摘要：Objectives. We determined whether racial/ethnic disparities existed in coverage by type of 100% smoke-free private workplace, restaurant, and bar laws from 2000 to 2009. Methods. We combined US census population data and the American Nonsmokers’ Rights Foundation US Tobacco Control Database to calculate the percentage of individuals in counties covered by each type of law by race/ethnicity from 2000 to 2009. Results. More of the US Hispanic and Asian populations were covered by 100% smoke-free restaurant and bar laws than non-Hispanic White and non-Hispanic Black populations. Asian coverage by smoke-free bars laws increased from 36% to 75%, and Hispanic coverage increased from 31% to 62%, compared with 6% to 41% for non-Hispanic Blacks and 8% to 49% for non-Hispanic Whites. Conclusions. Hispanics and Asians benefited more from the rapid spread of smoke-free law coverage, whereas non-Hispanic Blacks benefited less. These ethnic disparities suggest a likely effect of geographic region and may provide a basis for more effective, community-based, and tailored policy-related interventions, particularly regarding areas with high concentrations of non-Hispanic Blacks. Secondhand smoke causes disease, including lung cancer, heart disease, and breast cancer. 1–3 Implementation of strong smoke-free laws is followed by drops in hospital admissions for heart attacks 4–8 and decreased smoking prevalence. 9 Regional studies have suggested that the implementation of clean indoor air and tobacco control policies is not evenly distributed across the United States; communities with high education and income are more likely to adopt 100% smoke-free laws than communities with lower education and income; disparities also exist by race/ethnicity. 10–14 In Massachusetts, areas with large minority populations were less likely to have smoke-free laws, 12 whereas in Texas large minority urban areas were more likely to have smoke-free coverage. 13 Blue-collar and service workers were less likely to be covered by smoke-free laws than white-collar workers. 15–17 Although several investigators have studied the effects of smoke-free laws at a national level, there have been important limitations in how coverage was assessed. Some studies relied only on state laws, 18–20 despite the fact that states with weak or no state laws often have strong local laws. Other studies relied on individual survey responses about law coverage, such as the initial outcomes index that measures clean indoor air laws in incorporated places (such as towns) that is combined with a state-level aggregate of individual responses regarding smoke-free coverage from the Tobacco Use Supplement to the Current Population Survey. 21–23 Some studies have not considered the fact that some county laws cover incorporated areas inside the county, 24 and other studies have not considered subcounty coverage at all. 25 In states in which county laws apply only to unincorporated areas in the county, the incorporated areas (cities or towns that function as administrative units that provide “governmental functions for a concentration of people” 26 ) can pass laws. Assuming that county laws apply everywhere in these counties introduces 2 kinds of misclassification errors: (1) if a county law exists but does not cover incorporated areas, the number of people covered by the law will be overstated, and (2) if no county law exists, but laws in incorporated cities do, the number of people covered by the law will be underestimated. Because population is not evenly distributed across counties but often clustered in cities and towns, reliance on county-level laws can miss large, subcounty population clusters. For example, the largest city in Alabama, Birmingham, has a 100% smoke-free law, but neither the State of Alabama nor Jefferson County (which includes Birmingham) have 100% smoke-free laws. Thus, relying on the county law would completely miss coverage for the largest city in the state. Measures of policy coverage have varied in earlier studies. For example, in 1 study respondent zip code was used as a means of determining respondent location and thus subcounty clean indoor air law coverage. (This level of geocode information is usually available only for privately funded data sets. 27 ) Two studies have accounted for subcounty law coverage by using the existence of a city-level law as a proxy for county coverage (in these cases, the publicly funded data sets only included county-level geocodes). 28,29 Other studies accounted for subcounty smoke-free law coverage by calculating the percentage of individuals in a county covered by clean indoor air laws 30,31 using state laws and local ordinances in the single years of 1992 32 or 2007 33 ; none accounted for changes in population over time.
