摘要：Objectives. We examined the influence of tobacco control policies (tobacco control program expenditures, smoke-free air laws, youth access law compliance, and cigarette prices) on youth smoking outcomes (smoking susceptibility, past-year initiation, current smoking, and established smoking). Methods. We combined data from the 2002 to 2008 National Surveys on Drug Use and Health with state and municipality population data from the US Census Bureau to assess the associations between state tobacco control policy variables and youth smoking outcomes, focusing on youths aged 12 to 17 years. We also examined the influence of policy variables on youth access when these variables were held at 2002 levels. Results. Per capita funding for state tobacco control programs was negatively associated with all 4 smoking outcomes. Smoke-free air laws were negatively associated with all outcomes except past-year initiation, and cigarette prices were associated only with current smoking. We found no association between these outcomes and retailer compliance with youth access laws. Conclusions. Smoke-free air laws and state tobacco control programs are effective strategies for curbing youth smoking. After peaking in the mid- to late 1990s, youth smoking rates have declined substantially. According to the Monitoring the Future study, smoking rates from 1997 to 2010 declined among 8th, 10th, and 12th graders by 63%, 54%, and 47%, respectively. 1 However, most of these declines occurred from 1997 to 2007. Thereafter, rates remained fairly stable, with past-month smoking prevalence in 2010 of 7.1%, 13.6%, and 19.2% among 8th, 10th, and 12th graders, respectively. 1 Curbing youth smoking can reduce adult smoking 2 and the related significant health and economic burdens. Tobacco industry documents have indicated that tobacco companies have viewed smoking initiation as key to sustaining future profitability. 3 In fact, Healton et al. 3 estimated that the declines in youth smoking from 1997 to 2002 would ultimately lead to a $4 billion loss of tobacco company revenue. Declines in youth smoking occurred during a period of significant activity in tobacco control policy. In 1998, the Master Settlement Agreement between 46 states and major tobacco companies made more than $200 billion available to states. Although only a fraction of these funds have been dedicated to tobacco control, 4 the Master Settlement Agreement led to a marked increase in funding for state tobacco control programs. The average per capita funding for state tobacco control programs more than doubled in real terms from $1.21 in 1997 to $2.52 in 2008 (our calculations). Cigarette prices increased significantly in real terms from $2.62 to $4.35 over the same period, 5 largely as a result of cigarette excise tax increases. In addition, a higher percentage of the US population is now covered by state and local smoke-free air ordinances in workplaces, restaurants, and bars, an increase from 3.8% to 68.6% from 1997 to 2008. Finally, state and federal laws encourage states to enforce youth access laws. Several studies have examined the effect of cigarette prices, youth access restrictions, smoking bans, and tobacco control programs on youth smoking, but with the exception of Tauras et al., 6 analyses have often included only a subset of these factors. However, the Tauras et al. study covered 1991 to 2000, largely before the Master Settlement Agreement, when funding for state tobacco control programs increased substantially in some cases. In addition, Tauras et al. included an index of the comprehensiveness of state youth access laws but did not include any measure of compliance. To reduce the variables in their regression models, they constructed their youth access index and index of smoke-free air laws as scales, each covering 9 different laws and reflecting a strength score applied to each law. A limitation of these indexes is that they force a linear relationship between the index and the outcome of interest: each law in the index has the same effect (e.g., random store inspections and vending machine availability). Although some of the smoke-free air laws have a different weight in the scale, this choice was also somewhat arbitrary. Some studies examining the influence of cigarette prices or taxes on youth smoking have generally found that higher cigarette prices are associated with declines in youth smoking prevalence. 6–9 DeCicca et al., 10,11 however, in their longitudinal study, examined the transition from being a nonsmoker in 1992 to being a smoker in 2000 as a function of the 2000 state excise tax rate and found no association between the tax rate and smoking. 10,11 Recent studies have examined the association between youth smoking and various measures of smoke-free air laws. For example, a longitudinal study in Massachusetts 12 found that town-level smoke-free restaurant ordinances were associated with a reduced progression to established smoking by youths. Similar results were found in Minnesota. 13 Other cross-sectional studies have found a negative association between youth smoking and smoke-free air laws. 6,14–16 The literature is less clear regarding the influence of youth access laws and their enforcement on youth smoking. One systematic literature review 17 concluded that youth access interventions are ineffective. However, another review 18 showed the limitations of some of the measures of youth access and asserted that regularly enforced youth access restrictions can reduce youth smoking. One study examined the effect of funding for tobacco control programs, revealing that higher funding levels are associated with decreased youth smoking prevalence. 6 However, most studies on youth smoking have relied on school-based surveys and have not captured youths who are not in school. We aim to extend and enhance the evidence base for tobacco control policies by quantifying the influence of funding on state tobacco control programs, cigarette prices, smoke-free air laws, and retailer youth access law violation rate on youth smoking. Our analysis merges data on state-level tobacco control policies with the 2002–2008 National Surveys on Drug Use and Health (NSDUHs), an annual household-based survey with national and state representative samples of youths aged 12 to 17 years (including youths not in school).
