A community-based participatory approach to identify factors related to tobacco advertising at the community level

Background: The purpose of this study was to explore the relationship between socioeconomic and demographic factors believed to be related to tobacco advertising at the community level. The analysis used data from the StoreAlert survey, a report card filled out by youth in their communities to give stores a grade (A, B, C, D, F) with regard to the levels of interior and exterior tobacco advertising. The StoreAlert data were collected between 2004-2007 in Massachusetts by community groups funded by the Massachusetts Department of Public Health.

Methods: Univariate and multivariate analyses were conducted to identify independent variables associated with a city or town's outcome of “high-level of tobacco advertising” (defined as a score of D or F). The city/town-level variables tested were: percent with a high school degree or higher, household income, percent living below poverty, percent of non-English speakers at home, smoking prevalence and residents' race/Hispanic ethnicity.

Results: In the univariate analyses, education level, household income, smoking prevalence and percent of non-English speakers were each found to be associated with a high level of advertising. In the multivariate analyses, education level, household income, smoking prevalence and percent of non-English speakers were included in a stepwise logistic regression analysis to determine which of these factors, if any would be associated with higher likelihood of high levels of tobacco advertising. We found in particular that 20 percent or more of non-English speakers in a community resulted as the factor most strongly associated with the outcome.