STATE-LEVEL STRUCTURAL RACISM AND ALCOHOL AND TOBACCO USE BEHAVIORS IN A NATIONAL PROBABILITY SAMPLE OF AFRICAN AMERICANS

Abstract

Structural racism is how society maintains and promotes racial discrimination through established and interconnected systems. Structural racism is a historical driver of health disparities, including those in the area of cancer. One pathway posited for this effect is through the promotion of maladaptive coping behaviors, such as alcohol and tobacco use. This dissertation empirically assessed the association between state-level structural racism and alcohol and tobacco use behaviors among African Americans, including within various subgroups of African Americans based on age, gender, and household income. This project analyzed secondary self-report data collected from a national probability sample of 1,946 African Americans in the Religion and Health in African Americans (RHIAA) Study. Existing measures of state level structural racism were merged into the RHIAA dataset including a composite index of structural racism assessed using five dimensions (i.e., residential segregation, and economic, employment, education, and incarceration disparities). Analyses were performed in SPSS Version 28 using hierarchical linear and logistic regression models. In the first study, two models (Model A and Model B) were constructed for each of four outcomes, frequency of alcohol consumption (measured in days per month), frequency of binge drinking (measured in the number of occurrences per month), smoking status (current smoker or not a current smoker), and smoking frequency (never smoked, former smoker, currently smoke on some days, and currently smoke every day). Model A used the composite structural racism index measure to model the four alcohol and tobacco use measures and Model B analyzed the disaggregated dimensions of structural racism rather than the composite measure. All hierarchical analyses controlled for confounding variables (i.e., participant gender, age, education, income, and employment status). In the second study, analyses using the dimension-level approach in Model B from study one were repeated in subgroups stratified by participant age, gender, and income for the frequency of binge drinking and smoking status behaviors. Statistical comparisons of the slope estimates between corresponding subgroups (e.g., younger and older age) were used to test the moderation effects of age, gender, and income on the association between structural racism and alcohol and tobacco use behaviors. Results from these studies generally indicated a positive association between state level structural racism, especially in the incarceration dimension, and binge drinking and tobacco use behaviors. Stratified analyses generally did not support age, gender, or income as moderating variables of the association between structural racism and binge drinking and tobacco use behaviors. Current findings demonstrate a need for further research on structural racism and health and progress in structural racism measurement, including further emphasis of dimension-level measurement and analysis. Findings from the current dissertation highlight the importance of addressing structural racism, especially in incarceration, to reduce alcohol and tobacco use behaviors among African Americans and help address existing health disparities.