"Hardcore"definitions and their application to a population-based sample of smokers

Introduction: As smoking prevalence declines, some suggest that so-called "hardcore" smokers will come to represent a growing and irreducible proportion of current smokers ("hardening hypothesis"). Different definitions of a "hardcore" smoker have been used in the literature. This paper describes population-based definitions of "hardcore" smokers and compares estimates of the prevalence of "hardcore" smokers derived using these definitions. Methods: Definitions identified in a comprehensive literature search were reduced to their component constructs. We estimated the prevalence of "hardcore" smokers as a proportion of all current adult smokers in Ontario, Canada, using data from the Ontario Tobacco Survey (2005-2008;N = 4,130). Definition concordance was examined using bivariate cross-tabulations. Results: Six definitions were identified in the literature. Five definitions included constructs of quit intentions and quit attempts, four included nicotine dependence, three included long-term use, and one included a measure of smoker knowledge about smoking hazards and confronting substantial societal disapprobation of smoking. Estimates of "hardcore" smoker prevalence in Ontario based on these definitions ranged from 0.03% to 13.77%. Conclusions: Estimates of the prevalence of "hardcore" smokers in Ontario varied considerably between the six definitions of the "hardcore" smokers found in the population-based literature. This study underscores the need for consensus on the best definition of "hardcore" smoker.