The researchers who responded to my overview of brief alcohol intervention and implementation challenges for research and practice [1] seemed to agree with me on most issues. Perhaps there is little within this subject that is up ‘for debate’, then. Nevertheless, I would like to comment on some of the issues addressed by Allamani [2], Kaner [3] and Moyer & Finney [4].

Kaner emphasized the need for increased research on system-level factors. I agree that this is important to obtain knowledge for improved understanding of implementation issues. Brief intervention research would benefit from becoming more of a multi-disciplinary endeavour, involving specialists in fields other than medicine and health. Kaner and Allamani both addressed the importance of conducting more research with practitioners (although Allamani was pessimistic about finding anything much of importance). I agree that we must partner with practitioners to gain a better understanding of how interventions can be embedded within clinical practice. Brief intervention research has been predominantly quantitative, but investigations into implementation issues of ‘how’ and ‘why’ require more methodological pluralism.

Kaner also touches upon another issue, which I think needs more attention: the way we interpret efficacy and effectiveness in intervention trials. Kaner and colleagues made an admirable attempt in their landmark Cochrane review [5] to categorize interventions on an efficacy–effectiveness continuum, based upon a number of study protocol aspects, e.g. whether the patients, alcohol problems, practitioners and practice contexts were ‘clinically representative’. Most of the studies were found to be fairly effectiveness-orientated, suggesting that they were conducted under quite realistic, real-world circumstances. However, intervention duration was not part of the classification scheme. The median length of the interventions in the review was 25 minutes, which is far from realistic in most busy health care practice settings. I contend that most intervention studies in this field still are fairly efficacy-orientated, which contributes to the difficulties of moving evidence into practice. An important challenge is to conduct intervention research under more realistic circumstances.

The need to broaden secondary alcohol prevention research and implementation efforts geographically becomes very clear when reading Allamani's response. He paints a somewhat gloomy picture, referring to how general practitioners ‘can be inclined to become judgemental about the client and his/her drinking behaviour, raising guilt or shameful feelings in the consultation room’. Nearly all the brief intervention research thus far has been undertaken in a few English-speaking countries (United Kingdom, United Nations, Canada, Australia), Nordic countries (Sweden, Norway, Denmark, Finland) and a few other countries in Europe (Germany, Switzerland, France and Spain). We also need to broaden research to learn more about the impact and reach of brief interventions in other populations than those who are seen in primary health care settings. Moyer & Finney described a web-based brief intervention project targeting students, which seems highly relevant in order to provide more knowledge concerning how younger generations can be reached.

Kaner stresses the need to search for the ‘active ingredients’ of brief interventions, even though she states that ‘consistent benefits’ of these interventions have been reported. It is natural to want to know ‘what makes it work’ (there is similar research into motivational interviewing interventions), yet I think it is futile to try to isolate context-free ‘success factors’ of interventions. After all, a half-century of psychotherapy research has demonstrated that the most robust predictor of treatment success is the quality of the therapeutic alliance, i.e. the collaborative nature of the partnership between practitioner and client. Hence, the ‘ingredients’ that lead to successful outcomes might not be attributable to specific elements of the intervention itself. Again, I do not believe that quantitative methodology can provide all the answers we are looking for.

Although many researchers seem to agree upon the important challenges to achieve wider implementation of brief interventions, it is important that researchers, practitioners and policy makers in this field coordinate efforts and prioritize issues that need to be addressed. Researchers from outside the immediate brief intervention ‘community’ should be invited to contribute new perspectives and competencies, for increased learning.

Declaration of interests