Testing whether prolactin and glucose associations differ by sex

Lan and colleagues report a significant inverse correlation between prolactin and fasting blood glucose in males but not females with first-episode schizophrenia. Separate within-sex significance tests do not establish that these associations differ by sex. A pooled prolactin-by-sex interaction analysis, together with a confidence interval for the difference between sex-specific estimates, would directly test heterogeneity. Clarifying the medication-restricted sample denominator and reconciling inconsistent fasting-insulin descriptions would further define what the data support; until then, the findings should be treated as hypothesis-generating rather than evidence of a male-specific metabolic mechanism.