Reviewed research

Authors Horwitz H., Andersen J.T. and Dalhoff K.P.

Review Date December 2018

Citation J Intern Med. 2018 doi: 10.1111/joim.12850. [Epub ahead of print]



The use of androgenic steroids once the domain of elite athletes and competitive body-builders, is now popular in the general population, especially among young men attending gyms. These performance and image enhancing drugs are taken to increase muscle bulk and strength, and for faster recovery. For many men their use is associated with body dissatisfaction disorders including muscle dysmorphia, and eating disorders. Negative body image is a primary driver for androgen use among young men.

Since men who use androgens tend not to engage with primary healthcare professionals and there is limited research in the area, the long-term side effects aren’t fully clear. However, androgen use has been linked to serious conditions including cardiovascular disease, liver and brain damage, rhabdomyolysis and embolism. Other undesirable side effects include infertility, gynecomastia, acne and an increased risk of contracting infectious disease. Given that the prevalence of androgen use is estimated to be 6%, this practice represents a major public health concern. A study by Danish researchers performed a large-scale retrospective control-case study to investigate mortality and morbidity among androgen users.



To investigate long-term survival and the prevalence of co-morbidities among androgen users compared to age and gender matched controls.



Under the Danish anti-doping programme, collaborating gyms (capturing 80% of gym users in Denmark) undergo annual inspections and testing of gyms users suspected of using androgenic steroids. Between January 2006 and March 2018, 545 men who were laboratory-confirmed androgen users were included in the study. A second validation cohort included 644 men who were strongly suspected of using androgens but refused testing. Both cohorts were compared to age and gender matched control cohorts (n=5450 and n=6440, respectively).

Retrospective health outcomes were derived using national personal identification numbers linked to the healthcare system, and these outcomes were captured until May 2018. The primary endpoint recorded was all-cause mortality; secondary endpoints included the incidence and prevalence of somatic co-morbidities related to androgen use.



In the androgen use confirmed cohort (n=545), the average age in which men were tested positive was 26.2 years (standard deviation [SD] 6.3). The average length of follow-up was 7.4 years (SD 2.9), which was similar to the control group (7.3 years; SD 3.0). At the end of the follow-up period, the all-cause mortality rate (primary endpoint) was seven cases (1.3%) in androgen users, compared to 23 cases (0.4%) in controls (hazard ratio [HR] 3.0; 95% CI 1.3-7.0). Hospitalisations were higher among androgen users compared to control groups (1.26 versus 0.68 hospital contacts per person/year).

The hazard ratios (HR) for specific co-morbidities in androgen users are summarised in the following Table.

Disorder (secondary endpoints) HR for androgen users (95% CI) P-value (corrected for multiple comparisons)
Male infertility 2.42 (1.6-3.7) <0.001
Testicular dysfunction 21.86 (7.6-62.9) <0.0001
Non-ischaemic heart disease (incl. cardiomyopathy and atrial fibrillation) 2.93 (1.7-5.0) <0.01
Thromboembolic disorders  5.0 (1.7-14.6)  <0.05
Prescription for acne medication  2.27 (1.5-3.4)  <0.001
Prescription of erectile dysfunction medication  3.10 (2.2-4.4)  <0.0001
Gynaecomastia 13.29 (8.3-21.4) <0.0001

Findings related to primary and secondary endpoints were replicated in the cohort of presumed androgen users.



Men who use androgens have a significantly higher risk of mortality, more hospital presentations and an increased risk of multiple co-morbidities, compared to men of the same age, from the general population.


Points to Note

1. Although there was no significant association between androgen use and ischaemic heart disease in these young men (HR 1.7, 95% CI 0.6-4.8), longer term follow-up would be required to see if this conditions develops with advancing age.
2. After correcting for multiple comparisons, there were no significant differences in the prevalence of kidney disease, cancers or liver disease. Longer-term follow-up is needed to determine if these slower-progressing conditions are more prevalent as a result of androgen use.
3. The analysis found that hospitilisation contact for hernias, skin disorders (cutaneous abscesses), boils, hives, and rheumatological disorders were more common among AAS users.
4. The increased risk of mortality among androgen abusers agree with previous population studies conducted in Sweden and Finland.
5. A study limitation noted by the authors, was that based on population estimates, up to 6% of the control subjects could be androgen users, leading to possible under-estimation of their effects when comparing outcomes across the case-control groups.
6. Another acknowledged limitation was the possibility of confounding factors that were not measured, such as the association between the use of performance and image enhancing drugs, drug use and risky behaviours in general, being linked to higher rates of mortality and morbidity.


Website: https://www.ncbi.nlm.nih.gov/pubmed/30460728

Reviewed by
Updated on