Webb and colleagues examined the usefulness of 10 behaviour change theories for the field of addiction [1]. They seek to fill a gap left by Robert West's review of the applicability of theories for understanding the phenomenon of addiction [2]. They highlight further the importance of using theory to identify intervention targets and select behaviour change techniques. This is a very important endeavour. However, the paper also gives the—probably unintended—impression that the researchers are examining crumbs that were left over by West. Webb and colleagues argue that because ‘manifestations of addiction are largely behavioral’, behaviour change theories can be useful in understanding addiction. We agree that there are strong behavioural and cognitive components in addiction which can be understood by theories from social and health psychology. In fact, this has become standard practice in health communication interventions, particularly in smoking cessation, and this has led to efficacious tailored interventions [3,4]. Others have also discussed the applicability of social cognitive models for addiction [5]. The rationale for choosing these 10 particular theories, however, is not really explained, except that they were not covered by West. The authors argue that each theory has its own merits and implications for behaviour change strategies which are outlined by their Table 1 and they use Control Theory to organize them; but why Control Theory? Why not use motivation stadium as an integrating principle, as this is mentioned explicitly by the authors when they say that theories of behaviour change are heterogeneous? Indeed, the importance of certain psychological factors may differ depending upon the behavioural change process that an individual is in. For instance, if raising awareness, one may need to address different factors than changing motivation, or translating motivation into concrete actions. Mapping the various factors in this way may be helpful for researchers and practitioners alike. Additionally, testing and identifying the contributions of these theories using dismantling designs, a process not identified by the authors, is also needed.

Below the surface of the paper lies an intriguing question which is not being addressed. Is there a meaningful distinction between theories relevant for life-style behaviours and theories for addictive behaviours, or are the latter simply special cases of the former? Webb and colleagues do not go into this, while West argued that addictions deserve a more complete and complicated theory beyond relatively simple models [2]. Thus, an important question that is left unanswered is when to choose social cognitive theories, and when it is better to combine them with theories that address factors specific for addiction.

We applaud the attempt to improve the theoretical underpinnings of intervention development and to make this more transparent. In most reports on health interventions in the scientific literature, a good description is missing of how theory was used to develop an intervention: what was the causal chain between determinants, methods for behaviour change and practical intervention techniques? We know from experience that the art of grounding an intervention in theory is one of the most difficult tasks that intervention developers may face. Papers such as this one and others [6] help to provide intervention developers with a selection of theories from which to choose and to urge them to make their theoretical underpinnings more transparent, but are less useful for helping them to understand how they should go about developing their intervention based on selected theories [7]. The authors point out that the science of intervention development is still in its infancy. This might, indeed, very well be the case in the field of addiction. However, in the field of health education and health promotion there is a growing body of knowledge on how to conduct this, for instance by using planning models [8], or by integrating findings from social cognitive models in a persuasion–communication matrix [9]. This literature has accumulated in a method to design theory- and evidence-based interventions, called Intervention Mapping [10,11], which has been applied widely in health education, especially in human immunodeficiency virus prevention [12–14]. We believe that it is in the ‘how’ of applying theory to practice that the field of addiction can learn most from other fields.

Declaration of interests

Marc C. Willemsen has received a research grant from Pfizer.