of Effect Modification Analyses.

People with lived experience were involved in planning and interpretation of the study. This study was registered with PROSPERO, CRD420251022957.

We identified 33 eligible trials; 18 fixed-dose trials with 7062 participants (2168 females and 4894 males; of 5161 with race reported, 2954 White, 1187 Black, 752 Asian, and 268 of other or unspecified race; mean age 40·9 years; receiving haloperidol, olanzapine, paliperidone, risperidone, or placebo) were included in the main analyses. There was no evidence of sex differences in drug effects on overall symptoms (difference of standardised mean difference [DoSMD] 0·04, 95% CI -0·09 to 0·18).

Dose-response curves were similar in females and males, plateauing at 3-5 mg risperidone equivalents, and the network meta-analysis showed no clear sex-specific treatment effects. Antipsychotics reduced all-cause dropout more in females than in males (ratio of odds ratios 0·74, 95% CI 0·56-0·98; moderate credibility).

Prolactin increase was greater in females (DoSMD 0·83, 0·67-0·99; high credibility); other tolerability outcomes showed no clear sex difference.

We found no evidence of sex differences in the efficacy or overall adverse event profiles, and no different dose-response relationships, for the drugs studied.

However, we found some evidence that, among females, antipsychotics reduce all-cause dropout more, and increase antipsychotic-induced prolactin elevation versus placebo, as compared with males. Overall treatment efficacy estimates from previous meta-analyses are relevant to both sexes.

To promote sex and gender equity, female participation in clinical trials and sex-stratified reporting should be encouraged.