• Availability;
  • drug dependence;
  • drug treatment;
  • social influence;
  • social network;
  • social norms;
  • social theory;
  • substance abuse;
  • theory building

This commentary touches upon the role of theory in behavior change and addiction research, how the theories outlined by Webb, Sniehotta & Michie [1] address key constructs, and what factors should be addressed by future theoretical perspectives on addictive behaviors.

It has been argued persuasively that in the social sciences most theories are not disproven; rather, they tend to fall out of favor and scientists move on to wrestle with new theories. For example, although papers have been published in Addiction that cast serious doubts both empirically and theoretically on the Transtheoretical Model [2,3], it continues to be cited frequently in studies of substance abuse. Even with abundant empirical evidence that a theory is not supported, it can be argued that the variables were not measured correctly or the sample or behavior was unique and hence does not generalize.

Gergen and others suggest that good social science theories are generative [4,5]. They have the capacity to challenge the guiding assumptions of our scientific disciplines and culture and allow us to look at behaviors from new perspectives. They can even have a transforming impact on society, but theories cannot serve all purposes. Thorngate has argued that there is always a trade-off between theories being simple, or general, or accurate [6].

The downside of our theories is that they can act as blinders and restrict our scientific and clinical inquiry. For example, the Transtheoretical Model may lead clinicians to conclude that a drug treatment patient who relapses was not ‘ready’ to quit using drugs. Many theories are imbued with our assumptions of human nature [7]; consequently, theories from the West, especially from the United States, tend to be highly individualistic and emphasize free will, rational decision making, and cognitive processes [8]. This perspective may be at odds with the dynamics of addictive behaviours, and impede the examination of both physiological and social factors and the development of appropriate interventions.

In examining the theories most relevant to substance abuse present by Webb et al. there are several unique constructs that need to be considered. One problematic issue is that of intentions. Most drug-dependent individuals intend to quit using drugs, but they continue to use to stave off withdrawal symptoms. Consequently, intentions are not an optimal construct. Not all the delineated theories highlight intentions. The Prototype Willingness Model suggests that adolescent risk behaviors are often not planned or intentional [9]. In choosing theoretical perspectives, it is also important to assess whether the same theory is appropriate for intervention to prevent initiation or promote cessation of substance use. We know that prolonged substance use may lead to profound changes in neurological structures and that drug overdoses may have cognitive effects [10,11]. These physiological changes may alter self-regulation and other cognitive and affective processes.

Some theories, such as the Elaboration Likelihood Model, are likely to have the most relevance for mass communication approaches for prevention of drug abuse rather than for developing drug treatment programs [12]. Protection motivation theory is another communications theory, but is limited as an individual-focused cognitive theory based on Lazarus' (1966) stress and coping model [13].

Neglected and future areas for theory building

  1. Top of page
  2. Neglected and future areas for theory building
  3. Declaration of interest
  4. References

Substance use is often a social behavior, and our theories need to be more refined in analyses of the social environment. For example, Reifman found that it was not simply the number of alcohol drinkers in the network that predicted future drinking among college students [14]. It was the pattern of drinking among the small network of peers with whom they drank with that was most predictive of future drinking. Hence, our measures of social networks need to identify key social influences. This need for specificity is also seen in the social norms literature. Social norms interventions to reduce drinking on college campuses are not usually tailored to specific referent groups [15]. It is doubtful that information on the overall student body drinking norms will be as influential as information about the norms of one's own referent group. In general, theories that include a social norms component have not delineated the multi-dimensionality of norms [16]. The constructs of subjective or descriptive and injunctive norms do not capture the wide diversity of normative influences on behaviors and how to utilize norms in behavior change interventions.

Availability has been a key construct in alcohol and tobacco research, but has been less salient in studies of other types of drug use [17]. Moreover, physiological and personality factors may directly influence or mediate substance use behaviors and, hence, need to be addressed in behavior change theories [18]. The review of behaviour change theories by Webb et al. highlights the value of theoretical perspectives, but we also need to be aware of the pitfalls of theories.


  1. Top of page
  2. Neglected and future areas for theory building
  3. Declaration of interest
  4. References