Testosterone products are increasingly being prescribed to males for a variety of possible health benefits but the causal relationship between testosterone and health-related outcomes is unclear. Evidence from well-powered randomized controlled trials are difficult to obtain, particularly regarding effects on long-term or adverse outcomes. We sought to determine the effects of genetically-predicted calculated free testosterone (CFT) on 23 health outcomes.

Genetic variants associated with CFT were determined from 136,531 white British males in the UK Biobank. One-sample and two-sample Mendelian randomization (MR) analyses were performed to infer the effects of genetically-predicted CFT on 23 health outcomes selected based on relevance with known or suspected effects of testosterone therapy.

In males from the UK Biobank, 81 independent genetic variants were associated with CFT levels at genome-wide significance (p<5×10⁻⁸). Each 0.1 nmol/L increase in genetically-predicted CFT was associated with clinical benefits on increased heel bone mineral density (0.053 SD; 95% CI = 0.038 to 0.068; p=8.77×10⁻¹²) and decreased body fat percentage (−1.86%; 95% CI = −2.35 to −1.37; p=1.56×10⁻¹³), and adverse effects on increased risk of prostate cancer (OR=1.28; 95% CI=1.11 to 1.49; p=1.0×10⁻³), risk of androgenic alopecia (OR=1.82; 95% CI = 1.55 to 2.14; p=3.52×10⁻¹³), risk of benign prostate hyperplasia (BPH) (OR=1.81; 95% CI = 1.34 to 2.44; p=1.05×10⁻⁴) and hematocrit percentage (1.49%; 95% CI = 1.24 to 1.74; p=3.49×10⁻³²).

Long-term elevated free testosterone levels cause prostate cancer, BPH, and hair loss while reducing body fat percentage and increasing bone density. It also has a neutral effect on type 2 diabetes, cardiovascular and cognitive outcomes. Well powered randomized trials are needed to address the effects of shorter term use of exogenous testosterone on these outcomes.