Associations between diet quality and testosterone: evidence from mixture and mediation analysis

BACKGROUND: While higher diet quality is associated with improved testosterone levels, prior population-based evidence relied on binary testosterone deficiency outcomes. The continuous dose-response relationship, component-specific contributions and adiposity-mediated mechanisms linking diet quality to testosterone remain unclear. OBJECTIVE: To examine the association between overall diet quality and serum testosterone in adult men, identify key dietary components and evaluate adiposity-mediated pathways. METHODS: This cross-sectional study included 5434 adult males from the National Health and Nutrition Examination Survey (2011-2016). Diet quality was assessed using Healthy Eating Index-2020 (HEI-2020) from 24-h dietary recalls. Testosterone was quantified by liquid chromatography-tandem mass spectrometry. Associations were evaluated using multivariable linear regression, restricted cubic spline analysis, mixture models (weighted quantile sum, quantile g-computation and Bayesian kernel machine regression) and mediation analysis. RESULTS: Higher HEI-2020 scores were significantly associated with increased log-transformed total testosterone (β per 10-point increase = 0.015; 95% CI: 0.002-0.028; P = 0.028), demonstrating a significant positive linear dose-response relationship. Mixture analyses assigned highest relative weights to sodium and dairy components. The effects were potentially mediated through body mass index and waist circumference. CONCLUSIONS: Higher overall diet quality exhibits a positive linear association with elevated serum testosterone in adult men. In exploratory mixture and mediation analyses, sodium, dairy and adiposity were identified as possible contributors.