Finally, we were interested in the possibility that tobacco dependence (i.e., the HSI) might lie along the causal pathway and at least partially account for any significant relations of menthol use status and smoking abstinence. To address this possibility, we used a nonparametric bootstrapping procedure to estimate the significance of an indirect things effect in single mediation models with 5,000 resamples with replacement from the dataset (MacKinnon, 2008). Continuous abstinence was treated as an ordinal variable with one observation (i.e., abstinent through which postquit week) for every participant. Therefore, the bootstrapping was performed by resampling individual ordinal outcomes from the dataset.

Because the CR logit model operates through a logistic regression approach for a binary smoking abstinence variable, with observations conditional on being continuously abstinent through each time point (thus, each individual contributes a varied number of binary observations), the corresponding estimates of the coefficients were standardized in calculating the indirect effect and the proportion of total effect that was mediated [] (MacKinnon, Lockwood, Brown, Wang, & Hoffman, 2007). These analyses were conducted using R version 2.13.0 (R Core Team, 2012) and were adjusted for age, gender, total annual household income, educational level, employment status, partner status, and stage. RESULTS Participant Characteristics Participants (N = 183; 95 non-Hispanic White and 88 non-Hispanic Black smokers) were 45.9 years of age (��10.31) on average. In the overall sample, continuous abstinence rates were 25.

7% at week 1, 19.1% at week 2, and 15.9% at week 3. There were fewer menthol than nonmenthol using participants (83 menthol users vs. 100 nonmenthol users). Among menthol users, continuous short-term smoking abstinence rates were 19.3% at week 1, 14.5% at week 2, and 12.1% at week 3. Among nonmenthol users, continuous short-term smoking abstinence rates were 31.0% at week 1, 23.0% at week 2, and 19.0% at week 3. Among White participants, 18.9% (n = 18) endorsed current menthol cigarette use, whereas 73.9% (n = 65) of Black participants endorsed current menthol cigarette use. White menthol smokers had higher HSI scores (4.28��1.53) as compared with Black menthol (3.42��1.27) smokers (t = ?2.43, p = .02). Likewise, White menthol smokers had higher HSI scores than White nonmenthol smokers (3.

75��1.30), although this difference was not statistically significant (p = .14). Participant characteristics by menthol use Carfilzomib status are in Table 1. Table 1. Participant Characteristics by Menthol Use Status Preliminary Analyses Preliminary analyses indicated significant differences between menthol and nonmenthol users by race and employment status. Specifically, most menthol users were Black, whereas most nonmenthol users were White participants.