83, p < .0001). The proportion of daily smoking was higher among youth from low-affluence families compared with those from high-affluence families (7% and 4%; ��2 = 38.74, p < .0001). Table 2. Description of the sample sociodemographic characteristics by cigarette smoking status Table 1. Targets AG014699 for each item in smoking policy variables The Spearman rho coefficient showed significant bivariate correlations between smoking policy variables. For example, for middle school students, packaging was correlated with vending machines and free distribution (r = .75, for both). For the clean indoor air variables, government worksites was strongly correlated with private worksites (r = .72), schools and childcare facilities (r = .54, for both), and restaurants (r = .78; complete results available from first author).
For middle school students, no youth access variable was associated with smoking status in chi-square analyses. Among the high school sample, the youth access variables found to have significant associations (p < .05) with smoking status were packaging, vending machines, and free distribution. For both middle and high school students, clean indoor air laws that target government worksites, private worksites, schools, restaurants, retail stores, and recreational and cultural facilities were associated with cigarette smoking. Penalties was only associated with high school students. The magnitude of the association was greatest for laws targeting private worksites, retail stores, and recreational and cultural facilities for the high school sample (p < .001).
Logistic regression analyses Youth access. Although we found no association between the youth access restrictions and the cigarette smoking for middle school students, regression models were run to explore the possibility of suppression effects. Packing, vending machines, and free distribution were not associated with cigarette smoking status after adjusting for potential confounders in regression models. Despite this lack of association, a consistent trend was noted in the increased probability of smoking in the presence of lax restrictions. This increased probability was not significant within a 99% CI. For high school students, the daily versus never model showed that Dacomitinib no restrictions on vending machines was a predictor of daily smoking when compared with youth living in states where vending machines were placed in adult locations only and at least 20 feet from any entry (OR = 2.02, 99% CI = 1.02�C4.01). After adjusting for sociodemographic characteristics and cigarette price, vending machine was not significant (Table 3). Table 3.