Higher Serum Testosterone Protects Against the Development of Type 2 Diabetes in Middle‐Aged but Not Older Men

BACKGROUND: Low serum testosterone concentration has been reported to be causally linked to the risk of Type 2 diabetes (T2D). The benefit of T2D protection has been demonstrated to continue up to serum testosterone concentrations of 461 ng/dL (16 nmol/L). It is unclear if there is an upper threshold for this effect of testosterone nor whether this causal process is moderated by age, waist circumference (WC) or baseline glycaemia (A1c). AIM: To examine the effect of WC, age and baseline glycaemia on the relationship between testosterone and incident T2D risk. METHODS: Men without T2D, cancer or testosterone treatment and with baseline and 5-year follow-up assessments in the Men Androgen Inflammation Lifestyle and Environment Study cohort were included. The association between baseline testosterone and 5-year incident T2D was assessed using confounder-adjusted logistic regressions, with and without restricted cubic splines and pairwise interactions with age, WC and baseline A1c. Primary analyses were in the complete case set, with multiple imputation being performed as a sensitivity analysis. RESULTS: The analysis included 1299 men with 109 cases of 5-year incident T2D (8.4%). An independent inverse relationship between serum testosterone concentration and 5-year incident T2D was observed (OR = 0.92, 95% CI = [0.87, 0.97], p = 0.001). There were no detectable interactions between testosterone and either WC (p = 0.93) or baseline A1c (p = 0.38). There was, however, evidence for a strong interaction between testosterone and age (p = 0.003), with low testosterone being associated with elevated T2D risk only in younger men. CONCLUSIONS: These data suggest that higher concentrations of serum testosterone are protective against incident T2D in younger but not older men.