Design: Individual participant data pairwise, dose-response and network meta-analysis
Population: 7,062 adults with acute schizophrenia in 18 fixed-dose trials
Primary outcome: Sex-treatment interaction on PANSS total at about 6 weeks
Effect: DoSMD 0.04 (95% CI −0.09 to 0.18); prolactin DoSMD 0.83 (0.67–0.99) greater in women

This individual participant data meta-analysis drew on 18 fixed-dose trials from two data-sharing platforms, with 7,062 adults with acute schizophrenia (2,168 women) given haloperidol, olanzapine, paliperidone, risperidone or placebo. It tested formally whether sex changed the drug effect.

It did not, for overall symptoms at about six weeks (difference in standardised mean difference 0.04, 95% CI −0.09 to 0.18). Dose-response curves were alike in both sexes and flattened at about 3–5 mg risperidone equivalents. Two differences stood out: antipsychotics reduced all-cause dropout more in women (ratio of odds ratios 0.74, 0.56 to 0.98, moderate credibility), and raised prolactin more in women (difference 0.83, 0.67 to 0.99, high credibility).

For practice, this argues against dosing women lower on efficacy grounds, and for asking about prolactin effects deliberately. Galactorrhoea, amenorrhoea, sexual dysfunction and longer-term bone effects are under-reported and are common reasons women stop treatment.

Dose by response and tolerability, not by sex — efficacy and dose-response were the same.
Expect little extra benefit above about 3–5 mg/day risperidone equivalent in acute schizophrenia.
Ask women on risperidone, paliperidone or haloperidol directly about menstrual change, galactorrhoea and sexual function.
Check prolactin when symptoms suggest it, and consider a prolactin-sparing agent if raised levels cause problems.
These findings cover five drugs only; do not assume the same for others.

Why it matters

It removes a reason to dose by sex and puts prolactin at the centre of tolerability for women.

The statistics, in plain English

A difference of 0.04 with an interval from −0.09 to 0.18 means any sex difference in symptom response is small and could go either way — effectively none. A ratio of odds ratios of 0.74 means the drug cut dropout by a larger relative amount in women; its upper limit of 0.98 is close to 1, so that finding is less certain than the prolactin one.