In humans, serum growth hormone (GH) concentrations are significantly higher in women than in men, but the neuroendocrine mechanisms that underlie such gender differences are not known. We compared normal episodic GH secretion in males and females in three distinct settings: two human studies employing quite different assay techniques (immunoradiometric assay and a high-sensitivity immunofluorimetric method) and a rat study. To quantify the amount of regularity in data, we utilized approximate entropy (ApEn), a scale- and model-independent statistic. In each study, females exhibited significantly greater statistical irregularity in GH concentration series than their male counterparts (P < 10(-3) for each human study, P < 10(-6) for the rat study), implying that mass and mode of GH secretion are regulated differently in males and females. The regularity comparisons indicated complete gender separation (100% specificity and sensitivity) for the rat study and nearly complete separation for the immunofluorimetric assay study. The consistency and statistical significance of these findings suggest that this gender difference may be broadly based within higher animals and that this may be readily evaluated objectively by analysis of ApEn.