Horwitz · Journal of internal medicine 2019 · retrospective matched cohort study · n=13079

Health consequences of androgenic anabolic steroid use.

Cited 208 times in the scientific literature.

Level 3 - non-randomized controlled study

Retrospective matched cohort study with an external replication cohort

PubMed 30460728 · doi:10.1111/joim.12850 · record verified 2026-08-27

What was done

In a retrospective matched cohort study in Denmark (2006–2018), 545 male subjects who tested positive for androgenic anabolic steroids (AAS) in fitness centres were matched to 5,450 male controls. A replication cohort included 644 men sanctioned for refusing a doping test matched to 6,440 controls. Mortality, annual hospital contacts, and specific clinical morbidities were evaluated between groups.

What was found

Mortality was significantly elevated in AAS users compared to non-user controls (hazard ratio 3.0, 95% CI 1.3–7.0). The median annual number of hospital contacts was 0.81 in AAS users versus 0.36 in controls (P < 0.0001). Prevalence of acne, gynaecomastia, and erectile dysfunction each exceeded 10% in AAS users and was significantly higher than in controls (P < 0.0001). Results were replicated in the sanction-refusal cohort.

Why it matters

This study provides large-scale, population-matched observational evidence linking recreational anabolic steroid use to substantially increased all-cause mortality, healthcare utilization, and endocrine-related morbidity.

Limits

The study sample was identified through targeted anti-doping testing in fitness centres, which may introduce selection bias. Unmeasured confounders such as specific dosages, drug regimens, duration of use, poly-substance use, and lifestyle factors were not reported in the abstract. The study included only male participants.

Cited by