Understanding the motivation to use anabolic steroids: an application of the theory of reasoned action

The nonmedical use of anabolic-androgenic steroids (AAS) is an emerging public health concern. Despite a recent call for AAS prevention by the National Institute on Drug Abuse (Lin & Erinoff, 1990), little theory-based research has been conducted on the etiology of AAS use. In the present study, the Theory of Reasoned Action (TRA) was used as a model for understanding and explaining the psychosocial factors which motivate weightlifters to use AAS.

One-hundred-three male weightlifters (mean age = 23.3 years, range 18-39) completed a survey that assessed past drug use (including past AAS use) and constructs relevant to the TRA: attitudes about personal AAS use, subjective norms regarding personal AAS use, and intentions to use AAS over the next year. In addition, perceived behavioral control was assessed and the TRA was compared to the Theory of Planned Behavior (TPB).

Overall, seventeen (16.50/0) of participants reported the past use of AAS and nine (8.7%) reported the intention to use AAS at sometime over the next year. Multiple regression (MR) analyses revealed that respondents' attitudes (Beta = .72, D, < .001) towards personal AAS use significantly predicted AAS intentions and explained 55% of the variability in the intention to use AAS (i.e., R2 = .55). Further, when perceived behavioral control (PBC) was added to the model, both attitudes (Beta=.54, D,<.001) and PBC (Beta =-.37, D,<.001) were significant predictors of AAS intentions. With respect to other drug use behaviors, the past use of AAS was not significantly correlated with the use of alcohol, cigarettes, chewing tobacco, marijuana, or cocaine. Similar results were found for an adolescent sub-sample of 59 male weightlifters (Mean age = 19.9).

These results provided strong support for the TRA (TPB) as an accurate model for explaining male weightlifters' motivation (intentions) to use AAS. Implications for the primary prevention of AAS based on these models are discussed in detail.