The relation of AS to smoking outcome expectancies was significantly mediated by negative urgency (unadjusted: β [95% CI] = .106 [.051 – .172], 33% of total effect; adjusted for MASQ-AA and FTND: β [95% CI] = .078 [.025 – .142], 29% of total effect), and the remaining direct effect of AS on smoking outcome expectancies was significant (Table 2). Similarly, the relation of AS to smoking abstinence expectancies was significantly mediated by negative urgency (unadjusted: β [95% CI] = .057 [.014 – .112], 22% of total effect; adjusted: (β [95% CI] = .042 [.008 – .090], 19% of total effect), and the remaining direct effect of AS on smoking abstinence expectancies was significant (Table 2). The relation of AS to proportion of quit attempts involving abstinence of less than one month was significantly mediated by negative urgency only when not adjusting for covariates (β [95% CI] = .063 [.012 – .128], 33% of total effect), and the remaining direct effect of AS on proportion of quit attempts involving abstinence of less than one month was not significant (β = .13, p = .17).
