Do number of smoking friends and changes over time predict smoking relapse? Findings from the International Tobacco Control Four-Country Survey

Background: Past research indicates that smokers with a large number of smoking friends within their social network are less interested in quitting, less likely to attempt to quit, and less likely to successfully quit. The extent to which a pro-smoking social network may increase relapse risk among ex-smokers is unclear. This study investigated among ex-smokers whether the number of close friends who smoke and changes in this number influence relapse risk. Methods: The study was a prospective cohort study of 551 adults who participated in the Australian and UK arms of the International Tobacco Control (ITC) project and were ex-smokers at wave 9 (2013) and followed up to wave 10 (2014). Logistic models regressed smoking relapse at follow-up on the baseline number of their five closest friends who smoked and changes in this number over time. Results: Ex-smokers who reported having 4 or 5 smokers among their five closest friends were more likely to relapse than those who had no smokers among their five closest friends (adjusted odds ratio [AOR] = 4.86, 95% confidence interval [CI] = 1.48-15.99, p = .009). Ex-smokers who gained smoking friends over time, but not those who lost smoking friends, were also more likely to relapse compared to those with the same number of smoking friends over time (AOR = 4.52, 95% CI = 2.15-9.52, p < .001; AOR = 1.08, 95% CI = 0.49-2.36, p = .848, respectively). Conclusions: This study demonstrated that relapse risk was elevated among ex-smokers who had more smokers among their close friends and also among those where the number of smokers in their social network increased over time.