Pharmacoepidemiology of anabolic androgenic steroids : a review / Ingemar Thiblin, Anna Petersson. - (Fundamental & Clinical Pharmacology 19 (2005) 1 (February); p. 27-44)
- PMID: 15660958
- DOI: 10.1111/j.1472-8206.2004.00298.x
Non-prescribed use of anabolic androgenic steroids (AAS) has been associated with a number of physical and psychiatric/behavioural complications, some of which are potentially lethal. Here, we review both observational and experimental studies on human subjects concerned with such side-effects. The only physical complication of AAS use that receives definitive support from such investigations is unfavourable changes in blood lipid profiles. Support for various psychiatric complications has also been provided by a number of cross-sectional studies, most involving comparisons between weight-training individuals who use or do not use AAS. Certain of these complications, in particular hypomania and increased aggressiveness, have been confirmed in some, but not all, randomized controlled studies. Epidemiological attempts to determine whether AAS use triggers violent behaviour have failed, primarily because of high rates of non-participation. Studies regarding the prevalence of AAS use in different populations typically report life-time prevalences of 1-5% among adolescents. However, the life-time prevalence (i.e. use on at least one occasion) is of doubtful relevance in attempting to estimate the number of individuals at risk for side-effects, as most of these complications appear to develop during prolonged use of AAS. Furthermore, it is reasonable to assume that the symptoms and signs of AAS use are often overlooked by healthcare professionals, so that the number of cases of possible AAS-related complications is virtually unknown. These limitations, together with an apparently low prevalence of prolonged AAS use among the general population, indicate that future epidemiological research in this area should focus on retrospective case-control studies and, perhaps, also on prospective cohort studies of populations selected for a high prevalence of AAS use, rather than attempting to perform large-scale population-based studies.