Retail tobacco availability: Measurement considerations and relation to smoking initiation

A32

Laws prohibiting commercial sales of tobacco products to minors are an important dimension of tobacco control policy in the United States. Federal regulation of tobacco sales to minors was introduced in 1992, with the passage of the Synar Amendment. Under Synar, states are required to conduct random compliance checks of tobacco retailers and maintain an 80% annual compliance rate, or risk losing federal grant funds. While the passage of Synar has led to observable decreases in cigarette sales to minors, considerable controversy still exists regarding the effectiveness of sales to minor (STM) laws as a strategy for reducing youth smoking prevalence. Studies attempting to evaluate the impact of STM laws on teenage smoking have yielded mixed results, leading some critics to conclude that this policy tool is ineffective. Others have argued that methodological problems related to the definition and measurement of retail tobacco availability may be to blame for inconclusive findings. Experts in the field argue that even in communities with good merchant compliance, youth know which stores sell to minors, so compliance levels likely need to reach a critical threshold (i.e., 90%) before the availability of tobacco has been meaningfully restricted and effects on smoking prevalence will be observed.
 >The current study used multilevel random-effects regression analyses to examine the relation of Level 1 individual factors (i.e., sex; race; age; number of smoking peers; adult smoker in home) and Level 2 environmental factors (i.e., retail tobacco availability (RTA); household income) to smoking initiation in a cohort of young adolescents living in twenty-four communities in Illinois. All Level-1 variables were derived from self-report data obtained from a student survey. The primary Level 2 variable RTA was created using merchant compliance data from two independent sources, the State of Illinois Liquor Control Commission (ILCC) and study-appointed field agents who conducted independent compliance checks. In order to address the methodological concerns raised in previous studies, associations between RTA and smoking initiation were modeled under two different sets of parameters. First, the two independent sources of RTA data were entered into the model as continuous constructs. Then, the analysis was repeated using a dichotomous variable, in which towns with ≥90% merchant compliance on both RTA measures were coded as low availability towns. Towns not meeting this stringent criterion were coded as high availability towns.
 >In the second model, several Level 1 variables and RTA Level 2 were significant correlates of smoking initiation. RTA was significantly related to tobacco use such that high availability towns were associated with increased odds that a youth had initiated smoking. This association between RTA and smoking initiation was not observed in the first model, suggesting that modeling RTA as a continuous construct may mask its relationship with smoking initiation. The results suggest that STM laws are effective for reducing youth smoking prevalence only when a critical threshold of merchant compliance is achieved. From a policy standpoint, this finding could have important implications for the Synar Amendment’s currently mandated 80% compliance level.