Performance of cigarette susceptibility index among e-cigarette and hookah users.

Abstract

Susceptibility to cigarette smoking has been used since the mid-1990s as a measure to identify youth at risk of cigarette initiation. However, it is unclear how well this measure predicts future smoking among electronic (e-)cigarette or hookah users, or among those in tobacco-friendly social environments. We used prospective data from the Southern California Children's Health Study to evaluate the performance (sensitivity, specificity, predictive value) of a composite index assessing susceptibility to smoking, and to evaluate whether performance of the measure differed by use of e-cigarettes or hookah, or immersion in a tobacco-friendly social environment. Susceptibility to cigarette smoking was measured in 11th/12th grade (2014) among never cigarette-smokers (N=1266); follow-up data on smoking initiation were obtained approximately 16 months later. Overall, 16.4% of youth initiated smoking between baseline and follow-up. The sensitivity of the susceptibility to smoking index was low (46.4%), and specificity was high (79.0%). No difference in sensitivity was observed by baseline e-cigarette use; specificity was higher among never e- cigarette users. Differences in negative predictive value (NPV) and positive predictive value (PPV) were also observed by baseline e-cigarette and hookah use. Specificity was generally lower, and sensitivity was generally higher for those in tobacco-friendly social environments. Our findings suggest that use of the susceptibility to smoking index in cross-sectional studies of older adolescents to identify those likely to begin smoking may be inappropriate for those using alternative tobacco products (e.g., e-cigarettes or hookah).