Among those who were parents, those who had a more positive adolescent selleckbio attitude toward smoking expressed significantly less support for the policy (p < .01). Table 3. Results for Full Hierarchical Regression Models Predicting Support for Prohibiting Smoking in Restaurants, Separately for Nonparents and Parents (n = 4,831) Adolescent Attitude Toward Smoking by Adult Smoking Status Interactions For one policy, increasing taxes on cigarettes, there was a significant interaction involving adolescent attitude and adult smoking status. To probe this significant interaction, we split the sample by adult smoking status to test the effect of adolescent attitude on support for increasing taxes. However, as shown in Table 4, the effects of adolescent attitude on support for increasing taxes were not significant for nonsmoking and smoking adults when modeled separately.

Table 4. Results for Full Hierarchical Regression Models Predicting Support for Increasing Taxes on Cigarettes, Separately for Adult Non-Smokers and Smokers (n = 4,833) Adolescent Smoking Status by Adult Smoking Status Interactions For one policy, raising cigarette taxes, the role of adolescent smoking status interacted with adult smoking status. Again, we probed this significant interaction by splitting the sample by adult smoking status to test whether the effect of adolescent smoking status on support for raising taxes differed by adult smoking status. Both smoking and nonsmoking adults who smoked as adolescents expressed less support for increasing taxes than did those who were not smokers in adolescence, but the effect was significant only among adult smokers (p < .

05). Discussion This is the first study to use longitudinal data to prospectively predict support for tobacco control policy measures in adulthood from smoking status and smoking attitudes measured in adolescence. The first finding of note was the generally moderate to high levels of support for the six tobacco control policies. This could reflect the fact that this sample is relatively well educated overall, which would be consistent with previous studies that have shown that education level is positively associated with support for tobacco control policies Anacetrapib (Bernat et al., 2009; Doucet et al., 2007; Hamilton et al., 2005). However, the level of support was not equally high for all policies. For example, the mean level of support was 3.26 for prohibiting smoking in bars compared with 4.15 for prohibiting smoking in restaurants. This same finding was demonstrated in a sample of young adults (Bernat et al., 2009). In terms of the adult predictors of policy support tested in the current study, our findings were generally consistent with prior research.