Criminality among individuals testing positive for the presence of anabolic androgenic steroids

Arch Gen Psychiatry. 2006 Nov;63(11):1274-9. doi: 10.1001/archpsyc.63.11.1274.


Context: Observations suggest that the use of anabolic androgenic steroids (AAS) may trigger uncontrolled, violent rage. Other observations indicate that certain groups of criminals may use AAS with the intention of being capable of committing crime more efficiently.

Objective: To examine the proposed association between the use of AAS and criminality.

Design: A controlled retrospective cohort study of registered criminal activity among individuals tested for AAS use during the period of January 1, 1995, to December 31, 2001.

Setting: All individuals in Sweden who were tested for AAS use during this period. These individuals were referred for testing from both inpatient and outpatient clinics as well as from centers for treatment of substance abuse.

Participants: Individuals testing positive for AAS (n=241), with those testing negative for AAS during the same period (n=1199) serving as the control group.

Main outcome measures: The ratios (expressed as relative risk [RR]) of the incidences of several categories of crime in the 2 study groups.

Results: The risk of having been convicted for a weapons offense or fraud was higher among individuals testing positive for AAS than among those testing negative (RR, 2.090 and 1.511, respectively; 95% confidence interval [CI], 1.589-2.749 and 1.208-1.891, respectively) whereas there were no significant differences with respect to violent crimes (RR, 1.116; 95% CI, 0.981-1.269) or crimes against property (RR, 0.942; 95% CI, 0.850-1.044). When patients referred from substance abuse centers were excluded, a lower risk for crimes against property was observed for the individuals who tested positive for AAS (RR, 0.761; 95% CI, 0.649-0.893) and the risk for fraud in the 2 groups was equalized (RR, 1.117; 95% CI, 0.764-1.635). The increased risk for a weapons offense among the individuals testing positive for AAS remained virtually unchanged.

Conclusions: In addition to the impulsive violent behavior previously shown to be related to AAS use, such use might also be associated with an antisocial lifestyle involving various types of criminality. However, the existence and nature of this possible association remain unclear and call for further investigation.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adult
  • Anabolic Agents / administration & dosage
  • Anabolic Agents / adverse effects*
  • Androgens / administration & dosage
  • Androgens / adverse effects
  • Antisocial Personality Disorder / diagnosis
  • Antisocial Personality Disorder / epidemiology
  • Cohort Studies
  • Crime / legislation & jurisprudence
  • Crime / psychology*
  • Crime / statistics & numerical data*
  • Female
  • Firearms / legislation & jurisprudence
  • Fraud / legislation & jurisprudence
  • Fraud / statistics & numerical data
  • Humans
  • Incidence
  • Life Style
  • Male
  • Rage / drug effects
  • Retrospective Studies
  • Risk Factors
  • Substance Abuse Detection / statistics & numerical data*
  • Substance-Related Disorders / diagnosis
  • Substance-Related Disorders / epidemiology*
  • Substance-Related Disorders / psychology
  • Sweden / epidemiology
  • Violence / legislation & jurisprudence
  • Violence / statistics & numerical data


  • Anabolic Agents
  • Androgens