For example, experiences of discrimination are associated with negative affect (e.g., Brondolo et al., 2008) and free overnight delivery with smoking (e.g., Landrine & Klonoff, 2000). Given this, one might argue that the discrimination�Csmoking relation might be mediated by psychological distress. However, the analyses reported here show no relation of psychological distress and smoking for Blacks and Hispanics. It may be that the facets of distress captured by the K6��s focus on overall psychological distress are separate and distinct from those aroused by experiences of discrimination (for an example of specific types of affect having different effects on substance use in the context of discrimination, see Gibbons et al., 2010).

Future research is needed to address the role of sociocultural aspects of race, including discrimination, in the relation of psychological distress to smoking behavior. Finally, the findings about the relation between psychological distress and smoking behavior raise interesting questions about the potential nature of race and ethnic differences in smoking cessation. Success at quitting smoking may differ as a function of race and ethnicity, with Blacks being less likely to maintain abstinence (King et al., 2004; Messer, Trinidad, Al-Delaimy, & Pierce, 2008). Understanding factors contributing to that relationship potentially have implications for how we design targeted intervention programs. Negative affect is associated with smoking relapse (Shiffman & Waters, 2004); the findings concerning differences in the relation of psychological distress to smoking suggest that research examining differences by race in triggers of relapse should be conducted.

Limitations There are, of course, limitations to the study, which should be addressed. First, it is important to note that the study is cross-sectional, so the findings should be properly interpreted as associations between distress and smoking and differences in association by race rather than as patterns of causal relations. Second, it is important to note that the K6 measure is a measure of overall psychological distress. There are specific dimensions of negative affect, which are not well assessed by the measure (e.g., stress, anger; see Collins & Lepore, 2009; Webb & Carey, 2008). Future research should examine a broader spectrum of distress and affective states in relation to smoking status.

Finally, although the study is designed as a population-representative sample, the nature of survey design is such that there may be elements of the population who are not well represented (e.g., individuals without landline telephones, including those with cell phones and low SES individuals) and the point estimates in the survey may not fully mesh with those of other population AV-951 representative reports (e.g.