Confidence in your own abilities can be a key factor in overcoming addiction to drugs and alcohol. Many interventions aim to improve this so-called self efficacy, but can they really change behavior?

Theresa Marteau and colleagues at Kings College London are among the first to explore such questions. They have good evidence that interventions can improve participants’ self-efficacy – defined by Albert Bandura as "the degree to which an individual believes he or she is capable of performing a particular task” – but as yet not much to suggest that such changes directly affect behavior.

Marteau and her colleagues searched academic databases for rigorous evaluations of programs that alter how young people see themselves with the goal of reducing drug use. They confined themselves to studies that conformed to Albert Bandura’s description of self-efficacy.

Applying this and other criteria to all the studies uncovered by the search whittled down the list of 2,000 useful possibilities to just 10.

One of the studies that made the cut, was an evaluation of a Blueprints model program called Athletes Training and Learning to Avoid Steroids (ATLAS), a school-based intervention designed to support the development of young male athletes.

The program’s developers at the Oregon Health and Science University say that ATLAS teaches students the confidence, skills and attitudes to enable them to make healthy (anti-drug) life choices.

Other studies in the top ten focused on the prevention of smoking or alcohol consumption, some of them for high school students, others for medical outpatients.

All were scrutinized for their effectiveness in relation to improvements in self-efficacy and associated behavior change. Seven made the first grade; two demonstrated significant effects on addiction behaviors.

In the ATLAS study, participants reported levels of self confidence and drug resistance higher than other adolescents who did not receive the program. At one year follow-up there were significantly fewer steroid users in the program group than the control group, indicating an equivalent change in behavior .

Taken as a whole, the studies included in the review, show clearly that self efficacy can be enhanced by a variety of methods. Whether these improvements have an effect on addiction behavior is more difficult to establish. None of the studies actually tested the relationship between the two variables, and so the King’s College team’s efforts were hindered by a lack of appropriate data.

Writing in the Journal of Health Psychology, Marteau and her colleagues argue that “mediation analyses” of the sort they were generally unable to make are an important item for the prevention science research agenda and should be stipulated in the research design. As they point out, studying such relationships is easier in studies that succeed in recruiting large numbers of participants.

• Summary of Marteau T M, Hyde J, Hankins M, and Deale A “Interventions to Increase Self-efficacy in the Context of Addiction Behaviours: A systematic literature review”, Journal of Health Psychology, 13, 5, pp607-623 (2008)