The relationship between cigarette smoking and negative health outcomes at the individual level is well-known and has been thoroughly described in the literature. Individual smokers are at substantially increased risk of chronic disease and mortality from a variety of different causes of death. However, there has been less attention to the potential for differences in cigarette smoking to produce differentials in health and mortality across populations. The goal of this dissertation is to examine the extent to which smoking operates as a factor determining mortality at the population level. Chapter 1 extends a new indirect method for estimating smoking-attributable mortality and applies it to geographic variation in adult mortality in the United States. Chapter 2 examines whether the adult mortality advantage of Hispanics compared to non-Hispanic whites in the US derives from Hispanics’ relatively low rates of smoking. Chapter 3 investigates the underlying explanations for Mexican immigrants’ relatively favorable smoking behavior. The results suggest that cigarette smoking is an important factor producing geographic disparities in mortality and the principal reason for Hispanics’ survival advantage over whites. Furthermore, there is no evidence that selective migration of non-smokers explains low rates of smoking among Mexican migrants in the United States. This dissertation makes both methodological and substantive contributions to the study of cigarette smoking, expanding current methods for estimating smoking attributable deaths and demonstrating examples of the population level impact of smoking.