Number of sexual partners and serum testosterone levels in a population-based sample of unpartnered heterosexual men.

Abstract

The relationship between men's testosterone levels and their sexual behaviors is unclear as existing studies find mixed results. The key objective of this study is to assess whether men's testosterone levels are affected by sexual behaviors and to explore whether this relationship varies by age. Specifically, this study addresses the question: Are men's testosterone levels affected by the number of sexual partners they accrue? This analysis was based on a nationally representative sample of 977 unpartnered heterosexual men between the ages of 20 and 65 in the United States. Serum testosterone levels were measured from blood samples using isotope dilution liquid chromatography tandem mass spectrometry and the number of sexual partners in the past 12months was ascertained from an audio computer-assisted personal self-interview. Using ordinary least squares regression, serum testosterone levels were predicted as a function of the number of sexual partners that men reported, sociodemographic characteristics, and factors known to affect endocrine regulation. An interaction term between the number of sexual partners in the past 12months and age was included. The correlation between the number of sexual partners in the past 12months and testosterone levels is contingent upon age: The relationship is positive for younger men but negative for older men. Among 20-year-old unpartnered heterosexual men, an additional sexual partner in the past 12months is associated with an increase of 5.55ng/dl of testosterone (95% confidence interval (CI): 2.92, 8.19). This initial positive effect at age 20 is reduced by 0.25ng/dl of testosterone (95% CI: -0.44, -0.06) with each year that men age. Testosterone levels in men are related to reproductive behaviors, but the direction of this relationship changes as men age.