Path analysis of cognitive behavioral models in smoking: what is the relationship between concepts involved?

Abstract

Introduction: Several cognitive behaviour therapy models are used in stop smoking services. Despite their similarities, they do not focus on the same elements of the process of therapy. The aims of this project were (i) to assess if data support the propositions about the assumed causal relationships between concepts deduced from the most popular of these models for smokers (relapse prevention, dynamic regulatory, and cognitive model) and (ii) to assess if data support one model better than the others. Method: Smokers were either attending community stop smoking services or willing to. Mediational path analysis was performed. Concepts such as positive and negative expectancies, urges, self-efficacy, and cognitive schemas were examined. Change of dependence at 3 months was the outcome. Results: Ninety-nine smokers were recruited. Comparison between models suggested that a modified version of the dynamic regulatory model, in which self-efficacy and negative expectancies were the final mediators of outcome, was the best to fit the data (AIC = 82.42), with very good indices (CFI = −0.964 and RMSEA = 0.027). Conclusion: For smokers, confidence in their ability to stop smoking seems to be the final and most important factor in treatment effectiveness. Negative expectancies have a direct positive effect as well as an indirect one through self-efficacy.