Date of Graduation

2014

Document Type

Dissertation

Degree Type

PhD

College

School of Public Health

Department

Epidemiology

Committee Chair

Kimberly Horn

Committee Co-Chair

Steven Branstetter

Committee Member

Jamison Conley

Committee Member

Lan Guo

Committee Member

Michael Hendryx

Committee Member

Juhua Luo

Committee Member

Motao Zhu

Abstract

Despite the historic declines in the prevalence of cigarette smoking in the United States, tobacco use remains a major public health burden. The persistent use and susceptibility of tobacco arises from tobacco-related health disparities, which historically occur among vulnerable and underserved populations. Since rural/nonmetropolitan populations suffer disproportionately more tobacco-related health disparities than urban/metropolitan populations, the main goal of this project and its three component studies was to address rural tobacco-related health disparities. Study 1 utilizes an exploratory spatial data analysis approach to identify those rural areas disproportionately affected by tobacco. This study assessed the overall spatial association between county-level rurality and county-level adults smoking rates in the contiguous United States. Furthermore, local indicators of spatial association identified localized county clusters with either positive or negative spatial association. Results found an overall significant positive spatial association (Moran's I = 0.0438, pseudo p-value < 0.0001) between county-level rurality and county-level adult cigarette smoking prevalence. Among the significant (p-value < 0.05) rural county clusters identified, rural areas of Appalachia were a primary area of concern, while other rural clusters (e.g., spatial outliers), like those found in the Pacific Northwest had much lower rates of smoking than expected based upon their rural nature. While Study 1 identified the rural areas disproportionately affect by tobacco use, Study 2 and Study 3 examine the relationships between the retail tobacco environment (N=276) and public middle/high schools (N=33) within metropolitan (N=3) and nonmetropolitan (N=3) counties of West Virginia. Study 2 visualizes the location of each county's public middle/high school relative to the underlying density of tobacco retailers in each county. Additionally, Study 2 examines the spatial relationship between the locations of tobacco retailers and public middle/high schools in each of the metropolitan and nonmetropolitan counties in West Virginia. Results found that 77% of the public middle/high schools with an above average tobacco retailer density value were located within a metro county. While each county's density surface of tobacco retailers took on either a monocentric or polycentric urban pattern, which was dependent on the structure of each county's underlying built environment, but was independent of their metropolitan/nonmetropolitan status. Additionally, the spatial relationship between the locations of tobacco retailers and public middle/high schools in each county exhibited a significant spatial interaction. Study 3 examines the relationship between tobacco retailer density and tobacco retailer's proximity to a public middle/high school in each of the metropolitan and nonmetropolitan counties in West Virginia. Additionally, Study 3 assesses the effects of tobacco retailer proximity---density groups, metropolitan/nonmetropolitan status, and store types on tobacco retailer characteristics (e.g., tobacco marketing and cigarette pack price). A significant positive relationship (Chi-square = 116.54, p-value < 0.001) existed between levels of tobacco retailer density and tobacco retailer proximity to school levels. The amount of exterior tobacco marketing displayed in each tobacco retailer displayed a significant (F (6, 254) = 3.528, p-value < 0.01) between proximity -density groups and store types. While a significant (F (4, 254) = 5.593, p-value < 0.001) three-way interaction between proximity - density groups, metropolitan/nonmetropolitan status, and store types existed for the average pack price of non-menthol cigarettes. Together, these three studies identify the need in reducing tobacco-related health disparities among those most vulnerable---rural populations. However, because tobacco-related health disparities result from the complex interactions between numerous factors future research must identify how these factors affect the burden of tobacco use in these rural populations. Only then, can tobacco control specialist begin to implement tailored comprehensive, systemic tobacco control programs in these rural areas.

Download

Share

COinS