Objectives Inflammation contributes to chronic diseases. Lower serum testosterone among men is associated with less inflammation, yet immune defense is thought to trade‐off against reproduction with androgens adversely affecting immune function. Anti‐androgens are effective at castrate levels of serum testosterone, suggesting serum testosterone may not capture all androgen activity. The association of two androgen biomarkers with key markers of inflammation was examined. Methods The adjusted association of serum testosterone and androstanediol glucuronide with C‐reactive protein, white blood cell, granulocyte and lymphocyte count, fibrinogen, and hemoglobin, as a control outcome because testosterone administration raises hemoglobin, were examined in a nationally representative sample of 1,490 US men from the National Health and Nutrition Examination Survey III phase 1 (1988–1991) using multivariable linear regression. Results Serum testosterone and androstanediol glucuronide were weakly correlated (0.13). Serum testosterone was associated with lower white blood cell count [−0.26 × 10−9 per standard deviation, 95% confidence interval (CI) −0.37 to −0.14] and granulocyte count (−0.21 × 10−9, 95% CI −0.29 to −0.13) but not with hemoglobin (0.02 g/l, 95% CI −0.89 to 0.92), adjusted for age, education, race/ethnicity, smoking, and alcohol. Similarly adjusted, androstanediol glucuronide was not associated with white blood cell count (0.10 × 10−9, 95% CI −0.05 to −0.25), granulocyte count (0.12 × 10−9, 95% CI −0.02 to 0.25), or fibrinogen (0.05 g/l, 95% CI −0.004 to 0.11), but was with hemoglobin (0.70 g/l, 95% CI 0.07 to 1.32). Conclusions Different androgen biomarkers had different associations with inflammatory markers, highlighting the need to consider several androgen biomarkers. The possibility remains that androgens may generate inflammatory processes with implications for chronic diseases. Am. J. Hum. Biol., 2013. © 2013 Wiley Periodicals, Inc.