Project Summary Although enormous progress has been made in reducing cigarette smoking in the United States, large disparities remain across subpopulations such as racial minorities. Quitting smoking can greatly reduce the risk of disease and early death. However, African Americans (AA) are less likely to quit smoking successfully than are non-Hispanic Whites, and they are also three times more likely to smoke menthol cigarettes than Whites. There is a remarkable variation in the incidence of lung cancer due to cigarette smoking among racial/ethnic groups in the United States, and African Americans have significantly greater risks than Whites at lower levels of smoking. Tobacco use landscape is changing among U.S. adults with an estimated 4.5% of adults aged 18 years and older reporting current use of electronic cigarettes (EC). Evidence suggests that e-cigarette aerosol is substantially less toxic than combustible cigarette smoking, but e-cigarette use may also post elevated risk for former smokers to relapse back to combustible cigarette smoking. Biomarker can play an important role in assessing the potential health effects of tobacco products and studies have found that AA have higher levels of serum cotinine per cigarette smoked, higher urinary amount of TNE and NNAL than Whites. However, evidence on the racial disparities of biomarker outcomes of EC use is scarce, especially whether and to what extent they are different from use of various EC devices, flavors, and transitions in EC ↔ cigarettes. The overarching goal of this study is to examine the racial disparities in biomarkers of exposure and toxicants in association with EC use by analyzing the restricted Population Assessment of Tobacco and Health (PATH) Wave 1-4 biomarker data. We will link the biomarker data with the PATH adult surveys to identify the between- person and within-person differences in biomarkers by use of different vaping products, flavors, and transition in EC and combustible cigarettes across different waves (Aim 1). We will further leverage machine learning algorithms to develop a composite bio (biomarker)-socio (socio-demographics) -psycho (psychosocial factors) risk index score for each racial/ethnic group to predict subsequent abstinence from cigarette smoking and relapse to cigarette smoking since these 2 outcomes represent the most meaningful transitions from a public health perspective. This study is innovative in its focus on racial disparities of biomarkers by leveraging the nationally longitudinal data, and on using machine learning algorithms to predict future smoking cessation and relapse behaviors in EC ↔ cigarette transition. Fulfillment of these aims will build a scientific base for maximizing the value of existing biospecimen collections. Findings from this R21 study are expected to inform regulatory strategies to reduce tobacco-related morbidity and mortality and set the stage for a future R01 application that will develop personalized intervention st...